1
|
Levy JJ, Davis MJ, Chacko RS, Davis MJ, Fu LJ, Goel T, Pamal A, Nafi I, Angirekula A, Suvarna A, Vempati R, Christensen BC, Hayden MS, Vaickus LJ, LeBoeuf MR. Intraoperative margin assessment for basal cell carcinoma with deep learning and histologic tumor mapping to surgical site. NPJ Precis Oncol 2024; 8:2. [PMID: 38172524 PMCID: PMC10764333 DOI: 10.1038/s41698-023-00477-7] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/12/2022] [Accepted: 11/14/2023] [Indexed: 01/05/2024] Open
Abstract
Successful treatment of solid cancers relies on complete surgical excision of the tumor either for definitive treatment or before adjuvant therapy. Intraoperative and postoperative radial sectioning, the most common form of margin assessment, can lead to incomplete excision and increase the risk of recurrence and repeat procedures. Mohs Micrographic Surgery is associated with complete removal of basal cell and squamous cell carcinoma through real-time margin assessment of 100% of the peripheral and deep margins. Real-time assessment in many tumor types is constrained by tissue size, complexity, and specimen processing / assessment time during general anesthesia. We developed an artificial intelligence platform to reduce the tissue preprocessing and histological assessment time through automated grossing recommendations, mapping and orientation of tumor to the surgical specimen. Using basal cell carcinoma as a model system, results demonstrate that this approach can address surgical laboratory efficiency bottlenecks for rapid and complete intraoperative margin assessment.
Collapse
Affiliation(s)
- Joshua J Levy
- Department of Pathology and Laboratory Medicine, Cedars-Sinai Medical Center, Los Angeles, CA, 90048, USA.
- Department of Computational Biomedicine, Cedars-Sinai Medical Center, Los Angeles, CA, 90048, USA.
- Department of Dermatology, Geisel School of Medicine at Dartmouth, Hanover, NH, 03756, USA.
- Emerging Diagnostic and Investigative Technologies, Clinical Genomics and Advanced Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, NH, 03756, USA.
- Department of Epidemiology, Geisel School of Medicine at Dartmouth, Hanover, NH, 03756, USA.
- Program in Quantitative Biomedical Sciences, Geisel School of Medicine at Dartmouth, Hanover, NH, 03756, USA.
| | - Matthew J Davis
- Department of Dermatology, Geisel School of Medicine at Dartmouth, Hanover, NH, 03756, USA
| | | | - Michael J Davis
- Department of Dermatology, Geisel School of Medicine at Dartmouth, Hanover, NH, 03756, USA
| | - Lucy J Fu
- Geisel School of Medicine at Dartmouth, Hanover, NH, 03755, USA
| | - Tarushii Goel
- Thomas Jefferson High School for Science and Technology, Alexandria, VA, 22312, USA
- Massachusetts Institute of Technology, Cambridge, MA, 02139, USA
| | - Akash Pamal
- Thomas Jefferson High School for Science and Technology, Alexandria, VA, 22312, USA
- University of Virginia, Charlottesville, VA, 22903, USA
| | - Irfan Nafi
- Thomas Jefferson High School for Science and Technology, Alexandria, VA, 22312, USA
- Stanford University, Palo Alto, CA, 94305, USA
| | - Abhinav Angirekula
- Thomas Jefferson High School for Science and Technology, Alexandria, VA, 22312, USA
- University of Illinois Urbana-Champaign, Champaign, IL, 61820, USA
| | - Anish Suvarna
- Thomas Jefferson High School for Science and Technology, Alexandria, VA, 22312, USA
| | - Ram Vempati
- Thomas Jefferson High School for Science and Technology, Alexandria, VA, 22312, USA
| | - Brock C Christensen
- Department of Dermatology, Geisel School of Medicine at Dartmouth, Hanover, NH, 03756, USA
- Department of Molecular and Systems Biology, Geisel School of Medicine at Dartmouth, Hanover, NH, 03756, USA
- Department of Community and Family Medicine, Geisel School of Medicine at Dartmouth, Hanover, NH, 03756, USA
| | - Matthew S Hayden
- Department of Dermatology, Geisel School of Medicine at Dartmouth, Hanover, NH, 03756, USA
| | - Louis J Vaickus
- Emerging Diagnostic and Investigative Technologies, Clinical Genomics and Advanced Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, NH, 03756, USA
| | - Matthew R LeBoeuf
- Department of Dermatology, Geisel School of Medicine at Dartmouth, Hanover, NH, 03756, USA
| |
Collapse
|
2
|
Srinivasan G, Davis MJ, LeBoeuf MR, Fatemi M, Azher ZL, Lu Y, Diallo AB, Saldias Montivero MK, Kolling FW, Perrard L, Salas LA, Christensen BC, Palys TJ, Karagas MR, Palisoul SM, Tsongalis GJ, Vaickus LJ, Preum SM, Levy JJ. Potential to Enhance Large Scale Molecular Assessments of Skin Photoaging through Virtual Inference of Spatial Transcriptomics from Routine Staining. Pac Symp Biocomput 2024; 29:477-491. [PMID: 38160301 PMCID: PMC10813837] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Grants] [Subscribe] [Scholar Register] [Indexed: 01/03/2024]
Abstract
The advent of spatial transcriptomics technologies has heralded a renaissance in research to advance our understanding of the spatial cellular and transcriptional heterogeneity within tissues. Spatial transcriptomics allows investigation of the interplay between cells, molecular pathways, and the surrounding tissue architecture and can help elucidate developmental trajectories, disease pathogenesis, and various niches in the tumor microenvironment. Photoaging is the histological and molecular skin damage resulting from chronic/acute sun exposure and is a major risk factor for skin cancer. Spatial transcriptomics technologies hold promise for improving the reliability of evaluating photoaging and developing new therapeutics. Challenges to current methods include limited focus on dermal elastosis variations and reliance on self-reported measures, which can introduce subjectivity and inconsistency. Spatial transcriptomics offers an opportunity to assess photoaging objectively and reproducibly in studies of carcinogenesis and discern the effectiveness of therapies that intervene in photoaging and preventing cancer. Evaluation of distinct histological architectures using highly-multiplexed spatial technologies can identify specific cell lineages that have been understudied due to their location beyond the depth of UV penetration. However, the cost and interpatient variability using state-of-the-art assays such as the 10x Genomics Spatial Transcriptomics assays limits the scope and scale of large-scale molecular epidemiologic studies. Here, we investigate the inference of spatial transcriptomics information from routine hematoxylin and eosin-stained (H&E) tissue slides. We employed the Visium CytAssist spatial transcriptomics assay to analyze over 18,000 genes at a 50-micron resolution for four patients from a cohort of 261 skin specimens collected adjacent to surgical resection sites for basal cell and squamous cell keratinocyte tumors. The spatial transcriptomics data was co-registered with 40x resolution whole slide imaging (WSI) information. We developed machine learning models that achieved a macro-averaged median AUC and F1 score of 0.80 and 0.61 and Spearman coefficient of 0.60 in inferring transcriptomic profiles across the slides, and accurately captured biological pathways across various tissue architectures.
Collapse
Affiliation(s)
- Gokul Srinivasan
- Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, NH 03756, USA,
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
Collapse
|
3
|
Davis MJ, Srinivasan G, Chacko R, Chen S, Suvarna A, Vaickus LJ, Torres VC, Hodge S, Chen EY, Preum S, Samkoe KS, Christensen BC, LeBoeuf MR, Levy JJ. A deep learning algorithm to detect cutaneous squamous cell carcinoma on frozen sections in Mohs micrographic surgery: A retrospective assessment. Exp Dermatol 2024; 33:e14949. [PMID: 37864429 DOI: 10.1111/exd.14949] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/22/2023] [Revised: 09/13/2023] [Accepted: 09/30/2023] [Indexed: 10/22/2023]
Abstract
Intraoperative margin analysis is crucial for the successful removal of cutaneous squamous cell carcinomas (cSCC). Artificial intelligence technologies (AI) have previously demonstrated potential for facilitating rapid and complete tumour removal using intraoperative margin assessment for basal cell carcinoma. However, the varied morphologies of cSCC present challenges for AI margin assessment. The aim of this study was to develop and evaluate the accuracy of an AI algorithm for real-time histologic margin analysis of cSCC. To do this, a retrospective cohort study was conducted using frozen cSCC section slides. These slides were scanned and annotated, delineating benign tissue structures, inflammation and tumour to develop an AI algorithm for real-time margin analysis. A convolutional neural network workflow was used to extract histomorphological features predictive of cSCC. This algorithm demonstrated proof of concept for identifying cSCC with high accuracy, highlighting the potential for integration of AI into the surgical workflow. Incorporation of AI algorithms may improve efficiency and completeness of real-time margin assessment for cSCC removal, particularly in cases of moderately and poorly differentiated tumours/neoplasms. Further algorithmic improvement incorporating surrounding tissue context is necessary to remain sensitive to the unique epidermal landscape of well-differentiated tumours, and to map tumours to their original anatomical position/orientation.
Collapse
Affiliation(s)
- Matthew J Davis
- Department of Dermatology, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire, USA
| | | | | | - Sophie Chen
- Caddo Parish Magnet High School, Shreveport, Louisiana, USA
| | - Anish Suvarna
- Thomas Jefferson School for Science and Technology, Alexandria, Virginia, USA
| | - Louis J Vaickus
- Department of Pathology and Laboratory Medicine, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire, USA
| | - Veronica C Torres
- Thayer School of Engineering, Dartmouth College, Hanover, New Hampshire, USA
| | - Sassan Hodge
- Thayer School of Engineering, Dartmouth College, Hanover, New Hampshire, USA
| | - Eunice Y Chen
- Geisel School of Medicine, Hanover, New Hampshire, USA
- Department of Surgery, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire, USA
| | - Sarah Preum
- Department of Computer Science, Dartmouth College, Hanover, New Hampshire, USA
| | - Kimberley S Samkoe
- Geisel School of Medicine, Hanover, New Hampshire, USA
- Thayer School of Engineering, Dartmouth College, Hanover, New Hampshire, USA
| | - Brock C Christensen
- Department of Epidemiology, Dartmouth College Geisel School of Medicine, Hanover, New Hampshire, USA
- Department of Molecular and Systems Biology, Geisel School of Medicine at Dartmouth, Lebanon, New Hampshire, USA
- Department of Community and Family Medicine, Geisel School of Medicine at Dartmouth, Lebanon, New Hampshire, USA
| | - Matthew R LeBoeuf
- Department of Dermatology, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire, USA
| | - Joshua J Levy
- Department of Dermatology, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire, USA
- Dartmouth College, Hanover, New Hampshire, USA
- Geisel School of Medicine, Hanover, New Hampshire, USA
- Department of Pathology and Laboratory Medicine, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire, USA
- Department of Epidemiology, Dartmouth College Geisel School of Medicine, Hanover, New Hampshire, USA
- Program in Quantitative Biomedical Science, Dartmouth College Geisel School of Medicine, Hanover, New Hampshire, USA
| |
Collapse
|
4
|
Azher ZL, Fatemi M, Lu Y, Srinivasan G, Diallo AB, Christensen BC, Salas LA, Kolling FW, Perreard L, Palisoul SM, Vaickus LJ, Levy JJ. Spatial Omics Driven Crossmodal Pretraining Applied to Graph-based Deep Learning for Cancer Pathology Analysis. Pac Symp Biocomput 2024; 29:464-476. [PMID: 38160300 PMCID: PMC10783797] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Grants] [Subscribe] [Scholar Register] [Indexed: 01/03/2024]
Abstract
Graph-based deep learning has shown great promise in cancer histopathology image analysis by contextualizing complex morphology and structure across whole slide images to make high quality downstream outcome predictions (ex: prognostication). These methods rely on informative representations (i.e., embeddings) of image patches comprising larger slides, which are used as node attributes in slide graphs. Spatial omics data, including spatial transcriptomics, is a novel paradigm offering a wealth of detailed information. Pairing this data with corresponding histological imaging localized at 50-micron resolution, may facilitate the development of algorithms which better appreciate the morphological and molecular underpinnings of carcinogenesis. Here, we explore the utility of leveraging spatial transcriptomics data with a contrastive crossmodal pretraining mechanism to generate deep learning models that can extract molecular and histological information for graph-based learning tasks. Performance on cancer staging, lymph node metastasis prediction, survival prediction, and tissue clustering analyses indicate that the proposed methods bring improvement to graph based deep learning models for histopathological slides compared to leveraging histological information from existing schemes, demonstrating the promise of mining spatial omics data to enhance deep learning for pathology workflows.
Collapse
Affiliation(s)
- Zarif L Azher
- Thomas Jefferson High School for Science and Technology, Alexandria, VA 22312, USA,
| | | | | | | | | | | | | | | | | | | | | | | |
Collapse
|
5
|
Mindiola Romero AE, Tafe LJ, Green DC, Deharvengt SJ, Winnick KN, Tsongalis GJ, Baker ML, Linos K, Levy JJ, Kerr DA. Utility of Retrospective Molecular Analysis in Diagnostically Challenging Mesenchymal Neoplasms. Int J Surg Pathol 2023; 31:1473-1484. [PMID: 36911994 DOI: 10.1177/10668969231157783] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 03/14/2023]
Abstract
Introduction: Molecular analysis plays a growing role in the diagnosis of mesenchymal neoplasms. The aim of this study was to retrospectively apply broad, multiplex molecular assays (a solid tumor targeted next-generation sequencing [NGS]) assay and single nucleotide polymorphism [SNP] microarray) to selected tumors, exploring the current utility and limitations. Methods: We searched our database (2010-2020) for diagnostically challenging mesenchymal neoplasms. After histologic review of available slides, tissue blocks were selected for NGS, SNP microarray, or both. DNA and RNA were extracted using the AllPrep DNA/RNA FFPE Kit Protocol on the QIAcube instrument. The NGS platform used was the TruSight Tumor 170 (TST-170). For SNP array, copy number variant (CNV) analysis was performed using the OncoScanTM CNV Plus Assay. Results: DNA/RNA was successfully extracted from 50% of tumors (n = 10/20). Specimens not successfully extracted included 6 core biopsies, 3 incisional biopsies, and 1 resection; 4 were decalcified (3 hydrochloric acid, 1 ethylenediaminetetraacetic acid). Higher tumor proportion and number of tumor cells were parameters positively associated with sufficient DNA/RNA extraction whereas necrosis and decalcification were negatively associated with sufficient extraction. Molecular testing helped reach a definitive diagnosis in 50% of tumors (n = 5/10). Conclusions: Although the overall utility of this approach is limited, these molecular panels can be helpful in detecting a specific "driver" alteration.
Collapse
Affiliation(s)
- Andres E Mindiola Romero
- Department of Pathology and Laboratory Medicine, Dartmouth-Hitchcock Medical Center, Lebanon, NH, USA
- Geisel School of Medicine at Dartmouth, Dartmouth College, Hanover, NH, USA
| | - Laura J Tafe
- Department of Pathology and Laboratory Medicine, Dartmouth-Hitchcock Medical Center, Lebanon, NH, USA
- Geisel School of Medicine at Dartmouth, Dartmouth College, Hanover, NH, USA
| | - Donald C Green
- Department of Pathology and Laboratory Medicine, Dartmouth-Hitchcock Medical Center, Lebanon, NH, USA
| | - Sophie J Deharvengt
- Department of Pathology and Laboratory Medicine, Dartmouth-Hitchcock Medical Center, Lebanon, NH, USA
| | - Kimberly N Winnick
- Department of Pathology and Laboratory Medicine, Dartmouth-Hitchcock Medical Center, Lebanon, NH, USA
| | - Gregory J Tsongalis
- Department of Pathology and Laboratory Medicine, Dartmouth-Hitchcock Medical Center, Lebanon, NH, USA
- Geisel School of Medicine at Dartmouth, Dartmouth College, Hanover, NH, USA
| | - Michael L Baker
- Department of Pathology and Laboratory Medicine, Dartmouth-Hitchcock Medical Center, Lebanon, NH, USA
- Geisel School of Medicine at Dartmouth, Dartmouth College, Hanover, NH, USA
| | - Konstantinos Linos
- Department of Pathology and Laboratory Medicine, Dartmouth-Hitchcock Medical Center, Lebanon, NH, USA
- Geisel School of Medicine at Dartmouth, Dartmouth College, Hanover, NH, USA
| | - Joshua J Levy
- Department of Pathology and Laboratory Medicine, Dartmouth-Hitchcock Medical Center, Lebanon, NH, USA
- Geisel School of Medicine at Dartmouth, Dartmouth College, Hanover, NH, USA
| | - Darcy A Kerr
- Department of Pathology and Laboratory Medicine, Dartmouth-Hitchcock Medical Center, Lebanon, NH, USA
- Geisel School of Medicine at Dartmouth, Dartmouth College, Hanover, NH, USA
| |
Collapse
|
6
|
Martinez Coconubo D, Levy JJ, Kerr DA, Vaickus LJ, Vidis L, Glass RE, Gutmann EJ, Marotti JD, Liu X. Use of molecular testing results to analyze the overuse of atypia of undetermined significance in thyroid cytology. J Am Soc Cytopathol 2023; 12:451-460. [PMID: 37775434 DOI: 10.1016/j.jasc.2023.09.002] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/02/2023] [Revised: 08/31/2023] [Accepted: 09/03/2023] [Indexed: 10/01/2023]
Abstract
INTRODUCTION The suggested atypia of undetermined significance (AUS) rate for thyroid fine-needle aspiration biopsies is 10% or less. Prompted by a high institutional AUS rate, we examined using molecular testing results (MTR) as a potential quality metric tool to reduce the AUS rate. We correlated MTR with AUS cytologic findings, surgical pathology follow-up, and individual pathologist AUS rates. MATERIALS AND METHODS Demographic data, cytologic diagnoses, MTR, and surgical pathology diagnoses were retrospectively obtained. MTR were classified as either positive or negative. AUS rates and MTR proportions were compared among pathologists. The cytomorphologic features of 143 AUS cases were assessed and correlated with MTR. RESULTS Between 2017 and 2022, 710 of 3247 thyroid fine-needle aspirations were classified as AUS, with a yearly average rate of 22% (range = 19%-26%). AUS cases included: 331 (47%) with architectural atypia; 204 (29%) with oncocytic (Hürthle cell) atypia; 99 (14%) with combined architectural and cytologic atypia; and 76 (10%) with isolated cytologic atypia. Most AUS cases with molecular testing had negative MTR (360/492, 73%). AUS with cytologic atypia had higher positive MTR risk (logarithm of odds ratio = 1.27, 95% credible interval [0.5-2.04], P = 0.001). The average positive MTR rate by pathologist was 21.5% (range 0%-35%); higher positive MTR rates had better correlation with subsequent neoplastic/malignant histologic diagnoses. The MTR sensitivity for malignant disease was 89% and the negative predictive value was 91%. CONCLUSIONS MTR analysis reveals the importance of cytologic atypia as a determinant of malignancy risk in AUS cases. Periodic analysis of MTR data alongside individual pathologist AUS rates can help refine diagnostic criteria and potentially reduce AUS overuse.
Collapse
Affiliation(s)
- Daniel Martinez Coconubo
- Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire.
| | - Joshua J Levy
- Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire; Departments of Quantitative Biomedical Sciences, Dermatology and Epidemiology, Geisel School of Medicine, Hanover, New Hampshire
| | - Darcy A Kerr
- Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire; Geisel School of Medicine at Dartmouth, Hanover, New Hampshire
| | - Louis J Vaickus
- Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire; Geisel School of Medicine at Dartmouth, Hanover, New Hampshire
| | - Laura Vidis
- Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire
| | - Ryan E Glass
- Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire
| | - Edward J Gutmann
- Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire; Geisel School of Medicine at Dartmouth, Hanover, New Hampshire
| | - Jonathan D Marotti
- Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire; Geisel School of Medicine at Dartmouth, Hanover, New Hampshire
| | - Xiaoying Liu
- Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire; Geisel School of Medicine at Dartmouth, Hanover, New Hampshire
| |
Collapse
|
7
|
Fatemi MY, Lu Y, Sharma C, Feng E, Azher ZL, Diallo AB, Srinivasan G, Rosner GM, Pointer KB, Christensen BC, Salas LA, Tsongalis GJ, Palisoul SM, Perreard L, Kolling FW, Vaickus LJ, Levy JJ. Feasibility of Inferring Spatial Transcriptomics from Single-Cell Histological Patterns for Studying Colon Cancer Tumor Heterogeneity. medRxiv 2023:2023.10.09.23296701. [PMID: 37873186 PMCID: PMC10593064 DOI: 10.1101/2023.10.09.23296701] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 10/25/2023]
Abstract
Background Spatial transcriptomics involves studying the spatial organization of gene expression within tissues, offering insights into the molecular diversity of tumors. While spatial gene expression is commonly amalgamated from 1-10 cells across 50-micron spots, recent methods have demonstrated the capability to disaggregate this information at subspot resolution by leveraging both expression and histological patterns. However, elucidating such information from histology alone presents a significant challenge but if solved can better permit spatial molecular analysis at cellular resolution for instances where Visium data is not available, reducing study costs. This study explores integrating single-cell histological and transcriptomic data to infer spatial mRNA expression patterns in whole slide images collected from a cohort of stage pT3 colorectal cancer patients. A cell graph neural network algorithm was developed to align histological information extracted from detected cells with single cell RNA patterns through optimal transport methods, facilitating the analysis of cellular groupings and gene relationships. This approach leveraged spot-level expression as an intermediary to co-map histological and transcriptomic information at the single-cell level. Results Our study demonstrated that single-cell transcriptional heterogeneity within a spot could be predicted from histological markers extracted from cells detected within a spot. Furthermore, our model exhibited proficiency in delineating overarching gene expression patterns across whole-slide images. This approach compared favorably to traditional patch-based computer vision methods as well as other methods which did not incorporate single cell expression during the model fitting procedures. Topological nuances of single-cell expression within a Visium spot were preserved using the developed methodology. Conclusion This innovative approach augments the resolution of spatial molecular assays utilizing histology as a sole input through synergistic co-mapping of histological and transcriptomic datasets at the single-cell level, anchored by spatial transcriptomics. While initial results are promising, they warrant rigorous validation. This includes collaborating with pathologists for precise spatial identification of distinct cell types and utilizing sophisticated assays, such as Xenium, to attain deeper subcellular insights.
Collapse
|
8
|
Fatemi MY, Lu Y, Diallo AB, Srinivasan G, Azher ZL, Christensen BC, Salas LA, Tsongalis GJ, Palisoul SM, Perreard L, Kolling FW, Vaickus LJ, Levy JJ. The Overlooked Role of Specimen Preparation in Bolstering Deep Learning-Enhanced Spatial Transcriptomics Workflows. medRxiv 2023:2023.10.09.23296700. [PMID: 37873287 PMCID: PMC10593052 DOI: 10.1101/2023.10.09.23296700] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 10/25/2023]
Abstract
The application of deep learning methods to spatial transcriptomics has shown promise in unraveling the complex relationships between gene expression patterns and tissue architecture as they pertain to various pathological conditions. Deep learning methods that can infer gene expression patterns directly from tissue histomorphology can expand the capability to discern spatial molecular markers within tissue slides. However, current methods utilizing these techniques are plagued by substantial variability in tissue preparation and characteristics, which can hinder the broader adoption of these tools. Furthermore, training deep learning models using spatial transcriptomics on small study cohorts remains a costly endeavor. Necessitating novel tissue preparation processes enhance assay reliability, resolution, and scalability. This study investigated the impact of an enhanced specimen processing workflow for facilitating a deep learning-based spatial transcriptomics assessment. The enhanced workflow leveraged the flexibility of the Visium CytAssist assay to permit automated H&E staining (e.g., Leica Bond) of tissue slides, whole-slide imaging at 40x-resolution, and multiplexing of tissue sections from multiple patients within individual capture areas for spatial transcriptomics profiling. Using a cohort of thirteen pT3 stage colorectal cancer (CRC) patients, we compared the efficacy of deep learning models trained on slide prepared using an enhanced workflow as compared to the traditional workflow which leverages manual tissue staining and standard imaging of tissue slides. Leveraging Inceptionv3 neural networks, we aimed to predict gene expression patterns across matched serial tissue sections, each stemming from a distinct workflow but aligned based on persistent histological structures. Findings indicate that the enhanced workflow considerably outperformed the traditional spatial transcriptomics workflow. Gene expression profiles predicted from enhanced tissue slides also yielded expression patterns more topologically consistent with the ground truth. This led to enhanced statistical precision in pinpointing biomarkers associated with distinct spatial structures. These insights can potentially elevate diagnostic and prognostic biomarker detection by broadening the range of spatial molecular markers linked to metastasis and recurrence. Future endeavors will further explore these findings to enrich our comprehension of various diseases and uncover molecular pathways with greater nuance. Combining deep learning with spatial transcriptomics provides a compelling avenue to enrich our understanding of tumor biology and improve clinical outcomes. For results of the highest fidelity, however, effective specimen processing is crucial, and fostering collaboration between histotechnicians, pathologists, and genomics specialists is essential to herald this new era in spatial transcriptomics-driven cancer research.
Collapse
|
9
|
Levy JJ, Chan N, Marotti JD, Kerr DA, Gutmann EJ, Glass RE, Dodge CP, Suriawinata AA, Christensen BC, Liu X, Vaickus LJ. Large-scale validation study of an improved semiautonomous urine cytology assessment tool: AutoParis-X. Cancer Cytopathol 2023; 131:637-654. [PMID: 37377320 DOI: 10.1002/cncy.22732] [Citation(s) in RCA: 3] [Impact Index Per Article: 3.0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/01/2023] [Revised: 05/11/2023] [Accepted: 05/12/2023] [Indexed: 06/29/2023]
Abstract
BACKGROUND Adopting a computational approach for the assessment of urine cytology specimens has the potential to improve the efficiency, accuracy, and reliability of bladder cancer screening, which has heretofore relied on semisubjective manual assessment methods. As rigorous, quantitative criteria and guidelines have been introduced for improving screening practices (e.g., The Paris System for Reporting Urinary Cytology), algorithms to emulate semiautonomous diagnostic decision-making have lagged behind, in part because of the complex and nuanced nature of urine cytology reporting. METHODS In this study, the authors report on the development and large-scale validation of a deep-learning tool, AutoParis-X, which can facilitate rapid, semiautonomous examination of urine cytology specimens. RESULTS The results of this large-scale, retrospective validation study indicate that AutoParis-X can accurately determine urothelial cell atypia and aggregate a wide variety of cell-related and cluster-related information across a slide to yield an atypia burden score, which correlates closely with overall specimen atypia and is predictive of Paris system diagnostic categories. Importantly, this approach accounts for challenges associated with the assessment of overlapping cell cluster borders, which improve the ability to predict specimen atypia and accurately estimate the nuclear-to-cytoplasm ratio for cells in these clusters. CONCLUSIONS The authors developed a publicly available, open-source, interactive web application that features a simple, easy-to-use display for examining urine cytology whole-slide images and determining the level of atypia in specific cells, flagging the most abnormal cells for pathologist review. The accuracy of AutoParis-X (and other semiautomated digital pathology systems) indicates that these technologies are approaching clinical readiness and necessitates full evaluation of these algorithms in head-to-head clinical trials.
Collapse
Affiliation(s)
- Joshua J Levy
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire, USA
- Department of Dermatology, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire, USA
- Department of Epidemiology, Dartmouth College Geisel School of Medicine, Hanover, New Hampshire, USA
- Program in Quantitative Biomedical Sciences, Dartmouth College Geisel School of Medicine, Hanover, New Hampshire, USA
| | - Natt Chan
- Program in Quantitative Biomedical Sciences, Dartmouth College Geisel School of Medicine, Hanover, New Hampshire, USA
| | - Jonathan D Marotti
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire, USA
- Dartmouth College Geisel School of Medicine, Hanover, New Hampshire, USA
| | - Darcy A Kerr
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire, USA
- Dartmouth College Geisel School of Medicine, Hanover, New Hampshire, USA
| | - Edward J Gutmann
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire, USA
- Dartmouth College Geisel School of Medicine, Hanover, New Hampshire, USA
| | - Ryan E Glass
- Department of Pathology, University of Pittsburgh Medical Center East, Pittsburgh, Pennsylvania, USA
| | | | - Arief A Suriawinata
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire, USA
- Dartmouth College Geisel School of Medicine, Hanover, New Hampshire, USA
| | - Brock C Christensen
- Department of Epidemiology, Dartmouth College Geisel School of Medicine, Hanover, New Hampshire, USA
- Department of Molecular and Systems Biology, Dartmouth College Geisel School of Medicine, Hanover, New Hampshire, USA
- Department of Community and Family Medicine, Dartmouth College Geisel School of Medicine, Hanover, New Hampshire, USA
| | - Xiaoying Liu
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire, USA
- Dartmouth College Geisel School of Medicine, Hanover, New Hampshire, USA
| | - Louis J Vaickus
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire, USA
- Dartmouth College Geisel School of Medicine, Hanover, New Hampshire, USA
| |
Collapse
|
10
|
Levy JJ, Chan N, Marotti JD, Rodrigues NJ, Ismail AAO, Kerr DA, Gutmann EJ, Glass RE, Dodge CP, Suriawinata AA, Christensen B, Liu X, Vaickus LJ. Examining longitudinal markers of bladder cancer recurrence through a semiautonomous machine learning system for quantifying specimen atypia from urine cytology. Cancer Cytopathol 2023; 131:561-573. [PMID: 37358142 PMCID: PMC10527805 DOI: 10.1002/cncy.22725] [Citation(s) in RCA: 1] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/02/2023] [Revised: 03/31/2023] [Accepted: 04/20/2023] [Indexed: 06/27/2023]
Abstract
BACKGROUND Urine cytology is generally considered the primary approach for screening for recurrence of bladder cancer. However, it is currently unclear how best to use cytological examinations for assessment and early detection of recurrence, beyond identifying a positive finding that requires more invasive methods to confirm recurrence and decide on therapeutic options. Because screening programs are frequent, and can be burdensome, finding quantitative means to reduce this burden for patients, cytopathologists, and urologists is an important endeavor and can improve both the efficiency and reliability of findings. Additionally, identifying ways to risk-stratify patients is crucial for improving quality of life while reducing the risk of future recurrence or progression of the cancer. METHODS In this study, a computational machine learning tool, AutoParis-X, was leveraged to extract imaging features from urine cytology examinations longitudinally to study the predictive potential of urine cytology for assessing recurrence risk. This study examined how the significance of imaging predictors changes over time before and after surgery to determine which predictors and time periods are most relevant for assessing recurrence risk. RESULTS Results indicate that imaging predictors extracted using AutoParis-X can predict recurrence as well or better than traditional cytological/histological assessments alone and that the predictiveness of these features is variable across time, with key differences in overall specimen atypia identified immediately before tumor recurrence. CONCLUSIONS Further research will clarify how computational methods can be effectively used in high-volume screening programs to improve recurrence detection and complement traditional modes of assessment.
Collapse
Affiliation(s)
- Joshua J. Levy
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, NH, 03766
- Department of Dermatology, Dartmouth Hitchcock Medical Center, Lebanon, NH, 03766
- Department of Epidemiology, Dartmouth College Geisel School of Medicine, Hanover, NH, 03756
- Program in Quantitative Biomedical Sciences, Dartmouth College Geisel School of Medicine, Hanover, NH, 03756
| | - Natt Chan
- Program in Quantitative Biomedical Sciences, Dartmouth College Geisel School of Medicine, Hanover, NH, 03756
| | - Jonathan D. Marotti
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, NH, 03766
- Dartmouth College Geisel School of Medicine, Hanover, NH, 03756
| | - Nathalie J. Rodrigues
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, NH, 03766
| | - A. Aziz O. Ismail
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, NH, 03766
- White River Junction VA Medical Center, White River Junction, VT, 05009
| | - Darcy A. Kerr
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, NH, 03766
- Dartmouth College Geisel School of Medicine, Hanover, NH, 03756
| | - Edward J. Gutmann
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, NH, 03766
- Dartmouth College Geisel School of Medicine, Hanover, NH, 03756
| | | | | | - Arief A. Suriawinata
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, NH, 03766
- Dartmouth College Geisel School of Medicine, Hanover, NH, 03756
| | - Brock Christensen
- Department of Epidemiology, Dartmouth College Geisel School of Medicine, Hanover, NH, 03756
- Department of Molecular and Systems Biology, Dartmouth College Geisel School of Medicine, Hanover, NH, 03756
- Department of Community and Family Medicine, Dartmouth College Geisel School of Medicine, Hanover, NH, 03756
| | - Xiaoying Liu
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, NH, 03766
- Dartmouth College Geisel School of Medicine, Hanover, NH, 03756
| | - Louis J. Vaickus
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, NH, 03766
- Dartmouth College Geisel School of Medicine, Hanover, NH, 03756
| |
Collapse
|
11
|
Azher ZL, Suvarna A, Chen JQ, Zhang Z, Christensen BC, Salas LA, Vaickus LJ, Levy JJ. Assessment of emerging pretraining strategies in interpretable multimodal deep learning for cancer prognostication. BioData Min 2023; 16:23. [PMID: 37481666 PMCID: PMC10363299 DOI: 10.1186/s13040-023-00338-w] [Citation(s) in RCA: 2] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/22/2022] [Accepted: 07/05/2023] [Indexed: 07/24/2023] Open
Abstract
BACKGROUND Deep learning models can infer cancer patient prognosis from molecular and anatomic pathology information. Recent studies that leveraged information from complementary multimodal data improved prognostication, further illustrating the potential utility of such methods. However, current approaches: 1) do not comprehensively leverage biological and histomorphological relationships and 2) make use of emerging strategies to "pretrain" models (i.e., train models on a slightly orthogonal dataset/modeling objective) which may aid prognostication by reducing the amount of information required for achieving optimal performance. In addition, model interpretation is crucial for facilitating the clinical adoption of deep learning methods by fostering practitioner understanding and trust in the technology. METHODS Here, we develop an interpretable multimodal modeling framework that combines DNA methylation, gene expression, and histopathology (i.e., tissue slides) data, and we compare performance of crossmodal pretraining, contrastive learning, and transfer learning versus the standard procedure. RESULTS Our models outperform the existing state-of-the-art method (average 11.54% C-index increase), and baseline clinically driven models (average 11.7% C-index increase). Model interpretations elucidate consideration of biologically meaningful factors in making prognosis predictions. DISCUSSION Our results demonstrate that the selection of pretraining strategies is crucial for obtaining highly accurate prognostication models, even more so than devising an innovative model architecture, and further emphasize the all-important role of the tumor microenvironment on disease progression.
Collapse
Affiliation(s)
- Zarif L Azher
- Thomas Jefferson High School for Science and Technology, Alexandria, VA, USA
| | - Anish Suvarna
- Thomas Jefferson High School for Science and Technology, Alexandria, VA, USA
| | - Ji-Qing Chen
- Cancer Biology Graduate Program, Dartmouth College Geisel School of Medicine, Hanover, NH, USA
- Program in Quantitative Biomedical Sciences, Dartmouth College Geisel School of Medicine, Hanover, NH, USA
- Department of Epidemiology, Dartmouth College Geisel School of Medicine, Hanover, NH, USA
| | - Ze Zhang
- Program in Quantitative Biomedical Sciences, Dartmouth College Geisel School of Medicine, Hanover, NH, USA
- Department of Epidemiology, Dartmouth College Geisel School of Medicine, Hanover, NH, USA
| | - Brock C Christensen
- Department of Epidemiology, Dartmouth College Geisel School of Medicine, Hanover, NH, USA
- Department of Molecular and Systems Biology, Dartmouth College Geisel School of Medicine, Hanover, NH, USA
- Department of Community and Family Medicine, Dartmouth College Geisel School of Medicine, Hanover, NH, USA
| | - Lucas A Salas
- Department of Epidemiology, Dartmouth College Geisel School of Medicine, Hanover, NH, USA
- Department of Molecular and Systems Biology, Dartmouth College Geisel School of Medicine, Hanover, NH, USA
- Integrative Neuroscience at Dartmouth (IND) Graduate Program, Dartmouth College Geisel School of Medicine, Hanover, NH, USA
| | - Louis J Vaickus
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Health, Lebanon, NH, USA
| | - Joshua J Levy
- Program in Quantitative Biomedical Sciences, Dartmouth College Geisel School of Medicine, Hanover, NH, USA.
- Department of Epidemiology, Dartmouth College Geisel School of Medicine, Hanover, NH, USA.
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Health, Lebanon, NH, USA.
- Department of Dermatology, Dartmouth Health, Lebanon, NH, USA.
| |
Collapse
|
12
|
Torres VC, Hodge S, Levy JJ, Vaickus LJ, Chen EY, LeBouef M, Samkoe KS. Paired-agent imaging as a rapid en face margin screening method in Mohs micrographic surgery. Front Oncol 2023; 13:1196517. [PMID: 37427140 PMCID: PMC10325620 DOI: 10.3389/fonc.2023.1196517] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Grants] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/29/2023] [Accepted: 06/05/2023] [Indexed: 07/11/2023] Open
Abstract
Background Mohs micrographic surgery is a procedure used for non-melanoma skin cancers that has 97-99% cure rates largely owing to 100% margin analysis enabled by en face sectioning with real-time, iterative histologic assessment. However, the technique is limited to small and aggressive tumors in high-risk areas because the histopathological preparation and assessment is very time intensive. To address this, paired-agent imaging (PAI) can be used to rapidly screen excised specimens and identify tumor positive margins for guided and more efficient microscopic evaluation. Methods A mouse xenograft model of human squamous cell carcinoma (n = 8 mice, 13 tumors) underwent PAI. Targeted (ABY-029, anti-epidermal growth factor receptor (EGFR) affibody molecule) and untargeted (IRDye 680LT carboxylate) imaging agents were simultaneously injected 3-4 h prior to surgical tumor resection. Fluorescence imaging was performed on main, unprocessed excised specimens and en face margins (tissue sections tangential to the deep margin surface). Binding potential (BP) - a quantity proportional to receptor concentration - and targeted fluorescence signal were measured for each, and respective mean and maximum values were analyzed to compare diagnostic ability and contrast. The BP and targeted fluorescence of the main specimen and margin samples were also correlated with EGFR immunohistochemistry (IHC). Results PAI consistently outperformed targeted fluorescence alone in terms of diagnostic ability and contrast-to-variance ratio (CVR). Mean and maximum measures of BP resulted in 100% accuracy, while mean and maximum targeted fluorescence signal offered 97% and 98% accuracy, respectively. Moreover, maximum BP had the greatest average CVR for both main specimen and margin samples (average 1.7 ± 0.4 times improvement over other measures). Fresh tissue margin imaging improved similarity with EGFR IHC volume estimates compared to main specimen imaging in line profile analysis; and margin BP specifically had the strongest concordance (average 3.6 ± 2.2 times improvement over other measures). Conclusions PAI was able to reliably distinguish tumor from normal tissue in fresh en face margin samples using the single metric of maximum BP. This demonstrated the potential for PAI to act as a highly sensitive screening tool to eliminate the extra time wasted on real-time pathological assessment of low-risk margins.
Collapse
Affiliation(s)
- Veronica C. Torres
- Thayer School of Engineering, Dartmouth College, Hanover, NH, United States
| | - Sassan Hodge
- Thayer School of Engineering, Dartmouth College, Hanover, NH, United States
| | - Joshua J. Levy
- Geisel School of Medicine, Dartmouth College, Hanover, NH, United States
- Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, NH, United States
- Department of Dermatology, Dartmouth Hitchcock Medical Center, Lebanon, NH, United States
- Quantitative Biomedical Sciences, Geisel School of Medicine, Dartmouth College, Hanover, NH, United States
- Department of Epidemiology, Geisel School of Medicine, Dartmouth College, Hanover, NH, United States
| | - Louis J. Vaickus
- Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, NH, United States
| | - Eunice Y. Chen
- Geisel School of Medicine, Dartmouth College, Hanover, NH, United States
- Department of Surgery, Dartmouth Hitchcock Medical Center, Lebanon, NH, United States
| | - Matthew LeBouef
- Geisel School of Medicine, Dartmouth College, Hanover, NH, United States
- Department of Dermatology, Dartmouth Hitchcock Medical Center, Lebanon, NH, United States
| | | |
Collapse
|
13
|
Levy JJ, Zavras JP, Veziroglu EM, Nasir-Moin M, Kolling FW, Christensen BC, Salas LA, Barney RE, Palisoul SM, Ren B, Liu X, Kerr DA, Pointer KB, Tsongalis GJ, Vaickus LJ. Identification of Spatial Proteomic Signatures of Colon Tumor Metastasis: A Digital Spatial Profiling Approach. Am J Pathol 2023; 193:778-795. [PMID: 37037284 PMCID: PMC10284031 DOI: 10.1016/j.ajpath.2023.02.020] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 12/04/2022] [Revised: 01/29/2023] [Accepted: 02/24/2023] [Indexed: 04/12/2023]
Abstract
Over 150,000 Americans are diagnosed with colorectal cancer (CRC) every year, and annually >50,000 individuals are estimated to die of CRC, necessitating improvements in screening, prognostication, disease management, and therapeutic options. CRC tumors are removed en bloc with surrounding vasculature and lymphatics. Examination of regional lymph nodes at the time of surgical resection is essential for prognostication. Developing alternative approaches to indirectly assess recurrence risk would have utility in cases where lymph node yield is incomplete or inadequate. Spatially dependent, immune cell-specific (eg, tumor-infiltrating lymphocytes), proteomic, and transcriptomic expression patterns inside and around the tumor-the tumor immune microenvironment-can predict nodal/distant metastasis and probe the coordinated immune response from the primary tumor site. The comprehensive characterization of tumor-infiltrating lymphocytes and other immune infiltrates is possible using highly multiplexed spatial omics technologies, such as the GeoMX Digital Spatial Profiler. In this study, machine learning and differential co-expression analyses helped identify biomarkers from Digital Spatial Profiler-assayed protein expression patterns inside, at the invasive margin, and away from the tumor, associated with extracellular matrix remodeling (eg, granzyme B and fibronectin), immune suppression (eg, forkhead box P3), exhaustion and cytotoxicity (eg, CD8), Programmed death ligand 1-expressing dendritic cells, and neutrophil proliferation, among other concomitant alterations. Further investigation of these biomarkers may reveal independent risk factors of CRC metastasis that can be formulated into low-cost, widely available assays.
Collapse
Affiliation(s)
- Joshua J Levy
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Health, Lebanon, New Hampshire; Department of Dermatology, Dartmouth Health, Lebanon, New Hampshire; Department of Epidemiology, Dartmouth College Geisel School of Medicine, Hanover, New Hampshire; Program in Quantitative Biomedical Sciences, Dartmouth College Geisel School of Medicine, Hanover, New Hampshire.
| | | | - Eren M Veziroglu
- Dartmouth College Geisel School of Medicine, Hanover, New Hampshire
| | | | | | - Brock C Christensen
- Department of Epidemiology, Dartmouth College Geisel School of Medicine, Hanover, New Hampshire; Department of Molecular and Systems Biology, Dartmouth College Geisel School of Medicine, Hanover, New Hampshire; Department of Community and Family Medicine, Dartmouth College Geisel School of Medicine, Hanover, New Hampshire
| | - Lucas A Salas
- Department of Epidemiology, Dartmouth College Geisel School of Medicine, Hanover, New Hampshire; Department of Molecular and Systems Biology, Dartmouth College Geisel School of Medicine, Hanover, New Hampshire; Integrative Neuroscience at Dartmouth Graduate Program, Dartmouth College Geisel School of Medicine, Hanover, New Hampshire
| | - Rachael E Barney
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Health, Lebanon, New Hampshire
| | - Scott M Palisoul
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Health, Lebanon, New Hampshire
| | - Bing Ren
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Health, Lebanon, New Hampshire
| | - Xiaoying Liu
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Health, Lebanon, New Hampshire
| | - Darcy A Kerr
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Health, Lebanon, New Hampshire
| | - Kelli B Pointer
- Section of Radiation Oncology, Department of Medicine, Dartmouth College Geisel School of Medicine, Hanover, New Hampshire
| | - Gregory J Tsongalis
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Health, Lebanon, New Hampshire.
| | - Louis J Vaickus
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Health, Lebanon, New Hampshire
| |
Collapse
|
14
|
Davis MJ, Srinivasan G, Chacko R, Chen S, Suvarna A, Vaickus LJ, Torres VC, Hodge S, Chen EY, Preum S, Samkoe KS, Christensen BC, LeBoeuf M, Levy JJ. A deep learning algorithm to detect cutaneous squamous cell carcinoma on frozen sections in Mohs micrographic surgery: a retrospective assessment. medRxiv 2023:2023.05.14.23289960. [PMID: 37293008 PMCID: PMC10246018 DOI: 10.1101/2023.05.14.23289960] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 06/10/2023]
Abstract
Importance Intraoperative margin analysis is crucial for the successful removal of cutaneous squamous cell carcinomas (cSCC). Artificial intelligence technologies (AI) have previously demonstrated potential for facilitating rapid and complete tumor removal using intraoperative margin assessment for basal cell carcinoma. However, the varied morphologies of cSCC present challenges for AI margin assessment. Objective To develop and evaluate the accuracy of an AI algorithm for real-time histologic margin analysis of cSCC. Design A retrospective cohort study was conducted using frozen cSCC section slides and adjacent tissues. Setting This study was conducted in a tertiary care academic center. Participants Patients undergoing Mohs micrographic surgery for cSCC between January and March 2020. Exposures Frozen section slides were scanned and annotated, delineating benign tissue structures, inflammation, and tumor to develop an AI algorithm for real-time margin analysis. Patients were stratified by tumor differentiation status. Epithelial tissues including epidermis and hair follicles were annotated for moderate-well to well differentiated cSCC tumors. A convolutional neural network workflow was used to extract histomorphological features predictive of cSCC at 50-micron resolution. Main Outcomes and Measures The performance of the AI algorithm in identifying cSCC at 50-micron resolution was reported using the area under the receiver operating characteristic curve. Accuracy was also reported by tumor differentiation status and by delineation of cSCC from epidermis. Model performance using histomorphological features alone was compared to architectural features (i.e., tissue context) for well-differentiated tumors. Results The AI algorithm demonstrated proof of concept for identifying cSCC with high accuracy. Accuracy differed by differentiation status, driven by challenges in separating cSCC from epidermis using histomorphological features alone for well-differentiated tumors. Consideration of broader tissue context through architectural features improved the ability to delineate tumor from epidermis. Conclusions and Relevance Incorporating AI into the surgical workflow may improve efficiency and completeness of real-time margin assessment for cSCC removal, particularly in cases of moderately and poorly differentiated tumors/neoplasms. Further algorithmic improvement is necessary to remain sensitive to the unique epidermal landscape of well-differentiated tumors, and to map tumors to their original anatomical position/orientation. Future studies should assess the efficiency improvements and cost benefits and address other confounding pathologies such as inflammation and nuclei. Funding sources JL is supported by NIH grants R24GM141194, P20GM104416 and P20GM130454. Support for this work was also provided by the Prouty Dartmouth Cancer Center development funds. Key Points Question: How can the efficiency and accuracy of real-time intraoperative margin analysis for the removal of cutaneous squamous cell carcinoma (cSCC) be improved, and how can tumor differentiation be incorporated into this approach?Findings: A proof-of-concept deep learning algorithm was trained, validated, and tested on frozen section whole slide images (WSI) for a retrospective cohort of cSCC cases, demonstrating high accuracy in identifying cSCC and related pathologies. Histomorphology alone was found to be insufficient to delineate tumor from epidermis in histologic identification of well-differentiated cSCC. Incorporation of surrounding tissue architecture and shape improved the ability to delineate tumor from normal tissue.Meaning: Integrating artificial intelligence into surgical procedures has the potential to enhance the thoroughness and efficiency of intraoperative margin analysis for cSCC removal. However, accurately accounting for the epidermal tissue based on the tumor's differentiation status requires specialized algorithms that consider the surrounding tissue context. To meaningfully integrate AI algorithms into clinical practice, further algorithmic refinement is needed, as well as the mapping of tumors to their original surgical site, and evaluation of the cost and efficacy of these approaches to address existing bottlenecks.
Collapse
|
15
|
Bobak CA, Zhao Y, Levy JJ, O’Malley AJ. GRANDPA: GeneRAtive network sampling using degree and property augmentation applied to the analysis of partially confidential healthcare networks. Appl Netw Sci 2023; 8:23. [PMID: 37188323 PMCID: PMC10173245 DOI: 10.1007/s41109-023-00548-5] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Grants] [Track Full Text] [Figures] [Subscribe] [Scholar Register] [Received: 12/01/2022] [Accepted: 04/24/2023] [Indexed: 05/17/2023]
Abstract
Protecting medical privacy can create obstacles in the analysis and distribution of healthcare graphs and statistical inferences accompanying them. We pose a graph simulation model which generates networks using degree and property augmentation and provide a flexible R package that allows users to create graphs that preserve vertex attribute relationships and approximating the retention of topological properties observed in the original graph (e.g., community structure). We illustrate our proposed algorithm using a case study based on Zachary's karate network and a patient-sharing graph generated from Medicare claims data in 2019. In both cases, we find that community structure is preserved, and normalized root mean square error between cumulative distributions of the degrees across the generated and the original graphs is low (0.0508 and 0.0514 respectively).
Collapse
Affiliation(s)
- Carly A. Bobak
- Department of Biomedical Data Science, Dartmouth College, Hanover, NH USA
- The Dartmouth Institute for Health Policy and Clinical Practice, Dartmouth College, Hanover, NH USA
- Research Computing, Dartmouth College, Hanover, NH USA
| | - Yifan Zhao
- Department of Biomedical Data Science, Dartmouth College, Hanover, NH USA
- The Dartmouth Institute for Health Policy and Clinical Practice, Dartmouth College, Hanover, NH USA
| | - Joshua J. Levy
- Department of Pathology and Laboratory Medicine, Dartmouth College, Hanover, NH USA
- Department of Dermatology, Dartmouth College, Hanover, NH USA
- Department of Epidemiology, Dartmouth College, Hanover, NH USA
| | - A. James O’Malley
- Department of Biomedical Data Science, Dartmouth College, Hanover, NH USA
- The Dartmouth Institute for Health Policy and Clinical Practice, Dartmouth College, Hanover, NH USA
| |
Collapse
|
16
|
Fatemi M, Feng E, Sharma C, Azher Z, Goel T, Ramwala O, Palisoul SM, Barney RE, Perreard L, Kolling FW, Salas LA, Christensen BC, Tsongalis GJ, Vaickus LJ, Levy JJ. Inferring spatial transcriptomics markers from whole slide images to characterize metastasis-related spatial heterogeneity of colorectal tumors: A pilot study. J Pathol Inform 2023; 14:100308. [PMID: 37114077 PMCID: PMC10127126 DOI: 10.1016/j.jpi.2023.100308] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/24/2022] [Revised: 03/23/2023] [Accepted: 03/24/2023] [Indexed: 03/31/2023] Open
Abstract
Over 150 000 Americans are diagnosed with colorectal cancer (CRC) every year, and annually over 50 000 individuals will die from CRC, necessitating improvements in screening, prognostication, disease management, and therapeutic options. Tumor metastasis is the primary factor related to the risk of recurrence and mortality. Yet, screening for nodal and distant metastasis is costly, and invasive and incomplete resection may hamper adequate assessment. Signatures of the tumor-immune microenvironment (TIME) at the primary site can provide valuable insights into the aggressiveness of the tumor and the effectiveness of various treatment options. Spatially resolved transcriptomics technologies offer an unprecedented characterization of TIME through high multiplexing, yet their scope is constrained by cost. Meanwhile, it has long been suspected that histological, cytological, and macroarchitectural tissue characteristics correlate well with molecular information (e.g., gene expression). Thus, a method for predicting transcriptomics data through inference of RNA patterns from whole slide images (WSI) is a key step in studying metastasis at scale. In this work, we collected tissue from 4 stage-III (pT3) matched colorectal cancer patients for spatial transcriptomics profiling. The Visium spatial transcriptomics (ST) assay was used to measure transcript abundance for 17 943 genes at up to 5000 55-micron (i.e., 1-10 cells) spots per patient sampled in a honeycomb pattern, co-registered with hematoxylin and eosin (H&E) stained WSI. The Visium ST assay can measure expression at these spots through tissue permeabilization of mRNAs, which are captured through spatially (i.e., x-y positional coordinates) barcoded, gene specific oligo probes. WSI subimages were extracted around each co-registered Visium spot and were used to predict the expression at these spots using machine learning models. We prototyped and compared several convolutional, transformer, and graph convolutional neural networks to predict spatial RNA patterns at the Visium spots under the hypothesis that the transformer- and graph-based approaches better capture relevant spatial tissue architecture. We further analyzed the model's ability to recapitulate spatial autocorrelation statistics using SPARK and SpatialDE. Overall, the results indicate that the transformer- and graph-based approaches were unable to outperform the convolutional neural network architecture, though they exhibited optimal performance for relevant disease-associated genes. Initial findings suggest that different neural networks that operate on different scales are relevant for capturing distinct disease pathways (e.g., epithelial to mesenchymal transition). We add further evidence that deep learning models can accurately predict gene expression in whole slide images and comment on understudied factors which may increase its external applicability (e.g., tissue context). Our preliminary work will motivate further investigation of inference for molecular patterns from whole slide images as metastasis predictors and in other applications.
Collapse
Affiliation(s)
- Michael Fatemi
- Department of Computer Science, University of Virginia, Charlottesville, VA, USA
| | - Eric Feng
- Thomas Jefferson High School for Science and Technology, Alexandria, VA, USA
| | - Cyril Sharma
- Department of Computer Science, Purdue University, West Lafayette, IN, USA
| | - Zarif Azher
- Thomas Jefferson High School for Science and Technology, Alexandria, VA, USA
| | - Tarushii Goel
- Department of Computer Science, Massachusetts Institute of Technology, Cambridge, MA, USA
| | - Ojas Ramwala
- Department of Biomedical Informatics and Medical Education, University of Washington, Seattle, WA, USA
| | - Scott M. Palisoul
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Health, Lebanon, NH, USA
| | - Rachael E. Barney
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Health, Lebanon, NH, USA
| | | | | | - Lucas A. Salas
- Department of Epidemiology, Dartmouth College Geisel School of Medicine, Hanover, NH, USA
- Department of Molecular and Systems Biology, Dartmouth College Geisel School of Medicine, Hanover, NH, USA
- Integrative Neuroscience at Dartmouth (IND) graduate program, Dartmouth College Geisel School of Medicine, Hanover, NH, USA
| | - Brock C. Christensen
- Department of Epidemiology, Dartmouth College Geisel School of Medicine, Hanover, NH, USA
- Department of Molecular and Systems Biology, Dartmouth College Geisel School of Medicine, Hanover, NH, USA
- Department of Community and Family Medicine, Dartmouth College Geisel School of Medicine, Hanover, NH, USA
| | - Gregory J. Tsongalis
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Health, Lebanon, NH, USA
| | - Louis J. Vaickus
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Health, Lebanon, NH, USA
| | - Joshua J. Levy
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Health, Lebanon, NH, USA
- Department of Epidemiology, Dartmouth College Geisel School of Medicine, Hanover, NH, USA
- Department of Dermatology, Dartmouth Health, Lebanon, NH, USA
- Program in Quantitative Biomedical Sciences, Dartmouth College Geisel School of Medicine, Hanover, NH, USA
| |
Collapse
|
17
|
Levy JJ, Liu X, Marotti JD, Kerr DA, Gutmann EJ, Glass RE, Dodge CP, Suriawinata AA, Vaickus LJ. Uncovering additional predictors of urothelial carcinoma from voided urothelial cell clusters through a deep learning-based image preprocessing technique. Cancer Cytopathol 2023; 131:19-29. [PMID: 35997513 DOI: 10.1002/cncy.22633] [Citation(s) in RCA: 4] [Impact Index Per Article: 4.0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/30/2022] [Revised: 06/19/2022] [Accepted: 06/27/2022] [Indexed: 01/04/2023]
Abstract
BACKGROUND Urine cytology is commonly used as a screening test for high-grade urothelial carcinoma for patients with risk factors or hematuria and is an essential step in longitudinal monitoring of patients with previous bladder cancer history. However, the semisubjective nature of current reporting systems for urine cytology (e.g., The Paris System) can hamper reproducibility. For instance, the incorporation of urothelial cell clusters into the classification schema is still an item of debate and perplexity among expert cytopathologists because several previous works have disputed their diagnostic relevance. METHODS In this work, an automated preprocessing tool for urothelial cell cluster assessment was developed that divides urothelial cell clusters into meaningful components for downstream assessment (ie, population-based studies, workflow automation). RESULTS In this work, an automated preprocessing tool for urothelial cell cluster assessment was developed that divides urothelial cell clusters into meaningful components for downstream assessment (ie, population-based studies, workflow automation). Results indicate that cell cluster atypia (i.e., defined by whether the cell cluster harbored multiple atypical cells, thresholded by a minimum number of cells), cell border overlap and smoothness, and total number of clusters are important markers of specimen atypia when considering assessment of urothelial cell clusters. CONCLUSIONS Markers established through techniques to separate cell clusters may have wider applicability for the design and implementation of machine learning approaches for urine cytology assessment.
Collapse
Affiliation(s)
- Joshua J Levy
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire, USA.,Department of Dermatology, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire, USA.,Department of Epidemiology, Dartmouth College Geisel School of Medicine, Hanover, New Hampshire, USA.,Program in Quantitative Biomedical Sciences, Dartmouth College Geisel School of Medicine, Hanover, New Hampshire, USA
| | - Xiaoying Liu
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire, USA.,Dartmouth College Geisel School of Medicine, Hanover, New Hampshire, USA
| | - Jonathan D Marotti
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire, USA.,Dartmouth College Geisel School of Medicine, Hanover, New Hampshire, USA
| | - Darcy A Kerr
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire, USA.,Dartmouth College Geisel School of Medicine, Hanover, New Hampshire, USA
| | - Edward J Gutmann
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire, USA.,Dartmouth College Geisel School of Medicine, Hanover, New Hampshire, USA
| | - Ryan E Glass
- University of Pennsylvania Medical Center East, Pittsburgh, Pennsylvania, USA
| | - Caroline P Dodge
- Dartmouth College Geisel School of Medicine, Hanover, New Hampshire, USA.,Cambridge Health Alliance, Cambridge, Massachusetts, USA
| | - Arief A Suriawinata
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire, USA.,Dartmouth College Geisel School of Medicine, Hanover, New Hampshire, USA
| | - Louis J Vaickus
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire, USA.,Dartmouth College Geisel School of Medicine, Hanover, New Hampshire, USA
| |
Collapse
|
18
|
Copeland-Halperin LR, Divakar P, Stewart T, Demsas F, Levy JJ, Nigriny JF, Paydarfar JA. Predictors of Gastrostomy Tube Placement in Head and Neck Cancer Patients at a Rural Tertiary Care Hospital. J Reconstr Microsurg Open 2023. [DOI: 10.1055/s-0043-1760757] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 02/12/2023] Open
Abstract
Abstract
Background Head and neck cancer is a leading cause of cancer. Treatment often requires surgical resection, free-flap reconstruction, radiation, and/or chemotherapy. Tumor burden and pain may limit swallowing and impair nutrition, increasing complications and mortality. Patients commonly require gastrostomy tubes (G-tube), but predicting which patients are in need remains elusive. This study identifies predictors of G-tube among head and neck cancer patients undergoing immediate free-flap reconstruction.
Methods Institutional Review Board approval was obtained. Retrospective database review was performed of patients at 18 years of age or older with head and neck cancer who underwent resection with immediate free-flap reconstruction from 2011 to 2019. Patients who underwent nonfree-flap or delayed reconstruction or with mortality within 7 days postoperatively were excluded. Patient demographics and comorbidities, tumor/treatment characteristics, and need for G-tube were analyzed to identify univariate and multivariate predictors.
Results In total, 107 patients were included and 72 required G-tube placement. On multivariate analysis, tracheostomy (odds ratio [OR]: 81.78; confidence interval [CI]: 7.43–1,399.92; p < 0.01), anterolateral thigh flap reconstruction (OR: 16.18; CI: 1.14–429.66; p = 0.04), and age 65 years or younger (OR: 9.35; CI: 1.47–89.11; p = 0.02) were predictors of G-tube placement.
Conclusion Head and neck cancer treatment commonly involves extensive resection, reconstruction, and/or chemoradiation. These patients are at high risk for malnutrition and need G-tube. Determining who requires a pre- or postoperative G-tube remains a challenge. In this study, the need for tracheostomy or ALT flap reconstruction and age 65 years or younger were predictive of postoperative G-tube placement. Future research will guide a multidisciplinary perioperative pathway to facilitate the optimization of nutrition management.
Collapse
Affiliation(s)
| | - Prashanthi Divakar
- Department of Surgery, Memorial Sloan Kettering Cancer Center, New York, New York
| | - Talia Stewart
- Department of Surgery, MetroHealth Medical Center, Cleveland, Ohio
| | - Falen Demsas
- Department of Surgery, The Massachusetts General Hospital, Boston, Massachusetts
| | - Joshua J. Levy
- Department of Biomedical Sciences, Geisel School of Medicine, Hanover, New Hampshire
| | - John F. Nigriny
- Department of Surgery, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire
| | - Joseph A. Paydarfar
- Department of Surgery, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire
| |
Collapse
|
19
|
Hong J, Quon RJ, Song Y, Xie T, Levy JJ, D'Agostino E, Camp EJ, Roberts DW, Jobst BC. Functional Reorganization of the Mesial Frontal Premotor Cortex in Patients With Supplementary Motor Area Seizures. Neurosurgery 2023; 92:186-194. [PMID: 36255216 DOI: 10.1227/neu.0000000000002172] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/01/2021] [Accepted: 07/29/2022] [Indexed: 12/25/2022] Open
Abstract
BACKGROUND Direct cortical stimulation of the mesial frontal premotor cortex, including the supplementary motor area (SMA), is challenging in humans. Limited access to these brain regions impedes understanding of human premotor cortex functional organization and somatotopy. OBJECTIVE To test whether seizure onset within the SMA was associated with functional remapping of mesial frontal premotor areas in a cohort of patients with epilepsy who underwent awake brain mapping after implantation of interhemispheric subdural electrodes. METHODS Stimulation trials from 646 interhemispheric subdural electrodes were analyzed and compared between patients who had seizure onset in the SMA (n = 13) vs patients who had seizure onset outside of the SMA (n = 12). 1:1 matching with replacement between SMA and non-SMA data sets was used to ensure similar spatial distribution of electrodes. Centroids and 95% confidence regions were computed for clustered head, trunk, upper extremity, lower extremity, and vision responses. A generalized linear mixed-effects model was used to test for significant differences in the resulting functional maps. Clinical, radiographic, and histopathologic data were reviewed. RESULTS After analyzing direct cortical stimulation trials from interhemispheric electrodes, we found significant displacement of the head and trunk responses in SMA compared with non-SMA patients ( P < .01 for both). These differences remained significant after accounting for structural lesions, preexisting motor deficits, and seizure outcome. CONCLUSION The somatotopy of the mesial frontal premotor regions is significantly altered in patients who have SMA-onset seizures compared with patients who have seizure onset outside of the SMA, suggesting that functional remapping can occur in these brain regions.
Collapse
Affiliation(s)
- Jennifer Hong
- Department of Surgery, Section of Neurosurgery, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire, USA
| | - Robert J Quon
- Department of Quantitative Biomedical Sciences, Geisel School of Medicine at Dartmouth, Hanover, New Hampshire, USA
| | - Yinchen Song
- Department of Neurology, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire, USA
| | - Tiankang Xie
- Department of Quantitative Biomedical Sciences, Geisel School of Medicine at Dartmouth, Hanover, New Hampshire, USA
| | - Joshua J Levy
- Department of Quantitative Biomedical Sciences, Geisel School of Medicine at Dartmouth, Hanover, New Hampshire, USA
| | - Erin D'Agostino
- Department of Surgery, Section of Neurosurgery, University of Vermont Medical Center, Burlington, Vermont, USA
| | - Edward J Camp
- Department of Neurology, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire, USA
| | - David W Roberts
- Department of Surgery, Section of Neurosurgery, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire, USA
| | - Barbara C Jobst
- Department of Neurology, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire, USA
| |
Collapse
|
20
|
Farhadi F, Barnes MR, Sugito HR, Sin JM, Henderson ER, Levy JJ. Applications of artificial intelligence in orthopaedic surgery. Front Med Technol 2022; 4:995526. [PMID: 36590152 PMCID: PMC9797865 DOI: 10.3389/fmedt.2022.995526] [Citation(s) in RCA: 9] [Impact Index Per Article: 4.5] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/16/2022] [Accepted: 11/28/2022] [Indexed: 12/23/2022] Open
Abstract
The practice of medicine is rapidly transforming as a result of technological breakthroughs. Artificial intelligence (AI) systems are becoming more and more relevant in medicine and orthopaedic surgery as a result of the nearly exponential growth in computer processing power, cloud based computing, and development, and refining of medical-task specific software algorithms. Because of the extensive role of technologies such as medical imaging that bring high sensitivity, specificity, and positive/negative prognostic value to management of orthopaedic disorders, the field is particularly ripe for the application of machine-based integration of imaging studies, among other applications. Through this review, we seek to promote awareness in the orthopaedics community of the current accomplishments and projected uses of AI and ML as described in the literature. We summarize the current state of the art in the use of ML and AI in five key orthopaedic disciplines: joint reconstruction, spine, orthopaedic oncology, trauma, and sports medicine.
Collapse
Affiliation(s)
- Faraz Farhadi
- Geisel School of Medicine, Dartmouth College, Hanover, NH, United States,Radiology and Imaging Sciences, National Institutes of Health (NIH), Bethesda, United States,Correspondence: Faraz Farhadi Joshua J. Levy
| | - Matthew R. Barnes
- Geisel School of Medicine, Dartmouth College, Hanover, NH, United States
| | - Harun R. Sugito
- Geisel School of Medicine, Dartmouth College, Hanover, NH, United States
| | - Jessica M. Sin
- Department of Radiology, Dartmouth Health, Lebanon, United States
| | - Eric R. Henderson
- Department of Orthopaedics, Dartmouth Health, Lebanon, United States
| | - Joshua J. Levy
- Department of Pathology and Laboratory Medicine, Dartmouth Health, Lebanon, NH, United States,Correspondence: Faraz Farhadi Joshua J. Levy
| |
Collapse
|
21
|
Levy JJ, Liu X, Marotti JD, Kerr DA, Gutmann EJ, Glass RE, Dodge CP, Vaickus LJ. Large-scale longitudinal comparison of urine cytological classification systems reveals potential early adoption of The Paris System criteria. J Am Soc Cytopathol 2022; 11:394-402. [PMID: 36068164 DOI: 10.1016/j.jasc.2022.08.001] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/19/2022] [Revised: 07/27/2022] [Accepted: 08/09/2022] [Indexed: 06/15/2023]
Abstract
INTRODUCTION Urine cytology is used to screen for urothelial carcinoma in patients with hematuria or risk factors (eg, smoking, industrial dye exposure) and is an essential clinical triage and longitudinal monitoring tool for patients with known bladder cancer. However, urine cytology is semisubjective and thus susceptible to issues including specimen quality, interobserver variability, and "hedging" towards equivocal ("atypical") diagnoses. These factors limit the predictive value of urine cytology and increase reliance on invasive procedures (cystoscopy). The Paris System for Reporting Urine Cytology (TPS) was formulated to provide more quantitative/reproducible endpoints with well-defined criteria for urothelial atypia. TPS is often compared to other assessment techniques to justify its adoption. TPS results in decreased use of the atypical category and better reproducibility. Previous reports comparing diagnoses pre- and post-TPS have not considered temporal differences between diagnoses made under prior systems and TPS. By aggregating across time, studies may underestimate the magnitude of differences between assessment methods. MATERIALS AND METHODS We conducted a large-scale longitudinal reassessment of urine cytology using TPS criteria from specimens collected from 2008 to 2018, prior to the mid-2018 adoption of TPS at an academic medical center. RESULTS Findings indicate that differences in atypical assignment were largest at the start of the period and these differences progressively decreased towards insignificance just prior to TPS implementation. CONCLUSIONS This finding suggests that cytopathologists had begun to utilize the quantitative TPS criteria prior to official adoption, which may more broadly inform adoption strategies, communication, and understanding for evolving classification systems in cytology.
Collapse
Affiliation(s)
- Joshua J Levy
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire; Department of Dermatology, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire; Department of Epidemiology, Dartmouth College Geisel School of Medicine, Hanover, New Hampshire; Program in Quantitative Biomedical Sciences, Dartmouth College Geisel School of Medicine, Hanover, New Hampshire.
| | - Xiaoying Liu
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire; Dartmouth College Geisel School of Medicine, Hanover, New Hampshire
| | - Jonathan D Marotti
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire; Dartmouth College Geisel School of Medicine, Hanover, New Hampshire
| | - Darcy A Kerr
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire; Dartmouth College Geisel School of Medicine, Hanover, New Hampshire
| | - Edward J Gutmann
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire; Dartmouth College Geisel School of Medicine, Hanover, New Hampshire
| | | | - Caroline P Dodge
- Dartmouth College Geisel School of Medicine, Hanover, New Hampshire
| | - Louis J Vaickus
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire; Dartmouth College Geisel School of Medicine, Hanover, New Hampshire
| |
Collapse
|
22
|
Stewart T, Copeland-Halperin LR, Demsas F, Divakar P, Shank N, Blunt H, J Levy J, Nigriny JF, Paydarfar JA. Predictors of gastrostomy tube placement in patients with head and neck cancer undergoing resection and flap-based reconstruction: systematic review and meta-analysis. J Plast Reconstr Aesthet Surg 2022; 79:1-10. [PMID: 36780787 DOI: 10.1016/j.bjps.2022.08.040] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/13/2022] [Accepted: 08/17/2022] [Indexed: 10/15/2022]
Abstract
BACKGROUND Nutritional status may be impaired in patients with head and neck cancer undergoing surgical treatment, often necessitating gastrostomy tube (G-tube) placement. Identifying which patients will require a G-tube remains a challenge. This study identifies predictors of G-tube requirement in patients undergoing tumor resection and reconstruction with pedicled or free flaps. METHODS Systematic review of the PubMed, Cochrane, and Scopus databases was performed of English language articles, discussing risk factors of perioperative G-tube placement among patients >18 years. Data on patient, tumor, and treatment factors, as well as need for G-tube, were collected. Univariable meta-analysis was conducted to identify predictors for G-tube placement. RESULTS Eleven studies (1,112 patients) met inclusion criteria. Overall pooled prevalence of postoperative G-tube placement was 25%. Patients with advanced cancer stage IV/recurrence were more likely to require a G-tube (OR 2.81 [CI 1.03-7.69]; p<0.05), as were those who had undergone preoperative radiation (OR 3.55 [CI 2.03-6.20], p<0.05). Reconstruction with a radial forearm free flap was associated with a lower need for G-tube versus rectus abdominis (OR 0.25 [CI 0.08-0.83], p=0.02) and latissimus dorsi flap (OR 0.21 [CI 0.04-1.09], p=0.06). There was no difference in G-tube placement between those receiving pedicled flaps versus free flaps (OR 1.54 [CI 0.38-6.20], p=0.54). CONCLUSIONS Among patients with head and neck cancer undergoing resection with immediate pedicled or free flap reconstruction, advanced tumor stage and history of prior radiation therapy are associated with increased likelihood of G-tube placement. More randomized controlled trials are needed to develop a decision-making algorithm.
Collapse
Affiliation(s)
| | | | | | | | - Nina Shank
- Dartmouth-Hitchcock Medical Center; Lebanon, NH
| | - Heather Blunt
- Department of Quantitative Biomedical Sciences, Geisel School of Medicine; Hanover, NH
| | | | | | | |
Collapse
|
23
|
Ondrasik RM, Khan J, Szczepiorkowski ZM, Levy JJ, Dunbar NM. Passive order auditing associated with reductions in red blood cell utilization: National blood shortage experience. Transfusion 2022; 62:1551-1558. [PMID: 35815525 DOI: 10.1111/trf.17008] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/18/2022] [Revised: 06/22/2022] [Accepted: 06/22/2022] [Indexed: 01/28/2023]
Abstract
BACKGROUND Decreased blood collection during the Coronavirus Disease 2019 (COVID-19) pandemic resulted in long-term red blood cell (RBC) shortages in the United States. In an effort to conserve RBCs, the existing passive alert system for auditing inpatient transfusions was modified to activate at a lower hemoglobin threshold (6.5 g/dL instead of 7.0 g/dL for stable, nonbleeding inpatients) during a 9-month shortage at an academic medical center. Hemoglobin levels prior to RBC transfusions were compared for inpatients receiving RBC transfusions to determine whether RBC utilization changed during the intervention. STUDY DESIGN AND METHODS This retrospective study compared the number of single-unit RBC transfusions and hemoglobin levels prior to RBC transfusion among inpatients during the 9 months of the intervention (Period 2, 06/01/2021-2/28/2022) to the same period of the previous year (Period 1, 06/01/2020-2/28/2021). RESULTS Overall full unit RBC transfusions to inpatients decreased by 15% from 5182 to 4421. Of all transfusions, 50.3% and 49.8% were single-unit RBC transfusions in Period 1 and Period 2, respectively. The incidence rate difference and incidence rate ratio of single RBC units transfused per 1000 patient days were significantly decreased (p = 0.0007). The average pre-transfusion hemoglobin level significantly decreased from 7.18 g/dL to 7.05 g/dL (p = 0.0002), largely due to significant decreases in hemoglobin transfusion triggers for adult inpatient ward transfusions. DISCUSSION Modification of the passive alert system was associated with significantly decreased RBC utilization during a long-term RBC shortage. Modification of transfusion criteria recommended by passive alerts may be a feasible option to decrease RBC utilization at centers during long-term RBC shortages.
Collapse
Affiliation(s)
- Regina M Ondrasik
- Department of Pathology and Laboratory Medicine, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire, USA
| | - Jenna Khan
- Department of Pathology and Laboratory Medicine, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire, USA.,Department of Medicine, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire, USA.,Department of Pathology and Laboratory Medicine, Geisel School of Medicine at Dartmouth, Hanover, New Hampshire, USA
| | - Zbigniew M Szczepiorkowski
- Department of Pathology and Laboratory Medicine, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire, USA.,Department of Medicine, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire, USA.,Department of Pathology and Laboratory Medicine, Geisel School of Medicine at Dartmouth, Hanover, New Hampshire, USA.,Division of Transfusion Medicine, Institute of Hematology and Transfusion Medicine, Warsaw, Poland
| | - Joshua J Levy
- Department of Pathology and Laboratory Medicine, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire, USA.,Department of Pathology and Laboratory Medicine, Geisel School of Medicine at Dartmouth, Hanover, New Hampshire, USA
| | - Nancy M Dunbar
- Department of Pathology and Laboratory Medicine, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire, USA.,Department of Medicine, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire, USA.,Department of Pathology and Laboratory Medicine, Geisel School of Medicine at Dartmouth, Hanover, New Hampshire, USA
| |
Collapse
|
24
|
Levy JJ, Lima JF, Miller MW, Freed GL, O'Malley AJ, Emeny RT. Machine Learning Approaches for Hospital Acquired Pressure Injuries: A Retrospective Study of Electronic Medical Records. Front Med Technol 2022; 4:926667. [PMID: 35782577 PMCID: PMC9243224 DOI: 10.3389/fmedt.2022.926667] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/22/2022] [Accepted: 05/24/2022] [Indexed: 11/24/2022] Open
Abstract
Background Many machine learning heuristics integrate well with Electronic Medical Record (EMR) systems yet often fail to surpass traditional statistical models for biomedical applications. Objective We sought to compare predictive performances of 12 machine learning and traditional statistical techniques to predict the occurrence of Hospital Acquired Pressure Injuries (HAPI). Methods EMR information was collected from 57,227 hospitalizations acquired from Dartmouth Hitchcock Medical Center (April 2011 to December 2016). Twelve classification algorithms, chosen based upon classic regression and recent machine learning techniques, were trained to predict HAPI incidence and performance was assessed using the Area Under the Receiver Operating Characteristic Curve (AUC). Results Logistic regression achieved a performance (AUC = 0.91 ± 0.034) comparable to the other machine learning approaches. We report discordance between machine learning derived predictors compared to the traditional statistical model. We visually assessed important patient-specific factors through Shapley Additive Explanations. Conclusions Machine learning models will continue to inform clinical decision-making processes but should be compared to traditional modeling approaches to ensure proper utilization. Disagreements between important predictors found by traditional and machine learning modeling approaches can potentially confuse clinicians and need to be reconciled. These developments represent important steps forward in developing real-time predictive models that can be integrated into EMR systems to reduce unnecessary harm.
Collapse
Affiliation(s)
- Joshua J. Levy
- Department of Epidemiology, Geisel School of Medicine at Dartmouth, Hanover, NH, United States
- Department of Pathology, Dartmouth Hitchcock Medical Center, Lebanon, NH, United States
- Quantitative Biomedical Sciences, Geisel School of Medicine at Dartmouth, Hanover, NH, United States
| | - Jorge F. Lima
- Quantitative Biomedical Sciences, Geisel School of Medicine at Dartmouth, Hanover, NH, United States
| | - Megan W. Miller
- Department of Wound Care Services, Dartmouth Hitchcock Medical Center, Lebanon, NH, United States
| | - Gary L. Freed
- Department of Wound Care Services, Dartmouth Hitchcock Medical Center, Lebanon, NH, United States
- Department of Plastic Surgery, Dartmouth Hitchcock Medical Center, Lebanon, NH, United States
| | - A. James O'Malley
- Department of Biomedical Data Science, Geisel School of Medicine at Dartmouth, Hanover, NH, United States
- The Dartmouth Institute for Health Policy and Clinical Practice, Geisel School of Medicine at Dartmouth, Hanover, NH, United States
| | - Rebecca T. Emeny
- The Dartmouth Institute for Health Policy and Clinical Practice, Geisel School of Medicine at Dartmouth, Hanover, NH, United States
| |
Collapse
|
25
|
Kelliher MT, Levy JJ, Nerenz RD, Poore B, Johnston AA, Rogers AR, Stella MEO, Snow SE, Cervinski MA, Hubbard JA. Comparison of Symptoms and Antibody Response Following Administration of Moderna or Pfizer SARS-CoV-2 Vaccines. Arch Pathol Lab Med 2022; 146:677-685. [PMID: 35188563 DOI: 10.5858/arpa.2021-0607-sa] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.5] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Accepted: 02/14/2022] [Indexed: 11/06/2022]
Abstract
CONTEXT. – Moderna (mRNA-1272) and Pfizer (BNT162b2) SARS-CoV-2 vaccines demonstrate favorable safety and efficacy profiles, but direct comparison data is lacking. OBJECTIVE. – To determine the vaccines' side effect profiles and expected antibody responses. This data may help personalize vaccine selection and identify individuals with a suboptimal vaccine response. DESIGN. – One hundred forty-nine healthy, largely seronegative adults were assigned Moderna (n=79) or Pfizer (n=70). Following the second dose, participants completed a survey documenting their side effects. Serum was collected 0-4 days prior to dose 2, 14±4 days, 30±4 days, 90±10 days, and 180±20 days after dose 2. Convalescent serum specimens were collected 32-54 days from donors after a polymerase chain reaction (PCR)-confirmed SARS-CoV-2 infection (n=20). Anti-spike antibodies were measured using the Roche Diagnostics Elecys anti-SARS-CoV-2 S assay on a Roche cobas e801 instrument. RESULTS. – Participants receiving the Moderna vaccine experienced side effects with greater frequency and severity. Both vaccines elicited a robust antibody response but median signal was higher in Moderna recipients. Symptom severity decreased with age. Antibody response in Pfizer recipients negatively correlated with age. Antibody response decreased after 6 months (84% reduction in Moderna, 79% Pfizer), but values remained greater than for convalescent donors. Antibody response did not correlate with gender or symptom severity. CONCLUSIONS. – Moderna may be preferred in individuals in need of greater immune stimulation (e.g. older individuals) while Pfizer may be preferred in those concerned about vaccine reactions. Anti-spike antibody signal varies by vaccine, so specific reference intervals will be needed to identify individuals with a suboptimal response.
Collapse
Affiliation(s)
- Michael T Kelliher
- Department of Pathology and Laboratory Medicine, Dartmouth-Hitchcock Health System, Lebanon, NH.,Department of Pathology of Laboratory Medicine, The Geisel School of Medicine at Dartmouth, Hanover, NH
| | - Joshua J Levy
- Department of Pathology and Laboratory Medicine, Dartmouth-Hitchcock Health System, Lebanon, NH.,Department of Pathology of Laboratory Medicine, The Geisel School of Medicine at Dartmouth, Hanover, NH
| | - Robert D Nerenz
- Department of Pathology and Laboratory Medicine, Dartmouth-Hitchcock Health System, Lebanon, NH.,Department of Pathology of Laboratory Medicine, The Geisel School of Medicine at Dartmouth, Hanover, NH
| | - Bradley Poore
- Department of Pathology and Laboratory Medicine, Dartmouth-Hitchcock Health System, Lebanon, NH.,Department of Pathology of Laboratory Medicine, The Geisel School of Medicine at Dartmouth, Hanover, NH
| | - Abigail A Johnston
- Department of Pathology and Laboratory Medicine, Dartmouth-Hitchcock Health System, Lebanon, NH.,Department of Pathology of Laboratory Medicine, The Geisel School of Medicine at Dartmouth, Hanover, NH
| | - Amanda R Rogers
- Department of Pathology and Laboratory Medicine, Dartmouth-Hitchcock Health System, Lebanon, NH.,Department of Pathology of Laboratory Medicine, The Geisel School of Medicine at Dartmouth, Hanover, NH
| | - Mary E O Stella
- Department of Pathology and Laboratory Medicine, Dartmouth-Hitchcock Health System, Lebanon, NH.,Department of Pathology of Laboratory Medicine, The Geisel School of Medicine at Dartmouth, Hanover, NH
| | - Sarah E Snow
- Department of Pathology and Laboratory Medicine, Dartmouth-Hitchcock Health System, Lebanon, NH.,Department of Pathology of Laboratory Medicine, The Geisel School of Medicine at Dartmouth, Hanover, NH
| | - Mark A Cervinski
- Department of Pathology and Laboratory Medicine, Dartmouth-Hitchcock Health System, Lebanon, NH.,Department of Pathology of Laboratory Medicine, The Geisel School of Medicine at Dartmouth, Hanover, NH
| | - Jacqueline A Hubbard
- Department of Pathology and Laboratory Medicine, Dartmouth-Hitchcock Health System, Lebanon, NH.,Department of Pathology of Laboratory Medicine, The Geisel School of Medicine at Dartmouth, Hanover, NH
| |
Collapse
|
26
|
Levy JJ, Bobak CA, Nasir-Moin M, Veziroglu EM, Palisoul SM, Barney RE, Salas LA, Christensen BC, Tsongalis GJ, Vaickus LJ. Mixed Effects Machine Learning Models for Colon Cancer Metastasis Prediction using Spatially Localized Immuno-Oncology Markers. Pac Symp Biocomput 2022; 27:175-186. [PMID: 34890147 PMCID: PMC8669762] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Grants] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Indexed: 11/17/2022]
Abstract
Spatially resolved characterization of the transcriptome and proteome promises to provide further clarity on cancer pathogenesis and etiology, which may inform future clinical practice through classifier development for clinical outcomes. However, batch effects may potentially obscure the ability of machine learning methods to derive complex associations within spatial omics data. Profiling thirty-five stage three colon cancer patients using the GeoMX Digital Spatial Profiler, we found that mixed-effects machine learning (MEML) methods† may provide utility for overcoming significant batch effects to communicate key and complex disease associations from spatial information. These results point to further exploration and application of MEML methods within the spatial omics algorithm development life cycle for clinical deployment.
Collapse
Affiliation(s)
- Joshua J Levy
- Quantitative Biomedical Sciences, Geisel School of Medicine at Dartmouth, Lebanon, NH 03756, USA,
| | | | | | | | | | | | | | | | | | | |
Collapse
|
27
|
Levy JJ, Lebeaux RM, Hoen AG, Christensen BC, Vaickus LJ, MacKenzie TA. Using Satellite Images and Deep Learning to Identify Associations Between County-Level Mortality and Residential Neighborhood Features Proximal to Schools: A Cross-Sectional Study. Front Public Health 2021; 9:766707. [PMID: 34805078 PMCID: PMC8602058 DOI: 10.3389/fpubh.2021.766707] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/29/2021] [Accepted: 10/06/2021] [Indexed: 12/15/2022] Open
Abstract
What is the relationship between mortality and satellite images as elucidated through the use of Convolutional Neural Networks? Background: Following a century of increase, life expectancy in the United States has stagnated and begun to decline in recent decades. Using satellite images and street view images, prior work has demonstrated associations of the built environment with income, education, access to care, and health factors such as obesity. However, assessment of learned image feature relationships with variation in crude mortality rate across the United States has been lacking. Objective: We sought to investigate if county-level mortality rates in the U.S. could be predicted from satellite images. Methods: Satellite images of neighborhoods surrounding schools were extracted with the Google Static Maps application programming interface for 430 counties representing ~68.9% of the US population. A convolutional neural network was trained using crude mortality rates for each county in 2015 to predict mortality. Learned image features were interpreted using Shapley Additive Feature Explanations, clustered, and compared to mortality and its associated covariate predictors. Results: Predicted mortality from satellite images in a held-out test set of counties was strongly correlated to the true crude mortality rate (Pearson r = 0.72). Direct prediction of mortality using a deep learning model across a cross-section of 430 U.S. counties identified key features in the environment (e.g., sidewalks, driveways, and hiking trails) associated with lower mortality. Learned image features were clustered, and we identified 10 clusters that were associated with education, income, geographical region, race, and age. Conclusions: The application of deep learning techniques to remotely-sensed features of the built environment can serve as a useful predictor of mortality in the United States. Although we identified features that were largely associated with demographic information, future modeling approaches that directly identify image features associated with health-related outcomes have the potential to inform targeted public health interventions.
Collapse
Affiliation(s)
- Joshua J. Levy
- Department of Epidemiology, Geisel School of Medicine at Dartmouth, Lebanon, NH, United States
- Program in Quantitative Biomedical Sciences, Geisel School of Medicine at Dartmouth, Lebanon, NH, United States
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, NH, United States
| | - Rebecca M. Lebeaux
- Department of Epidemiology, Geisel School of Medicine at Dartmouth, Lebanon, NH, United States
- Program in Quantitative Biomedical Sciences, Geisel School of Medicine at Dartmouth, Lebanon, NH, United States
| | - Anne G. Hoen
- Department of Epidemiology, Geisel School of Medicine at Dartmouth, Lebanon, NH, United States
- Program in Quantitative Biomedical Sciences, Geisel School of Medicine at Dartmouth, Lebanon, NH, United States
- Department of Biomedical Data Science, Geisel School of Medicine at Dartmouth, Lebanon, NH, United States
| | - Brock C. Christensen
- Department of Epidemiology, Geisel School of Medicine at Dartmouth, Lebanon, NH, United States
| | - Louis J. Vaickus
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, NH, United States
| | - Todd A. MacKenzie
- Department of Epidemiology, Geisel School of Medicine at Dartmouth, Lebanon, NH, United States
- Department of Biomedical Data Science, Geisel School of Medicine at Dartmouth, Lebanon, NH, United States
- The Dartmouth Institute for Health Policy and Clinical Practice, Geisel School of Medicine at Dartmouth, Lebanon, NH, United States
| |
Collapse
|
28
|
Glass RE, Marotti JD, Kerr DA, Levy JJ, Vaickus LJ, Gutmann EJ, Tafe LJ, Motanagh SA, Sorensen MJ, Davies L, Liu X. Using molecular testing to improve the management of thyroid nodules with indeterminate cytology: an institutional experience with review of molecular alterations. J Am Soc Cytopathol 2021; 11:79-86. [PMID: 34627720 DOI: 10.1016/j.jasc.2021.08.004] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/18/2021] [Revised: 08/09/2021] [Accepted: 08/11/2021] [Indexed: 12/12/2022]
Abstract
INTRODUCTION Molecular testing has helped clinicians and cytopathologists to further categorize indeterminate thyroid fine needle aspiration (FNA) specimens. The purpose of the present study was to evaluate the accuracy of commercially available molecular tests, review their effects on patient treatment, and correlate the molecular alterations with the histologic findings. MATERIALS AND METHODS A pathology laboratory information system search identified thyroid FNAs performed at our institution between January 1, 2015 and June 30, 2020. The results of surgical follow-up and ancillary molecular testing were collected. We evaluated the accuracy of these tests and whether they could reduce the number of surgeries performed. RESULTS Our laboratory information system search identified 510 cases reported as atypia of undetermined significance, 94 as suspicious for follicular neoplasm, and 44 as suspicious for follicular neoplasm, Hurthle cell type. Of the specimens, 343 had no ancillary molecular testing, 146 were sent for ThyGenX/ThyraMIR, and 136 were sent for ThyroSeq. Of the patients without molecular testing, 50.4% had undergone follow-up surgery compared with 30.8% after ThyGenX/ThyraMIR and 38.2% after ThyroSeq testing, resulting in 38.9% and 24.2% fewer surgeries and an odds ratio of 0.04 (95% confidence interval, 0.00-0.33) and 0.14 (95% confidence interval, 0.01-0.95), respectively. For ThyGenX/ThyraMIR testing, the risk of malignancy for high and moderate to high risk alterations was 80%, 28.6% for moderate and low to moderate risk alterations, and 23.1% for low risk alterations. For ThyroSeq, the risk of malignancy was 87.5% for high risk alterations, 36.8% for intermediate to high risk alterations, 27.3% for intermediate risk alterations, and 0% for low risk alterations. The areas under the curve for ThyGenX/ThyraMIR and ThyroSeq testing were 0.65 and 0.85, respectively. CONCLUSIONS These findings suggest that, at our institution, both ThygenX/ThyraMIR and ThyroSeq can be used to effectively stratify cytology specimens based on the risk of malignancy and reduce the number of surgeries performed at our institution.
Collapse
Affiliation(s)
- Ryan E Glass
- Department of Pathology and Laboratory Medicine, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire; Geisel School of Medicine at Dartmouth, Hanover, New Hampshire.
| | - Jonathan D Marotti
- Department of Pathology and Laboratory Medicine, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire; Geisel School of Medicine at Dartmouth, Hanover, New Hampshire
| | - Darcy A Kerr
- Department of Pathology and Laboratory Medicine, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire; Geisel School of Medicine at Dartmouth, Hanover, New Hampshire
| | - Joshua J Levy
- Geisel School of Medicine at Dartmouth, Hanover, New Hampshire
| | - Louis J Vaickus
- Department of Pathology and Laboratory Medicine, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire; Geisel School of Medicine at Dartmouth, Hanover, New Hampshire
| | - Edward J Gutmann
- Department of Pathology and Laboratory Medicine, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire; Geisel School of Medicine at Dartmouth, Hanover, New Hampshire
| | - Laura J Tafe
- Department of Pathology and Laboratory Medicine, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire; Geisel School of Medicine at Dartmouth, Hanover, New Hampshire
| | - Samaneh A Motanagh
- Department of Pathology and Laboratory Medicine, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire
| | - Meredith J Sorensen
- Department of Endocrine Surgery, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire
| | - Louise Davies
- Department of Endocrine Surgery, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire
| | - Xiaoying Liu
- Department of Pathology and Laboratory Medicine, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire; Geisel School of Medicine at Dartmouth, Hanover, New Hampshire
| |
Collapse
|
29
|
Levy JJ, Chen Y, Azizgolshani N, Petersen CL, Titus AJ, Moen EL, Vaickus LJ, Salas LA, Christensen BC. MethylSPWNet and MethylCapsNet: Biologically Motivated Organization of DNAm Neural Networks, Inspired by Capsule Networks. NPJ Syst Biol Appl 2021; 7:33. [PMID: 34417465 PMCID: PMC8379254 DOI: 10.1038/s41540-021-00193-7] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/17/2020] [Accepted: 07/01/2021] [Indexed: 02/07/2023] Open
Abstract
DNA methylation (DNAm) alterations have been heavily implicated in carcinogenesis and the pathophysiology of diseases through upstream regulation of gene expression. DNAm deep-learning approaches are able to capture features associated with aging, cell type, and disease progression, but lack incorporation of prior biological knowledge. Here, we present modular, user-friendly deep-learning methodology and software, MethylCapsNet and MethylSPWNet, that group CpGs into biologically relevant capsules-such as gene promoter context, CpG island relationship, or user-defined groupings-and relate them to diagnostic and prognostic outcomes. We demonstrate these models' utility on 3,897 individuals in the classification of central nervous system (CNS) tumors. MethylCapsNet and MethylSPWNet provide an opportunity to increase DNAm deep-learning analyses' interpretability by enabling a flexible organization of DNAm data into biologically relevant capsules.
Collapse
Affiliation(s)
- Joshua J Levy
- Program in Quantitative Biomedical Sciences, Geisel School of Medicine at Dartmouth, Hanover, NH, USA.
- Department of Epidemiology, Geisel School of Medicine at Dartmouth, Hanover, NH, USA.
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, NH, USA.
| | - Youdinghuan Chen
- Program in Quantitative Biomedical Sciences, Geisel School of Medicine at Dartmouth, Hanover, NH, USA
- Department of Epidemiology, Geisel School of Medicine at Dartmouth, Hanover, NH, USA
| | - Nasim Azizgolshani
- Department of Epidemiology, Geisel School of Medicine at Dartmouth, Hanover, NH, USA
| | - Curtis L Petersen
- Department of Epidemiology, Geisel School of Medicine at Dartmouth, Hanover, NH, USA
- The Dartmouth Institute for Health Policy and Clinical Practice, Lebanon, NH, USA
| | - Alexander J Titus
- Department of Life Sciences, University of New Hampshire, Manchester, NH, USA
| | - Erika L Moen
- The Dartmouth Institute for Health Policy and Clinical Practice, Lebanon, NH, USA
- Department of Biomedical Data Science, Geisel School of Medicine at Dartmouth, Hanover, NH, USA
| | - Louis J Vaickus
- Emerging Diagnostic and Investigative Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, NH, USA
| | - Lucas A Salas
- Department of Epidemiology, Geisel School of Medicine at Dartmouth, Hanover, NH, USA
- Department of Molecular and Systems Biology, Geisel School of Medicine at Dartmouth, Hanover, NH, USA
| | - Brock C Christensen
- Department of Epidemiology, Geisel School of Medicine at Dartmouth, Hanover, NH, USA
- Department of Molecular and Systems Biology, Geisel School of Medicine at Dartmouth, Hanover, NH, USA
- Department of Community and Family Medicine, Geisel School of Medicine at Dartmouth, Hanover, NH, USA
| |
Collapse
|
30
|
Glass RE, Levy JJ, Motanagh SA, Vaickus LJ, Liu X. Atypia of undetermined significance in thyroid cytology: Nuclear atypia and architectural atypia are associated with different molecular alterations and risks of malignancy. Cancer Cytopathol 2021; 129:966-972. [PMID: 34399035 DOI: 10.1002/cncy.22495] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/07/2021] [Revised: 07/08/2021] [Accepted: 07/11/2021] [Indexed: 12/21/2022]
Abstract
BACKGROUND The Bethesda System for Reporting Thyroid Cytopathology contains an atypia of undetermined significance (AUS) category with heterogeneous and distinct inclusion criteria. The purpose of this study was to investigate differences in malignancy rates and molecular alterations based on the presence of different criteria. METHODS A laboratory information search was conducted to identify thyroid fine-needle aspiration specimens signed out as AUS. The cases were reclassified as architectural atypia (3A), cytologic atypia (3C), both architectural and cytologic atypia (3B), or Hürthle cell aspirate (3H). Surgical follow-up and concurrent molecular test results, if available, were collected. RESULTS Five hundred ten specimens, including 258 reclassified as 3A, 40 reclassified as 3B, 119 reclassified as 3C, and 86 reclassified as 3H, were identified. The risks of malignancy for the subcategories were 13.4%, 26.3%, 44.1%, and 13.8%, respectively. Additionally, BRAF V600E mutations were more prevalent in specimens with cytologic atypia (3B/3C), whereas low-risk alterations, including KRAS, PTEN, and PAX8-PPARy2, were more prevalent in those with architectural atypia (3A). CONCLUSIONS Subdividing AUS specimens on the basis of the type of atypia can yield categories associated with distinct molecular alterations and risks of malignancy.
Collapse
Affiliation(s)
- Ryan E Glass
- Department of Pathology, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire
| | - Joshua J Levy
- Department of Quantitative Biomedical Sciences, Lebanon, New Hampshire
| | - Samaneh A Motanagh
- Department of Pathology, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire
| | - Louis J Vaickus
- Department of Pathology, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire
| | - Xiaoying Liu
- Department of Pathology, Dartmouth Hitchcock Medical Center, Lebanon, New Hampshire
| |
Collapse
|
31
|
Levy JJ, Azizgolshani N, Andersen MJ, Suriawinata A, Liu X, Lisovsky M, Ren B, Bobak CA, Christensen BC, Vaickus LJ. A large-scale internal validation study of unsupervised virtual trichrome staining technologies on nonalcoholic steatohepatitis liver biopsies. Mod Pathol 2021; 34:808-822. [PMID: 33299110 PMCID: PMC7985027 DOI: 10.1038/s41379-020-00718-1] [Citation(s) in RCA: 15] [Impact Index Per Article: 5.0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/03/2020] [Revised: 11/06/2020] [Accepted: 11/08/2020] [Indexed: 02/07/2023]
Abstract
Non-alcoholic steatohepatitis (NASH) is a fatty liver disease characterized by accumulation of fat in hepatocytes with concurrent inflammation and is associated with morbidity, cirrhosis and liver failure. After extraction of a liver core biopsy, tissue sections are stained with hematoxylin and eosin (H&E) to grade NASH activity, and stained with trichrome to stage fibrosis. Methods to computationally transform one stain into another on digital whole slide images (WSI) can lessen the need for additional physical staining besides H&E, reducing personnel, equipment, and time costs. Generative adversarial networks (GAN) have shown promise for virtual staining of tissue. We conducted a large-scale validation study of the viability of GANs for H&E to trichrome conversion on WSI (n = 574). Pathologists were largely unable to distinguish real images from virtual/synthetic images given a set of twelve Turing Tests. We report high correlation between staging of real and virtual stains ([Formula: see text]; 95% CI: 0.84-0.88). Stages assigned to both virtual and real stains correlated similarly with a number of clinical biomarkers and progression to End Stage Liver Disease (Hazard Ratio HR = 2.06, 95% CI: 1.36-3.12, p < 0.001 for real stains; HR = 2.02, 95% CI: 1.40-2.92, p < 0.001 for virtual stains). Our results demonstrate that virtual trichrome technologies may offer a software solution that can be employed in the clinical setting as a diagnostic decision aid.
Collapse
Affiliation(s)
- Joshua J Levy
- Emerging Diagnostic and Investigative Technologies, Clinical Genomics and Advanced Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, NH, 03756, USA.
- Department of Epidemiology, Geisel School of Medicine at Dartmouth, Lebanon, NH, 03756, USA.
- Program in Quantitative Biomedical Sciences, Geisel School of Medicine at Dartmouth, Lebanon, NH, 03756, USA.
| | - Nasim Azizgolshani
- Department of Epidemiology, Geisel School of Medicine at Dartmouth, Lebanon, NH, 03756, USA
| | - Michael J Andersen
- Emerging Diagnostic and Investigative Technologies, Clinical Genomics and Advanced Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, NH, 03756, USA
| | - Arief Suriawinata
- Emerging Diagnostic and Investigative Technologies, Clinical Genomics and Advanced Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, NH, 03756, USA
| | - Xiaoying Liu
- Emerging Diagnostic and Investigative Technologies, Clinical Genomics and Advanced Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, NH, 03756, USA
| | - Mikhail Lisovsky
- Emerging Diagnostic and Investigative Technologies, Clinical Genomics and Advanced Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, NH, 03756, USA
| | - Bing Ren
- Emerging Diagnostic and Investigative Technologies, Clinical Genomics and Advanced Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, NH, 03756, USA
| | - Carly A Bobak
- Program in Quantitative Biomedical Sciences, Geisel School of Medicine at Dartmouth, Lebanon, NH, 03756, USA
- Thayer School of Engineering, Dartmouth College, Hanover, NH, 03755, USA
- Department of Biomedical Data Science, Geisel School of Medicine at Dartmouth, Lebanon, NH, 03756, USA
- The Dartmouth Institute for Health Policy and Clinical Practice, Geisel School of Medicine at Dartmouth, Lebanon, NH, 03756, USA
| | - Brock C Christensen
- Department of Epidemiology, Geisel School of Medicine at Dartmouth, Lebanon, NH, 03756, USA
- Department of Molecular and Systems Biology, Geisel School of Medicine at Dartmouth, Lebanon, NH, 03756, USA
- Department of Community and Family Medicine, Geisel School of Medicine at Dartmouth, Lebanon, NH, 03756, USA
| | - Louis J Vaickus
- Emerging Diagnostic and Investigative Technologies, Clinical Genomics and Advanced Technologies, Department of Pathology and Laboratory Medicine, Dartmouth Hitchcock Medical Center, Lebanon, NH, 03756, USA
| |
Collapse
|
32
|
Brady RE, Badour CL, Arega EA, Levy JJ, Adams TG. Evaluating the mediating effects of perceived vulnerability to disease in the relation between disgust and contamination-based OCD. J Anxiety Disord 2021; 79:102384. [PMID: 33774559 DOI: 10.1016/j.janxdis.2021.102384] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 06/14/2020] [Revised: 02/17/2021] [Accepted: 03/01/2021] [Indexed: 10/21/2022]
Abstract
Contamination-based obsessive-compulsive disorder (OCD) is thought to develop and be maintained by excessive propensity to experience disgust, particularly in response to perceived contaminants, and dysfunctional threat appraisals pertaining to illness. The present studies attempted to integrate these lines of research by testing the degree to which contamination-based OCD is associated with individual differences in disgust propensity and sensitivity, affective distress in response to perceived contaminants, and perceived threat of illness. In Study 1, a convenience sample of 185 adults completed self-report scales assessing obsessive-compulsive symptoms, disgust propensity and sensitivity, germ aversion, and perceived infectability. Multivariate regression showed that disgust propensity and germ aversion were the only significant predictors of contamination-based obsessions and compulsions. Exploratory analyses suggested that there was a significant indirect effect of disgust propensity on contamination-based obsessions and compulsions via germ aversion. Findings from Study 1 were replicated using a sample of twenty-six obsessive-compulsive participants. Despite the substantially smaller sample, the proportion of the total effects attributable to the mediating effect of germ aversion was comparable, consistent with a significant partial mediation in both samples. These results together suggest that contamination-based OCD symptoms are likely maintained and motivated by basic affective processes.
Collapse
Affiliation(s)
- Robert E Brady
- Geisel School of Medicine at Dartmouth, United States; Dartmouth-Hitchcock Medical Center, United States.
| | | | - Enat A Arega
- Geisel School of Medicine at Dartmouth, United States
| | - Joshua J Levy
- Geisel School of Medicine at Dartmouth, United States
| | | |
Collapse
|
33
|
Gordon SP, Contreras-Moreira B, Levy JJ, Djamei A, Czedik-Eysenberg A, Tartaglio VS, Session A, Martin J, Cartwright A, Katz A, Singan VR, Goltsman E, Barry K, Dinh-Thi VH, Chalhoub B, Diaz-Perez A, Sancho R, Lusinska J, Wolny E, Nibau C, Doonan JH, Mur LAJ, Plott C, Jenkins J, Hazen SP, Lee SJ, Shu S, Goodstein D, Rokhsar D, Schmutz J, Hasterok R, Catalan P, Vogel JP. Gradual polyploid genome evolution revealed by pan-genomic analysis of Brachypodium hybridum and its diploid progenitors. Nat Commun 2020; 11:3670. [PMID: 32728126 PMCID: PMC7391716 DOI: 10.1038/s41467-020-17302-5] [Citation(s) in RCA: 47] [Impact Index Per Article: 11.8] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/13/2019] [Accepted: 06/19/2020] [Indexed: 02/08/2023] Open
Abstract
Our understanding of polyploid genome evolution is constrained because we cannot know the exact founders of a particular polyploid. To differentiate between founder effects and post polyploidization evolution, we use a pan-genomic approach to study the allotetraploid Brachypodium hybridum and its diploid progenitors. Comparative analysis suggests that most B. hybridum whole gene presence/absence variation is part of the standing variation in its diploid progenitors. Analysis of nuclear single nucleotide variants, plastomes and k-mers associated with retrotransposons reveals two independent origins for B. hybridum, ~1.4 and ~0.14 million years ago. Examination of gene expression in the younger B. hybridum lineage reveals no bias in overall subgenome expression. Our results are consistent with a gradual accumulation of genomic changes after polyploidization and a lack of subgenome expression dominance. Significantly, if we did not use a pan-genomic approach, we would grossly overestimate the number of genomic changes attributable to post polyploidization evolution.
Collapse
Affiliation(s)
- Sean P Gordon
- DOE Joint Genome Institute, Berkeley, CA, 94720, USA
| | - Bruno Contreras-Moreira
- Estación Experimental de Aula Dei (EEAD-CSIC), Zaragoza, Spain
- Fundación ARAID, Zaragoza, Spain
- Grupo de Bioquímica, Biofísica y Biología Computacional (BIFI, UNIZAR), Unidad Asociada al CSIC, Zaragoza, Spain
| | - Joshua J Levy
- DOE Joint Genome Institute, Berkeley, CA, 94720, USA
- University California, Berkeley, Berkeley, CA, 94720, USA
| | - Armin Djamei
- Gregor Mendel Institute of Molecular Plant Biology GmbH, Vienna, Austria
- Leibniz Institute of Plant Genetics and Crop Plant Research (IPK) Gatersleben. Stadt Seeland, Seeland, Germany
| | | | - Virginia S Tartaglio
- DOE Joint Genome Institute, Berkeley, CA, 94720, USA
- University California, Berkeley, Berkeley, CA, 94720, USA
| | - Adam Session
- DOE Joint Genome Institute, Berkeley, CA, 94720, USA
| | - Joel Martin
- DOE Joint Genome Institute, Berkeley, CA, 94720, USA
| | | | - Andrew Katz
- DOE Joint Genome Institute, Berkeley, CA, 94720, USA
| | | | | | - Kerrie Barry
- DOE Joint Genome Institute, Berkeley, CA, 94720, USA
| | - Vinh Ha Dinh-Thi
- Organization and evolution of complex genomes (OECG) Institut national de la Recherche agronomique (INRA), Université d'Evry Val d'Essonne (UEVE), Evry, France
| | - Boulos Chalhoub
- Organization and evolution of complex genomes (OECG) Institut national de la Recherche agronomique (INRA), Université d'Evry Val d'Essonne (UEVE), Evry, France
- Institute of Crop Science, Zhejiang University, 866 Yu-Hang-Tang Road, 310058, Hangzhou, China
| | - Antonio Diaz-Perez
- Universidad de Zaragoza-Escuela Politécnica Superior de Huesca, 22071, Huesca, Spain
- Instituto de Genética, Facultad de Agronomía, Universidad Central de Venezuela, 2102, Maracay, Venezuela
| | - Ruben Sancho
- Universidad de Zaragoza-Escuela Politécnica Superior de Huesca, 22071, Huesca, Spain
| | - Joanna Lusinska
- Plant Cytogenetics and Molecular Biology Group, Institute of Biology, Biotechnology and Environmental Protection, Faculty of Natural Sciences, University of Silesia in Katowice, 40-032, Katowice, Poland
| | - Elzbieta Wolny
- Plant Cytogenetics and Molecular Biology Group, Institute of Biology, Biotechnology and Environmental Protection, Faculty of Natural Sciences, University of Silesia in Katowice, 40-032, Katowice, Poland
| | - Candida Nibau
- Institute of Biological, Environmental and Rural Sciences (IBERS), Aberystwyth University, Aberystwyth, Wales, UK
| | - John H Doonan
- Institute of Biological, Environmental and Rural Sciences (IBERS), Aberystwyth University, Aberystwyth, Wales, UK
| | - Luis A J Mur
- Institute of Biological, Environmental and Rural Sciences (IBERS), Aberystwyth University, Aberystwyth, Wales, UK
| | - Chris Plott
- HudsonAlpha Institute for Biotechnology, Huntsville, AL, 35806, USA
| | - Jerry Jenkins
- HudsonAlpha Institute for Biotechnology, Huntsville, AL, 35806, USA
| | - Samuel P Hazen
- Biology Department, University of Massachusetts Amherst, Amherst, MA, 01003, USA
| | - Scott J Lee
- Biology Department, University of Massachusetts Amherst, Amherst, MA, 01003, USA
| | | | | | - Daniel Rokhsar
- DOE Joint Genome Institute, Berkeley, CA, 94720, USA
- University California, Berkeley, Berkeley, CA, 94720, USA
| | - Jeremy Schmutz
- DOE Joint Genome Institute, Berkeley, CA, 94720, USA
- HudsonAlpha Institute for Biotechnology, Huntsville, AL, 35806, USA
| | - Robert Hasterok
- Plant Cytogenetics and Molecular Biology Group, Institute of Biology, Biotechnology and Environmental Protection, Faculty of Natural Sciences, University of Silesia in Katowice, 40-032, Katowice, Poland
| | - Pilar Catalan
- Grupo de Bioquímica, Biofísica y Biología Computacional (BIFI, UNIZAR), Unidad Asociada al CSIC, Zaragoza, Spain.
- Universidad de Zaragoza-Escuela Politécnica Superior de Huesca, 22071, Huesca, Spain.
- Institute of Biology, Tomsk State University, Tomsk, 634050, Russia.
| | - John P Vogel
- DOE Joint Genome Institute, Berkeley, CA, 94720, USA.
- University California, Berkeley, Berkeley, CA, 94720, USA.
| |
Collapse
|
34
|
Levy JJ, Titus AJ, Salas LA, Christensen BC. PyMethylProcess-convenient high-throughput preprocessing workflow for DNA methylation data. Bioinformatics 2020; 35:5379-5381. [PMID: 31368477 DOI: 10.1093/bioinformatics/btz594] [Citation(s) in RCA: 11] [Impact Index Per Article: 2.8] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/14/2019] [Revised: 06/27/2019] [Accepted: 07/26/2019] [Indexed: 01/28/2023] Open
Abstract
SUMMARY Performing highly parallelized preprocessing of methylation array data using Python can accelerate data preparation for downstream methylation analyses, including large scale production-ready machine learning pipelines. We present a highly reproducible, scalable pipeline (PyMethylProcess) that can be quickly set-up and deployed through Docker and PIP. AVAILABILITY AND IMPLEMENTATION Project Home Page: https://github.com/Christensen-Lab-Dartmouth/PyMethylProcess. Available on PyPI (pymethylprocess), Docker (joshualevy44/pymethylprocess). SUPPLEMENTARY INFORMATION Supplementary data are available at Bioinformatics online.
Collapse
Affiliation(s)
- Joshua J Levy
- Department of Epidemiology, Geisel School of Medicine at Dartmouth.,Program in Quantitative Biomedical Sciences, Geisel School of Medicine at Dartmouth, Hanover, NH, USA
| | | | - Lucas A Salas
- Department of Epidemiology, Geisel School of Medicine at Dartmouth
| | - Brock C Christensen
- Department of Epidemiology, Geisel School of Medicine at Dartmouth.,Department of Molecular and Systems Biology, Hanover, NH, USA
| |
Collapse
|
35
|
Levy JJ, O'Malley AJ. Don't dismiss logistic regression: the case for sensible extraction of interactions in the era of machine learning. BMC Med Res Methodol 2020; 20:171. [PMID: 32600277 PMCID: PMC7325087 DOI: 10.1186/s12874-020-01046-3] [Citation(s) in RCA: 25] [Impact Index Per Article: 6.3] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/17/2019] [Accepted: 06/10/2020] [Indexed: 01/08/2023] Open
Abstract
BACKGROUND Machine learning approaches have become increasingly popular modeling techniques, relying on data-driven heuristics to arrive at its solutions. Recent comparisons between these algorithms and traditional statistical modeling techniques have largely ignored the superiority gained by the former approaches due to involvement of model-building search algorithms. This has led to alignment of statistical and machine learning approaches with different types of problems and the under-development of procedures that combine their attributes. In this context, we hoped to understand the domains of applicability for each approach and to identify areas where a marriage between the two approaches is warranted. We then sought to develop a hybrid statistical-machine learning procedure with the best attributes of each. METHODS We present three simple examples to illustrate when to use each modeling approach and posit a general framework for combining them into an enhanced logistic regression model building procedure that aids interpretation. We study 556 benchmark machine learning datasets to uncover when machine learning techniques outperformed rudimentary logistic regression models and so are potentially well-equipped to enhance them. We illustrate a software package, InteractionTransformer, which embeds logistic regression with advanced model building capacity by using machine learning algorithms to extract candidate interaction features from a random forest model for inclusion in the model. Finally, we apply our enhanced logistic regression analysis to two real-word biomedical examples, one where predictors vary linearly with the outcome and another with extensive second-order interactions. RESULTS Preliminary statistical analysis demonstrated that across 556 benchmark datasets, the random forest approach significantly outperformed the logistic regression approach. We found a statistically significant increase in predictive performance when using hybrid procedures and greater clarity in the association with the outcome of terms acquired compared to directly interpreting the random forest output. CONCLUSIONS When a random forest model is closer to the true model, hybrid statistical-machine learning procedures can substantially enhance the performance of statistical procedures in an automated manner while preserving easy interpretation of the results. Such hybrid methods may help facilitate widespread adoption of machine learning techniques in the biomedical setting.
Collapse
Affiliation(s)
- Joshua J Levy
- Program in Quantitative Biomedical Sciences, Geisel School of Medicine at Dartmouth, Hanover, USA.
- Department of Epidemiology, Geisel School of Medicine at Dartmouth, Hanover, USA.
- Department of Pathology, Geisel School of Medicine at Dartmouth, Hanover, USA.
| | - A James O'Malley
- Department of Biomedical Data Science, Geisel School of Medicine at Dartmouth, Hanover, USA
- The Dartmouth Institute for Health Policy and Clinical Practice, Geisel School of Medicine at Dartmouth, Hanover, USA
| |
Collapse
|
36
|
Levy JJ, Titus AJ, Petersen CL, Chen Y, Salas LA, Christensen BC. MethylNet: an automated and modular deep learning approach for DNA methylation analysis. BMC Bioinformatics 2020; 21:108. [PMID: 32183722 PMCID: PMC7076991 DOI: 10.1186/s12859-020-3443-8] [Citation(s) in RCA: 41] [Impact Index Per Article: 10.3] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/11/2019] [Accepted: 03/04/2020] [Indexed: 12/13/2022] Open
Abstract
BACKGROUND DNA methylation (DNAm) is an epigenetic regulator of gene expression programs that can be altered by environmental exposures, aging, and in pathogenesis. Traditional analyses that associate DNAm alterations with phenotypes suffer from multiple hypothesis testing and multi-collinearity due to the high-dimensional, continuous, interacting and non-linear nature of the data. Deep learning analyses have shown much promise to study disease heterogeneity. DNAm deep learning approaches have not yet been formalized into user-friendly frameworks for execution, training, and interpreting models. Here, we describe MethylNet, a DNAm deep learning method that can construct embeddings, make predictions, generate new data, and uncover unknown heterogeneity with minimal user supervision. RESULTS The results of our experiments indicate that MethylNet can study cellular differences, grasp higher order information of cancer sub-types, estimate age and capture factors associated with smoking in concordance with known differences. CONCLUSION The ability of MethylNet to capture nonlinear interactions presents an opportunity for further study of unknown disease, cellular heterogeneity and aging processes.
Collapse
Affiliation(s)
- Joshua J Levy
- Program in Quantitative Biomedical Sciences, Geisel School of Medicine at Dartmouth, Hanover, NH, 03755, USA.
- Department of Epidemiology, Geisel School of Medicine at Dartmouth, Hanover, NH, 03755, USA.
| | - Alexander J Titus
- Department of Defense, Office of the Under Secretary of Defense for Research & Engineering, Washington, DC, USA
| | - Curtis L Petersen
- Program in Quantitative Biomedical Sciences, Geisel School of Medicine at Dartmouth, Hanover, NH, 03755, USA
- Department of Epidemiology, Geisel School of Medicine at Dartmouth, Hanover, NH, 03755, USA
- The Dartmouth Institute for Health Policy and Clinical Practice, Lebanon, NH, 03766, USA
| | - Youdinghuan Chen
- Program in Quantitative Biomedical Sciences, Geisel School of Medicine at Dartmouth, Hanover, NH, 03755, USA
- Department of Epidemiology, Geisel School of Medicine at Dartmouth, Hanover, NH, 03755, USA
| | - Lucas A Salas
- Department of Epidemiology, Geisel School of Medicine at Dartmouth, Hanover, NH, 03755, USA
| | - Brock C Christensen
- Department of Epidemiology, Geisel School of Medicine at Dartmouth, Hanover, NH, 03755, USA
- Department of Molecular and Systems Biology, Geisel School of Medicine at Dartmouth, Hanover, NH, 03755, USA
| |
Collapse
|
37
|
Levy JJ, Salas LA, Christensen BC, Sriharan A, Vaickus LJ. PathFlowAI: A High-Throughput Workflow for Preprocessing, Deep Learning and Interpretation in Digital Pathology. Pac Symp Biocomput 2020; 25:403-414. [PMID: 31797614 PMCID: PMC6919317] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Indexed: 11/03/2022]
Abstract
The diagnosis of disease often requires analysis of a biopsy. Many diagnoses depend not only on the presence of certain features but on their location within the tissue. Recently, a number of deep learning diagnostic aids have been developed to classify digitized biopsy slides. Clinical workflows often involve processing of more than 500 slides per day. But, clinical use of deep learning diagnostic aids would require a preprocessing workflow that is cost-effective, flexible, scalable, rapid, interpretable, and transparent. Here, we present such a workflow, optimized using Dask and mixed precision training via APEX, capable of handling any patch-level or slide level classification and prediction problem. The workflow uses a flexible and fast preprocessing and deep learning analytics pipeline, incorporates model interpretation and has a highly storage-efficient audit trail. We demonstrate the utility of this package on the analysis of a prototypical anatomic pathology specimen, liver biopsies for evaluation of hepatitis from a prospective cohort. The preliminary data indicate that PathFlowAI may become a cost-effective and time-efficient tool for clinical use of Artificial Intelligence (AI) algorithms.
Collapse
Affiliation(s)
| | - Lucas A. Salas
- Department of Epidemiology, Geisel School of Medicine at Dartmouth Lebanon, NH 03756
| | - Brock C. Christensen
- Department of Epidemiology, Geisel School of Medicine at Dartmouth Lebanon, NH 03756
| | - Aravindhan Sriharan
- Department of Pathology, Dartmouth Hitchcock Medical Center Lebanon, NH 03756
| | - Louis J. Vaickus
- Department of Pathology, Dartmouth Hitchcock Medical Center Lebanon, NH 03756
| |
Collapse
|
38
|
Gordon SP, Levy JJ, Vogel JP. PolyCRACKER, a robust method for the unsupervised partitioning of polyploid subgenomes by signatures of repetitive DNA evolution. BMC Genomics 2019; 20:580. [PMID: 31299888 PMCID: PMC6626429 DOI: 10.1186/s12864-019-5828-5] [Citation(s) in RCA: 4] [Impact Index Per Article: 0.8] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/22/2019] [Accepted: 05/23/2019] [Indexed: 11/10/2022] Open
Abstract
BACKGROUND Our understanding of polyploid genomes is limited by our inability to definitively assign sequences to a specific subgenome without extensive prior knowledge like high resolution genetic maps or genome sequences of diploid progenitors. In theory, existing methods for assigning sequences to individual species from metagenome samples could be used to separate subgenomes in polyploid organisms, however, these methods rely on differences in coarse genome properties like GC content or sequences from related species. Thus, these approaches do not work for subgenomes where gross features are indistinguishable and related genomes are lacking. Here we describe a method that uses rapidly evolving repetitive DNA to circumvent these limitations. RESULTS By using short, repetitive, DNA sequences as species-specific signals we separated closely related genomes from test datasets and subgenomes from two polyploid plants, tobacco and wheat, without any prior knowledge. CONCLUSION This approach is ideal for separating the subgenomes of polyploid species with unsequenced or unknown progenitor genomes.
Collapse
Affiliation(s)
- Sean P. Gordon
- DOE Joint Genome Institute, 2800 Mitchell Dr, Walnut Creek, CA 94598 USA
| | - Joshua J. Levy
- DOE Joint Genome Institute, 2800 Mitchell Dr, Walnut Creek, CA 94598 USA
- University of California Berkeley, Berkeley, CA USA
| | - John P. Vogel
- DOE Joint Genome Institute, 2800 Mitchell Dr, Walnut Creek, CA 94598 USA
- University of California Berkeley, Berkeley, CA USA
| |
Collapse
|
39
|
Plucker JA, Levy JJ. The downside of being talented. Am Psychol 2001; 56:75-6. [PMID: 11242991] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [MESH Headings] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 02/19/2023]
|
40
|
Levy JJ, Samson JM, Lopez F, Picod-bernard C, Maticka-tyndale E. [Unsafe sex and contraception among students from France, Quebec and Spain]. Contracept Fertil Sex (Paris) 1993; 21:914-9. [PMID: 12318995] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Key Words] [MESH Headings] [Subscribe] [Scholar Register] [Indexed: 02/26/2023]
|
41
|
Chorev M, Caulfield MP, Roubini E, McKee RL, Gibbons SW, Leu CT, Levy JJ, Rosenblatt M. A novel, mild, specific and indirect maleimido-based radioiodolabeling method. Radiolabeling of analogs derived from parathyroid hormone (PTH) and PTH-related protein (PTHrP). Int J Pept Protein Res 1992; 40:445-55. [PMID: 1336486] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [MESH Headings] [Subscribe] [Scholar Register] [Indexed: 12/26/2022]
Abstract
In an effort to design a mild, non-oxidative and site-specific means of radiolabeling bioactive molecules we have employed maleimido-sulfhydryl chemistry to produce bioactive hormone radioligands. We have prepared two novel radioiodolabeled reagents, 3'-maleimidopropanoyl-3-125I-tyramide and its retro analog, N-maleoyl-N'-3-(4-hydroxy-3-125I-phenyl)propanoyl ethylenediamide, by either oxidative radioiodination of the precursors or radiolabeling of the phenolic component prior to its incorporation into the radiolabeling reagents. These reagents were then used to radiolabel analogs of parathyroid hormone (PTH) and parathyroid hormone-related protein (PTHrP) in an efficient way, yielding reaction mixtures which were easily purified. The radioligands obtained are stable upon storage and bind in a reversible manner to a single population of binding sites displaying affinity in the low nanomolar range. The potencies of these analogs are comparable to the non-modified PTH and PTHrP analogs. This study demonstrates the utility of the novel maleimido-based indirect radioiodination approach and highlights some of its advantages over either direct oxidative procedures or acylation using the Bolton-Hunter reagent.
Collapse
Affiliation(s)
- M Chorev
- Department of Pharmaceutical Chemistry, Faculty of Medicine, Hebrew University of Jerusalem, Israel
| | | | | | | | | | | | | | | |
Collapse
|
42
|
Goud NA, McKee RL, Sardana MK, DeHaven PA, Huelar E, Syed MM, Goud RA, Gibbons SW, Fisher JE, Levy JJ. Solid-phase synthesis and biologic activity of human parathyroid hormone (1-84). J Bone Miner Res 1991; 6:781-9. [PMID: 1664643 DOI: 10.1002/jbmr.5650060802] [Citation(s) in RCA: 8] [Impact Index Per Article: 0.2] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 12/28/2022]
Abstract
We have chemically synthesized the full-length, 84 amino acid, human parathyroid hormone (hPTH) on a greater than 100 mg scale by the Merrifield solid-phase technique of stepwise peptide synthesis using a benzhydrylamine support. The peptide was purified by high-performance liquid chromatography and found to be greater than 96% pure. The authenticity or the sequence of the synthetic peptide was confirmed by repetitive Edman degradation. Furthermore, tryptic digestion of hPTH generated the predicted fragments. The synthetic full-length hormone was evaluated for biologic activity in assays of PTH receptor binding and stimulation of adenylate cyclase activity (using bovine renal cortical membranes and rat and human bone cells). Synthetic hPTH (1-84) was found to be highly potent in binding to PTH receptors (Kb = 1-25 nM) and stimulating adenylate cyclase (Km = 1-14 nM). The availability of significant quantities of synthetic full-length hPTH and future analogs will permit widespread use in multiple in vitro and in vivo assays to delineate their spectrum of biologic properties. Available supplies of the synthetic hormone will also enable evaluation of the effectiveness of PTH antagonists at inhibiting the action of native sequence hormone at its receptors.
Collapse
Affiliation(s)
- N A Goud
- Bachem, Inc., Torrance, California
| | | | | | | | | | | | | | | | | | | |
Collapse
|
43
|
Abstract
A light-efficient folded perfect shuffle network is described. Two-dimensional (2-D) raster encoding of the processing element nodes is used to accommodate large arrays with simple imaging optics. The network uses 2-D arrays of lenslets and prisms to correct for magnification and replication losses. Simulation results show the required prism arrays to be of low complexity and suggest that the network is tolerant to imperfections in the prism parameters.
Collapse
|
44
|
Tosteson MT, Alvarez O, Hubbell W, Bieganski RM, Attenbach C, Caporales LH, Levy JJ, Nutt RF, Rosenblatt M, Tosteson DC. Primary structure of peptides and ion channels. Role of amino acid side chains in voltage gating of melittin channels. Biophys J 1990; 58:1367-75. [PMID: 1703448 PMCID: PMC1281090 DOI: 10.1016/s0006-3495(90)82483-8] [Citation(s) in RCA: 47] [Impact Index Per Article: 1.4] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/28/2022] Open
Abstract
Melittin produces a voltage-dependent increase in the conductance of planar lipid bilayers. The conductance increases when the side of the membrane to which melittin has been added (cis-side) is made positive. This paper reports observations on the effect of modifying two positively charged amino acid residues within the NH2-terminal region of the molecule: lysine at position 7 (K7), and the NH2-terminal glycine (G1). We have synthesized melittin analogues in which K7 is replaced by asparagine (K7-N), G1 is blocked by a formyl group (G1-f), and in which both modifications of the parent compound were introduced (G1-f, K7-N). The time required to reach peak conductance during a constant voltage pulse was shorter in membranes exposed to the analogues than in membranes modified by melittin. The apparent number of monomers producing a conducting unit for [K7-N]-melittin and [G1-f]-melittin, eight, was found to be greater than the one for [G1-f], K7-N]-melittin and for melittin itself, four. The apparent gating charge per monomer was less for the analogues, 0.5-0.3 than for melittin, one. Essentially similar results were obtained with melittin analogues in which the charge on K7 or G1 or both was blocked by an uncharged N-linked spin label. These results show that the positive charges in the NH2-terminal region of melittin play a major but not exclusive role in the voltage gating of melittin channels in bilayers.
Collapse
Affiliation(s)
- M T Tosteson
- Department of Cellular Molecular Physiology, Harvard Medical School, Boston, Massachusetts 02115
| | | | | | | | | | | | | | | | | | | |
Collapse
|
45
|
Caulfield MP, McKee RL, Goldman ME, Duong LT, Fisher JE, Gay CT, DeHaven PA, Levy JJ, Roubini E, Nutt RF. The bovine renal parathyroid hormone (PTH) receptor has equal affinity for two different amino acid sequences: the receptor binding domains of PTH and PTH-related protein are located within the 14-34 region. Endocrinology 1990; 127:83-7. [PMID: 2163327 DOI: 10.1210/endo-127-1-83] [Citation(s) in RCA: 76] [Impact Index Per Article: 2.2] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 12/30/2022]
Abstract
Previous studies examining the interaction of PTH and PTH-related protein (PTHrP) with target tissue have for the most part emphasized the similarity between the two hormones in binding to and activating receptors. This observation that two peptides with limited homology have equal affinities for the same receptor is unusual. In this report we investigated two aspects of PTH/PTHrP-receptor interactions. First, the nonhomologous 14-34 regions of PTH and PTHrP were synthesized and evaluated. Second, hybrid peptides containing the 7-18 fragment of one hormone combined with the 19-34 region of the other hormone were studied to determine whether interactions between these two regions are required for receptor recognition. All four peptides were examined in bovine renal cortical membrane and rat osteosarcoma (ROS 17/2.8) cell PTH-binding and PTH-stimulated adenylate cyclase assays. The results indicate that the receptor-binding domains of PTH and PTHrP lie outside of the 1-13 region, the region containing sequence homology shared by the two hormones, and that two peptides of different amino acid sequence bind with equal affinity to the bovine renal PTH receptor. However, in the absence of the N-terminal region, the rat bone PTH receptor displays a preference for the C-terminal (19-34 sequence) region of PTHrP.
Collapse
Affiliation(s)
- M P Caulfield
- Department of Biological Research and Molecular Biology, Merck, Sharp, and Dohme Research Laboratories, West Point, Pennsylvania 19486
| | | | | | | | | | | | | | | | | | | |
Collapse
|
46
|
Nutt RF, Caulfield MP, Levy JJ, Gibbons SW, Rosenblatt M, McKee RL. Removal of partial agonism from parathyroid hormone (PTH)-related protein-(7-34)NH2 by substitution of PTH amino acids at positions 10 and 11. Endocrinology 1990; 127:491-3. [PMID: 2163325 DOI: 10.1210/endo-127-1-491] [Citation(s) in RCA: 51] [Impact Index Per Article: 1.5] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 12/30/2022]
Abstract
PTHrP(7-34)NH2 and [D-Trp12]PTHrP(7-34)NH2 have previously been shown to be shown to be more potent antagonists than the corresponding PTH peptide, [Tyr34]bPTH(7-34)NH2. However, these peptides also display partial agonism for adenylate cyclase activity in ROS 17/2.8 cells. In this study, design of a pure potent antagonist of PTH and PTHrP by removal of agonism from PTHrP(7-34)NH2 with retention of antagonist potency was accomplished. Since [Tyr34]bPTH(7-34)NH2 lacks agonist activity, we introduced two amino acids native to the PTH sequence into their respective positions in PTHrP and the potent D-Trp12 analog. [Asn10Leu11]- and [Asn10,leu11,D-Trp12]-PTHrP(7-34)NH2 were found to be 23- and 26-fold more potent as antagonists in ROS cells than PTHrP(7-34)NH2 and [D-Trp12]PTHrP(7-34)NH2, respectively. In addition, these peptides did not display partial agonism, even in an assay based on highly responsive cells pretreated with dexamethasone and pertussis toxin. In contrast, when the PTHrP sequence Asp10,Lys11 was inserted into [Tyr34]hPTH(7-34)NH2, antagonist potency declined by more than 6-fold and PTH-like agonist activity was installed. These results demonstrate that the activation domain of both PTH and PTHrP can be extended to include the 1-12 region and that the 10-12 region, in addition to the N-terminal hexapeptide, is important not only for receptor binding but also for hormonal signal transduction.
Collapse
Affiliation(s)
- R F Nutt
- Parathyroid Hormone Laboratory, Merck Sharp and Dohme Research Laboratories, West Point, PA 19486
| | | | | | | | | | | |
Collapse
|
47
|
Chorev M, Goldman ME, McKee RL, Roubini E, Levy JJ, Gay CT, Reagan JE, Fisher JE, Caporale LH, Golub EE. Modifications of position 12 in parathyroid hormone and parathyroid hormone related protein: toward the design of highly potent antagonists. Biochemistry 1990; 29:1580-6. [PMID: 2334716 DOI: 10.1021/bi00458a032] [Citation(s) in RCA: 56] [Impact Index Per Article: 1.6] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/31/2022]
Abstract
Truncated N-terminal fragments of parathyroid hormone (PTH), [Tyr34]bovine PTH(7-34)NH2, and parathyroid hormone related protein (PTHrP), PTHrP(7-34)NH2, inhibit [Nle8,18,[125I]iodo-Tyr34]-bPTH(1-34)NH2 binding and PTH-stimulated adenylate cyclase in bone and kidney assays. However, the receptor interactions of these peptides are 2-3 orders of magnitude weaker than those of their agonist counterparts. To produce an antagonist with increased receptor-binding affinity but lacking agonist-like properties, structure-function studies were undertaken. Glycine at position 12 (present in all homologues of PTH and in PTHrP), which is predicted in both hormones to participate in a beta-turn, was examined by substituting conformational reporters, such as D- or L-Ala, Pro, and alpha-aminoisobutyric acid (Aib), in both agonist and antagonist analogues. Except for N-substituted amino acids, which substantially diminished potency, substitutions were well tolerated, indicating that this site can accept a wide latitude of modifications. To augment receptor avidity, hydrophobic residues compatible with helical secondary structure were introduced. Incorporation of the nonnatural amino acids D-Trp, D-alpha-naphthylalanine (D-alpha-Nal), or D-beta-Nal into either [Tyr34]bPTH(7-34)NH2 or [Nle8,18,Tyr34]bPTH(7-34)NH2 resulted in antagonists that were about 10-fold more active than their respective 7-34 parent compound. Similarly, [D-Trp12]PTHrP(7-34)NH2 was 6 times more potent than the unsubstituted peptide but retained partial agonistic properties, although markedly reduced, similar to PTHrP(7-34)NH2. The antagonistic potentiating effect was configurationally specific.(ABSTRACT TRUNCATED AT 250 WORDS)
Collapse
Affiliation(s)
- M Chorev
- Merck Sharp & Dohme Research Laboratories, West Point, Pennsylvania 19486
| | | | | | | | | | | | | | | | | | | |
Collapse
|
48
|
Caulfield MP, McKee RL, Goldman ME, Thiede MA, Thompson DD, Fisher JE, Levy JJ, Seedor JG, Horiuchi N, Clemens TL. Parathyroid hormone-related protein (PTHrP): studies with synthetic peptides indicate that parathyroid hormone and PTHrP interact with the same receptor. Int J Rad Appl Instrum B 1990; 17:633-7. [PMID: 2175735 DOI: 10.1016/0883-2897(90)90076-d] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.1] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 12/30/2022]
Affiliation(s)
- M P Caulfield
- Merck Sharp and Dohme Research Laboratories, West Point, PA 19486
| | | | | | | | | | | | | | | | | | | |
Collapse
|
49
|
Rosenberg J, Pines M, Levy JJ, Nutt RF, Caulfield MP, Russell J, Sherwood LM, Hurwitz S. Renal and adrenal adenosine 3',5'-monophosphate production and corticosteroid secretion in response to synthetic chicken parathyroid hormone-(1-34). Endocrinology 1989; 125:1082-9. [PMID: 2546736 DOI: 10.1210/endo-125-2-1082] [Citation(s) in RCA: 9] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 01/01/2023]
Abstract
The activity of synthetic chicken (c) PTH-(1-34) amide was tested in dispersed chicken and rat kidney and adrenocortical cells. In the adrenal cells the effect of intact cPTH was also evaluated. In chicken kidney cells, the time- and dose-response patterns of cAMP production were similar for cPTH-(1-34) amide and human (h) PTH-(1-34), whereas rat kidney cells were considerably more sensitive to hPTH-(1-34) than to cPTH-(1-34) amide. The agonist effects of both hPTH-(1-34) and cPTH-(1-34) amide in kidney cells were inhibited by the bovine PTH-(3-34) analog. In chicken adrenocortical cells, cPTH-(1-34) amide stimulated cAMP production and steroid secretion. This action of the peptide was inhibited by bovine PTH-(3-34) and hPTH-(1-34), which by themselves showed no agonist effects. The maximal response of steroid secretion to cPTH-(1-34) amide was significantly lower than that to ACTH, but intact cPTH (supplied as a semipurified parathyroid extract) stimulated steriodogenesis to the same extent as ACTH. In rat adrenocortical cells, intact cPTH stimulated both cAMP formation and steriodogenesis, but cPTH-(1-34) amine showed no agonist effect. The action of the intact hormone in the rat adrenal could be inhibited by cPTH-(1-34) amide. The present results demonstrate the interaction of cPTH-(1-34) with kidney and adrenocortical cells of either chicken or rat. The cAMP and steroidogenic responses of the adrenocortical cells to PTH appear to be dependent (completely in the rat and partially in the chicken) on some sequence beyond the 1-34 region.
Collapse
Affiliation(s)
- J Rosenberg
- Institute of Animal Science, Volcani Center, Bet Dagan, Israel
| | | | | | | | | | | | | | | |
Collapse
|
50
|
Levy JJ, Samson JM, Dupras A, Tessier D. [Children of AIDS]. Nurs Que 1989; 9:44-5. [PMID: 2927782] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [MESH Headings] [Subscribe] [Scholar Register] [Indexed: 01/03/2023]
|