Clustering Algorithms: Their Application to Gene Expression Data

MousiPLIER: A Mouse Pathway-Level Information Extractor Model

High-throughput gene expression profiling measures individual gene expression across conditions. However, genes are regulated in complex networks, not as individual entities, limiting the interpretability of gene expression data. Machine learning models that incorporate prior biological knowledge are a powerful tool to extract meaningful biology from gene expression data. Pathway-level information extractor (PLIER) is an unsupervised machine learning method that defines biological pathways by leveraging the vast amount of published transcriptomic data. PLIER converts gene expression data into known pathway gene sets, termed latent variables (LVs), to substantially reduce data dimensionality and improve interpretability. In the current study, we trained the first mouse PLIER model on 190,111 mouse brain RNA-sequencing samples, the greatest amount of training data ever used by PLIER. We then validated the mousiPLIER approach in a study of microglia and astrocyte gene expression across mouse brain aging. mousiPLIER identified biological pathways that are significantly associated with aging, including one latent variable (LV41) corresponding to striatal signal. To gain further insight into the genes contained in LV41, we performed k-means clustering on the training data to identify studies that respond strongly to LV41. We found that the variable was relevant to striatum and aging across the scientific literature. Finally, we built a Web server (http://mousiplier.greenelab.com/) for users to easily explore the learned latent variables. Taken together, this study defines mousiPLIER as a method to uncover meaningful biological processes in mouse brain transcriptomic studies.

Study of selenium enrichment metabolomics in Bacillus subtilis BSN313 via transcriptome analysis

In this study, the transcriptome analysis was practiced to identify potential genes of probiotic Bacillus subtilis BSN313 involved in selenium (Se) enrichment metabolism. The transcriptomic variation of the strain was deliberated in presence of three different sodium selenite concentrations (0, 3, and 20 μg/mL). The samples were taken at 1 and 13 h subsequent to inoculation of selenite and gene expression profiles in Se metabolism were analyzed through RNA sequencing. The gene expression levels of the pre log phase were lower than the stationary phase. It is because, the bacteria has maximum grown with high concentration of Se (enriched with organic Se), at stationary phase. Bacterial culture containing 3 μg/mL concentration of inorganic Se (sodium selenite) has shown highest gene expression as compared to no or high concentration of Se. This concentration (3 μg/mL) of sodium selenite (as Se) in the medium promoted the upregulation of thioredoxin reductase expression, whereas its higher Se concentration inhibited the formation of selenomethionine (SeMet). The result of 5 L bioreactor fermentation showed that SeMet was also detected in the fermentation supernatant as the growth entered in the late stationary phase and reached up to 857.3 ng/mL. The overall intracellular SeMet enriched content in BSN313 was extended up to 23.4 μg/g dry cell weight. The other two selenoamino acids (Se-AAs), methyl-selenocysteine, and selenocysteine were hardly detected in medium supernatant. From this study, it was concluded that SeMet was the highest content of organic Se byproduct biosynthesized by B. subtilis BSN313 strain in Se-enriched medium during stationary phase. Thus, B. subtilis BSN313 can be considered a commercial probiotic strain that can be used in the food and pharmaceutical industries. This is because it can meet the commercial demand for Se-AAs (SeMet) in both industries.

Impact of tRNA-induced proline-to-serine mistranslation on the transcriptome of Drosophila melanogaster

Mistranslation is the misincorporation of an amino acid into a polypeptide. Mistranslation has diverse effects on multicellular eukaryotes and is implicated in several human diseases. In Drosophila melanogaster, a serine transfer RNA (tRNA) that misincorporates serine at proline codons (P→S) affects male and female flies differently. The mechanisms behind this discrepancy are currently unknown. Here, we compare the transcriptional response of male and female flies to P→S mistranslation to identify genes and cellular processes that underlie sex-specific differences. Both males and females downregulate genes associated with various metabolic processes in response to P→S mistranslation. Males downregulate genes associated with extracellular matrix organization and response to negative stimuli such as wounding, whereas females downregulate aerobic respiration and ATP synthesis genes. Both sexes upregulate genes associated with gametogenesis, but females also upregulate cell cycle and DNA repair genes. These observed differences in the transcriptional response of male and female flies to P→S mistranslation have important implications for the sex-specific impact of mistranslation on disease and tRNA therapeutics.

The impact of FASTQ and alignment read order on structural variant calling from long-read sequencing data

Background

Structural variant (SV) calling from DNA sequencing data has been challenging due to several factors, including the ambiguity of short-read alignments, multiple complex SVs in the same genomic region, and the lack of "truth" datasets for benchmarking. Additionally, caller choice, parameter settings, and alignment method are known to affect SV calling. However, the impact of FASTQ read order on SV calling has not been explored for long-read data.

Results

Here, we used PacBio DNA sequencing data from 15 Caenorhabditis elegans strains and four Arabidopsis thaliana ecotypes to evaluate the sensitivity of different SV callers on FASTQ read order. Comparisons of variant call format files generated from the original and permutated FASTQ files demonstrated that the order of input data affected the SVs predicted by each caller. In particular, pbsv was highly sensitive to the order of the input data, especially at the highest depths where over 70% of the SV calls generated from pairs of differently ordered FASTQ files were in disagreement. These demonstrate that read order sensitivity is a complex, multifactorial process, as the differences observed both within and between species varied considerably according to the specific combination of aligner, SV caller, and sequencing depth. In addition to the SV callers being sensitive to the input data order, the SAMtools alignment sorting algorithm was identified as a source of variability following read order randomization.

Conclusion

The results of this study highlight the sensitivity of SV calling on the order of reads encoded in FASTQ files, which has not been recognized in long-read approaches. These findings have implications for the replication of SV studies and the development of consistent SV calling protocols. Our study suggests that researchers should pay attention to the input order sensitivity of read alignment sorting methods when analyzing long-read sequencing data for SV calling, as mitigating a source of variability could facilitate future replication work. These results also raise important questions surrounding the relationship between SV caller read order sensitivity and tool performance. Therefore, tool developers should also consider input order sensitivity as a potential source of variability during the development and benchmarking of new and improved methods for SV calling.

Unsupervised analysis of whole transcriptome data from human pluripotent stem cells cardiac differentiation

The main objective of the present work was to highlight differences and similarities in gene expression patterns between different pluripotent stem cell cardiac differentiation protocols, using a workflow based on unsupervised machine learning algorithms to analyse the transcriptome of cells cultured as a 2D monolayer or as 3D aggregates. This unsupervised approach effectively allowed to portray the transcriptomic changes that occurred throughout the differentiation processes, with a visual representation of the entire transcriptome. The results allowed to corroborate previously reported data and also to unveil new gene expression patterns. In particular, it was possible to identify a correlation between low cardiomyocyte differentiation efficiencies and the early expression of a set of non-mesodermal genes, which can be further explored as predictive markers of differentiation efficiency. The workflow here developed can also be applied to analyse other stem cell differentiation transcriptomic datasets, envisaging future clinical implementation of cellular therapies.

A Preliminary Investigation into Search and Matching for Tumor Discrimination in World Health Organization Breast Taxonomy Using Deep Networks

Breast cancer is one of the most common cancers affecting women worldwide. It includes a group of malignant neoplasms with a variety of biological, clinical, and histopathologic characteristics. There are more than 35 different histologic forms of breast lesions that can be classified and diagnosed histologically according to cell morphology, growth, and architecture patterns. Recently, deep learning, in the field of artificial intelligence, has drawn a lot of attention for the computerized representation of medical images. Searchable digital atlases can provide pathologists with patch-matching tools, allowing them to search among evidently diagnosed and treated archival cases, a technology that may be regarded as computational second opinion. In this study, we indexed and analyzed the World Health Organization breast taxonomy (Classification of Tumors fifth ed.) spanning 35 tumor types. We visualized all tumor types using deep features extracted from a state-of-the-art deep-learning model, pretrained on millions of diagnostic histopathology images from the Cancer Genome Atlas repository. Furthermore, we tested the concept of a digital "atlas" as a reference for search and matching with rare test cases. The patch similarity search within the World Health Organization breast taxonomy data reached >88% accuracy when validating through "majority vote" and >91% accuracy when validating using top n tumor types. These results show for the first time that complex relationships among common and rare breast lesions can be investigated using an indexed digital archive.

Multimodality imaging reveals angiogenic evolution in vivo during calvarial bone defect healing

The healing of calvarial bone defects is a pressing clinical problem that involves the dynamic interplay between angiogenesis and osteogenesis within the osteogenic niche. Although structural and functional vascular remodeling (i.e., angiogenic evolution) in the osteogenic niche is a crucial modulator of oxygenation, inflammatory and bone precursor cells, most clinical and pre-clinical investigations have been limited to characterizing structural changes in the vasculature and bone. Therefore, we developed a new multimodality imaging approach that for the first time enabled the longitudinal (i.e., over four weeks) and dynamic characterization of multiple in vivo functional parameters in the remodeled vasculature and its effects on de novo osteogenesis, in a preclinical calvarial defect model. We employed multi-wavelength intrinsic optical signal (IOS) imaging to assess microvascular remodeling, intravascular oxygenation (SO2), and osteogenesis; laser speckle contrast (LSC) imaging to assess concomitant changes in blood flow and vascular maturity; and micro-computed tomography (μCT) to validate volumetric changes in calvarial bone. We found that angiogenic evolution was tightly coupled with calvarial bone regeneration and corresponded to distinct phases of bone healing, such as injury, hematoma formation, revascularization, and remodeling. The first three phases occurred during the initial two weeks of bone healing and were characterized by significant in vivo changes in vascular morphology, blood flow, oxygenation, and maturity. Overall, angiogenic evolution preceded osteogenesis, which only plateaued toward the end of bone healing (i.e., four weeks). Collectively, these data indicate the crucial role of angiogenic evolution in osteogenesis. We believe that such multimodality imaging approaches have the potential to inform the design of more efficacious tissue-engineering calvarial defect treatments.

MetageneCluster: a Python package for filtering conflicting signal trends in metagene plots

BACKGROUND

Metagene plots provide a visualization of biological signal trends over subsections of the genome and are used to perform high-level analysis of experimental data by aggregating genome-level data to create an average profile. The generation of metagene plots is useful for summarizing the results of many sequencing-based applications. Despite their prevalence and utility, the standard metagene plot is blind to conflicting signals within data. If multiple distinct trends occur, they can interact destructively, creating a plot that does not accurately represent any of the underlying trends.

RESULTS

We present MetageneCluster, a Python tool to generate a collection of representative metagene plots based on k-means clustering of genomic regions of interest. Clustering the data by similarity allows us to identify patterns within the features of interest. We are then able to summarize each pattern present in the data, rather than averaging across the entire feature space. We show that our method performs well when used to identify conflicting signals in real-world genome-level data.

CONCLUSIONS

Overall, MetageneCluster is a user-friendly tool for the creation of metagene plots that capture distinct patterns in underlying sequence data.

Cellular clarity: a logistic regression approach to identify root epidermal regulators of iron deficiency response

BACKGROUND

Plants respond to stress through highly tuned regulatory networks. While prior works identified master regulators of iron deficiency responses in A. thaliana from whole-root data, identifying regulators that act at the cellular level is critical to a more comprehensive understanding of iron homeostasis. Within the root epidermis complex molecular mechanisms that facilitate iron reduction and uptake from the rhizosphere are known to be regulated by bHLH transcriptional regulators. However, many questions remain about the regulatory mechanisms that control these responses, and how they may integrate with developmental processes within the epidermis. Here, we use transcriptional profiling to gain insight into root epidermis-specific regulatory processes.

RESULTS

Set comparisons of differentially expressed genes (DEGs) between whole root and epidermis transcript measurements identified differences in magnitude and timing of organ-level vs. epidermis-specific responses. Utilizing a unique sampling method combined with a mutual information metric across time-lagged and non-time-lagged windows, we identified relationships between clusters of functionally relevant differentially expressed genes suggesting that developmental regulatory processes may act upstream of well-known Fe-specific responses. By integrating static data (DNA motif information) with time-series transcriptomic data and employing machine learning approaches, specifically logistic regression models with LASSO, we also identified putative motifs that served as crucial features for predicting differentially expressed genes. Twenty-eight transcription factors (TFs) known to bind to these motifs were not differentially expressed, indicating that these TFs may be regulated post-transcriptionally or post-translationally. Notably, many of these TFs also play a role in root development and general stress response.

CONCLUSIONS

This work uncovered key differences in -Fe response identified using whole root data vs. cell-specific root epidermal data. Machine learning approaches combined with additional static data identified putative regulators of -Fe response that would not have been identified solely through transcriptomic profiles and reveal how developmental and general stress responses within the epidermis may act upstream of more specialized -Fe responses for Fe uptake.

Identification of genes associated with abiotic stress tolerance in sweetpotato using weighted gene co-expression network analysis

Sweetpotato, Ipomoea batatas (L.), a key food security crop, is negatively impacted by heat, drought, and salinity stress. The orange-fleshed sweetpotato cultivar "Beauregard" was exposed to heat, salt, and drought treatments for 24 and 48 h to identify genes responding to each stress condition in leaves. Analysis revealed both common (35 up regulated, 259 down regulated genes in the three stress conditions) and unique sets of up regulated (1337 genes by drought, 516 genes by heat, and 97 genes by salt stress) and down regulated (2445 genes by drought, 678 genes by heat, and 204 genes by salt stress) differentially expressed genes (DEGs) suggesting common, yet stress-specific transcriptional responses to these three abiotic stressors. Gene Ontology analysis of down regulated DEGs common to both heat and salt stress revealed enrichment of terms associated with "cell population proliferation" suggestive of an impact on the cell cycle by the two stress conditions. To identify shared and unique gene co-expression networks under multiple abiotic stress conditions, weighted gene co-expression network analysis was performed using gene expression profiles from heat, salt, and drought stress treated 'Beauregard' leaves yielding 18 co-expression modules. One module was enriched for "response to water deprivation," "response to abscisic acid," and "nitrate transport" indicating synergetic crosstalk between nitrogen, water, and phytohormones with genes encoding osmotin, cell expansion, and cell wall modification proteins present as key hub genes in this drought-associated module. This research lays the groundwork for exploring to a further degree, mechanisms for abiotic stress tolerance in sweetpotato.

Inferring circadian gene regulatory relationships from gene expression data with a hybrid framework

BACKGROUND

The central biological clock governs numerous facets of mammalian physiology, including sleep, metabolism, and immune system regulation. Understanding gene regulatory relationships is crucial for unravelling the mechanisms that underlie various cellular biological processes. While it is possible to infer circadian gene regulatory relationships from time-series gene expression data, relying solely on correlation-based inference may not provide sufficient information about causation. Moreover, gene expression data often have high dimensions but a limited number of observations, posing challenges in their analysis.

METHODS

In this paper, we introduce a new hybrid framework, referred to as Circadian Gene Regulatory Framework (CGRF), to infer circadian gene regulatory relationships from gene expression data of rats. The framework addresses the challenges of high-dimensional data by combining the fuzzy C-means clustering algorithm with dynamic time warping distance. Through this approach, we efficiently identify the clusters of genes related to the target gene. To determine the significance of genes within a specific cluster, we employ the Wilcoxon signed-rank test. Subsequently, we use a dynamic vector autoregressive method to analyze the selected significant gene expression profiles and reveal directed causal regulatory relationships based on partial correlation.

CONCLUSION

The proposed CGRF framework offers a comprehensive and efficient solution for understanding circadian gene regulation. Circadian gene regulatory relationships are inferred from the gene expression data of rats based on the Aanat target gene. The results show that genes Pde10a, Atp7b, Prok2, Per1, Rhobtb3 and Dclk1 stand out, which have been known to be essential for the regulation of circadian activity. The potential relationships between genes Tspan15, Eprs, Eml5 and Fsbp with a circadian rhythm need further experimental research.

MousiPLIER: A Mouse Pathway-Level Information Extractor Model

High throughput gene expression profiling is a powerful approach to generate hypotheses on the underlying causes of biological function and disease. Yet this approach is limited by its ability to infer underlying biological pathways and burden of testing tens of thousands of individual genes. Machine learning models that incorporate prior biological knowledge are necessary to extract meaningful pathways and generate rational hypothesis from the vast amount of gene expression data generated to date. We adopted an unsupervised machine learning method, Pathway-level information extractor (PLIER), to train the first mouse PLIER model on 190,111 mouse brain RNA-sequencing samples, the greatest amount of training data ever used by PLIER. mousiPLER converted gene expression data into a latent variables that align to known pathway or cell maker gene sets, substantially reducing data dimensionality and improving interpretability. To determine the utility of mousiPLIER, we applied it to a mouse brain aging study of microglia and astrocyte transcriptomic profiling. We found a specific set of latent variables that are significantly associated with aging, including one latent variable (LV41) corresponding to striatal signal. We next performed k-means clustering on the training data to identify studies that respond strongly to LV41, finding that the variable is relevant to striatum and aging across the scientific literature. Finally, we built a web server (http://mousiplier.greenelab.com/) for users to easily explore the learned latent variables. Taken together this study provides proof of concept that mousiPLIER can uncover meaningful biological processes in mouse transcriptomic studies.

Secretory MPP3 reinforce myeloid differentiation trajectory and amplify myeloid cell production

Hematopoietic stem cells (HSC) and downstream lineage-biased multipotent progenitors (MPP) tailor blood production and control myelopoiesis on demand. Recent lineage tracing analyses revealed MPPs to be major functional contributors to steady-state hematopoiesis. However, we still lack a precise resolution of myeloid differentiation trajectories and cellular heterogeneity in the MPP compartment. Here, we found that myeloid-biased MPP3 are functionally and molecularly heterogeneous, with a distinct subset of myeloid-primed secretory cells with high endoplasmic reticulum (ER) volume and FcγR expression. We show that FcγR+/ERhigh MPP3 are a transitional population serving as a reservoir for rapid production of granulocyte/macrophage progenitors (GMP), which directly amplify myelopoiesis through inflammation-triggered secretion of cytokines in the local bone marrow (BM) microenvironment. Our results identify a novel regulatory function for a secretory MPP3 subset that controls myeloid differentiation through lineage-priming and cytokine production and acts as a self-reinforcing amplification compartment in inflammatory stress and disease conditions.

Artificial Intelligence Can Define and Predict the "Optimal Observed Outcome" After Anterior Shoulder Instability Surgery: An Analysis of 200 Patients With 11-Year Mean Follow-Up

Purpose

The purpose of this study was to use unsupervised machine learning clustering to define the "optimal observed outcome" after surgery for anterior shoulder instability (ASI) and to identify predictors for achieving it.

Methods

Medical records, images, and operative reports were reviewed for patients <40 years old undergoing surgery for ASI. Four unsupervised machine learning clustering algorithms partitioned subjects into "optimal observed outcome" or "suboptimal outcome" based on combinations of actually observed outcomes. Demographic, clinical, and treatment variables were compared between groups using descriptive statistics and Kaplan-Meier survival curves. Variables were assessed for prognostic value through multivariate stepwise logistic regression.

Results

Two hundred patients with a mean follow-up of 11 years were included. Of these, 146 (64%) obtained the "optimal observed outcome," characterized by decreased: postoperative pain (23% vs 52%; P < 0.001), recurrent instability (12% vs 41%; P < 0.001), revision surgery (10% vs 24%; P = 0.015), osteoarthritis (OA) (5% vs 19%; P = 0.005), and restricted motion (161° vs 168°; P = 0.001). Forty-one percent of patients had a "perfect outcome," defined as ideal performance across all outcomes. Time from initial instability to presentation (odds ratio [OR] = 0.96; 95% confidence interval [CI], 0.92-0.98; P = 0.006) and habitual/voluntary instability (OR = 0.17; 95% CI, 0.04-0.77; P = 0.020) were negative predictors of achieving the "optimal observed outcome." A predilection toward subluxations rather than dislocations before surgery (OR = 1.30; 95% CI, 1.02-1.65; P = 0.030) was a positive predictor. Type of surgery performed was not a significant predictor.

Conclusion

After surgery for ASI, 64% of patients achieved the "optimal observed outcome" defined as minimal postoperative pain, no recurrent instability or OA, low revision surgery rates, and increased range of motion, of whom only 41% achieved a "perfect outcome." Positive predictors were shorter time to presentation and predilection toward preoperative subluxations over dislocations.

Level of Evidence

Retrospective cohort, level IV.

MOBILE pipeline enables identification of context-specific networks and regulatory mechanisms

Robust identification of context-specific network features that control cellular phenotypes remains a challenge. We here introduce MOBILE (Multi-Omics Binary Integration via Lasso Ensembles) to nominate molecular features associated with cellular phenotypes and pathways. First, we use MOBILE to nominate mechanisms of interferon-γ (IFNγ) regulated PD-L1 expression. Our analyses suggest that IFNγ-controlled PD-L1 expression involves BST2, CLIC2, FAM83D, ACSL5, and HIST2H2AA3 genes, which were supported by prior literature. We also compare networks activated by related family members transforming growth factor-beta 1 (TGFβ1) and bone morphogenetic protein 2 (BMP2) and find that differences in ligand-induced changes in cell size and clustering properties are related to differences in laminin/collagen pathway activity. Finally, we demonstrate the broad applicability and adaptability of MOBILE by analyzing publicly available molecular datasets to investigate breast cancer subtype specific networks. Given the ever-growing availability of multi-omics datasets, we envision that MOBILE will be broadly useful for identification of context-specific molecular features and pathways.

Bayesian cluster analysis

Bayesian cluster analysis offers substantial benefits over algorithmic approaches by providing not only point estimates but also uncertainty in the clustering structure and patterns within each cluster. An overview of Bayesian cluster analysis is provided, including both model-based and loss-based approaches, along with a discussion on the importance of the kernel or loss selected and prior specification. Advantages are demonstrated in an application to cluster cells and discover latent cell types in single-cell RNA sequencing data to study embryonic cellular development. Lastly, we focus on the ongoing debate between finite and infinite mixtures in a model-based approach and robustness to model misspecification. While much of the debate and asymptotic theory focuses on the marginal posterior of the number of clusters, we empirically show that quite a different behaviour is obtained when estimating the full clustering structure. This article is part of the theme issue 'Bayesian inference: challenges, perspectives, and prospects'.

Topological biclustering ARTMAP for identifying within bicluster relationships

Biclustering is a powerful tool for exploratory data analysis in domains such as social networking, data reduction, and differential gene expression studies. Topological learning identifies connected regions that are difficult to find using other traditional clustering methods and produces a graphical representation. Therefore, to improve the quality of biclustering and module extraction, this work combines the adaptive resonance theory (ART)-based methods of biclustering ARTMAP (BARTMAP) and topological ART (TopoART), to produce TopoBARTMAP. The latter inherits the ability to detect topological associations while performing data reduction. The capabilities of TopoBARTMAP were benchmarked using 35 real world cancer datasets and contrasted with other (bi)clustering methods, where it showed a statistically significant improvement over the other assessed methods on ordered and shuffled data experiments. In experiments with 12 synthetic datasets, the method was observed to perform better at identifying constant, scale, shift, and shift scale type biclusters. The produced graphical representation was refined to represent gene bicluster associations and was assessed on the NCBI GSE89116 dataset containing expression levels of 39,326 probes sampled over 38 observations.

Characterization of Continuous Transcriptional Heterogeneity in High-Risk Blastemal-Type Wilms' Tumors Using Unsupervised Machine Learning

Wilms' tumors are pediatric malignancies that are thought to arise from faulty kidney development. They contain a wide range of poorly differentiated cell states resembling various distorted developmental stages of the fetal kidney, and as a result, differ between patients in a continuous manner that is not well understood. Here, we used three computational approaches to characterize this continuous heterogeneity in high-risk blastemal-type Wilms' tumors. Using Pareto task inference, we show that the tumors form a triangle-shaped continuum in latent space that is bounded by three tumor archetypes with "stromal", "blastemal", and "epithelial" characteristics, which resemble the un-induced mesenchyme, the cap mesenchyme, and early epithelial structures of the fetal kidney. By fitting a generative probabilistic "grade of membership" model, we show that each tumor can be represented as a unique mixture of three hidden "topics" with blastemal, stromal, and epithelial characteristics. Likewise, cellular deconvolution allows us to represent each tumor in the continuum as a unique combination of fetal kidney-like cell states. These results highlight the relationship between Wilms' tumors and kidney development, and we anticipate that they will pave the way for more quantitative strategies for tumor stratification and classification.

Examining unsupervised ensemble learning using spectroscopy data of organic compounds

One solution to the challenge of choosing an appropriate clustering algorithm is to combine different clusterings into a single consensus clustering result, known as cluster ensemble (CE). This ensemble learning strategy can provide more robust and stable solutions across different domains and datasets. Unfortunately, not all clusterings in the ensemble contribute to the final data partition. Cluster ensemble selection (CES) aims at selecting a subset from a large library of clustering solutions to form a smaller cluster ensemble that performs as well as or better than the set of all available clustering solutions. In this paper, we investigate four CES methods for the categorization of structurally distinct organic compounds using high-dimensional IR and Raman spectroscopy data. Single quality selection (SQI) forms a subset of the ensemble by selecting the highest quality ensemble members. The Single Quality Selection (SQI) method is used with various quality indices to select subsets by including the highest quality ensemble members. The Bagging method, usually applied in supervised learning, ranks ensemble members by calculating the normalized mutual information (NMI) between ensemble members and consensus solutions generated from a randomly sampled subset of the full ensemble. The hierarchical cluster and select method (HCAS-SQI) uses the diversity matrix of ensemble members to select a diverse set of ensemble members with the highest quality. Furthermore, a combining strategy can be used to combine subsets selected using multiple quality indices (HCAS-MQI) for the refinement of clustering solutions in the ensemble. The IR + Raman hybrid ensemble library is created by merging two complementary "views" of the organic compounds. This inherently more diverse library gives the best full ensemble consensus results. Overall, the Bagging method is recommended because it provides the most robust results that are better than or comparable to the full ensemble consensus solutions.

Single nucleotide polymorphisms of enamel formation genes and early childhood caries - systematic review, gene-based, gene cluster and meta-analysis

Introduction

Genetic polymorphisms of genes regulating amelogenesis can alter susceptibility to Early Childhood Caries (ECC). This systematic review aims to analyze associations between single-nucleotide polymorphisms of enamel formation genes and ECC.

Methods

Search was conducted across PUBMED, CINAHL, LILACS, SCOPUS, EMBASE, Web of Science, Genome-Wide Association Studies databases from January 2003 to September 2022. This was supplemented by hand search. Totally 7124 articles were identified and 21 articles that satisfied the inclusion criteria proceeded to data extraction. Quality assessment was done using the Q-Genie tool.

Results

Quantitative synthesis revealed that homozygous genotype AA of rs12640848 was significantly higher in children with ECC with an odds ratio of 2.36. Gene-based analysis revealed significant association between six variants of AMBN, four variants of KLK4, two variants of MMP20, and a single variant of each of MMP9 and MMP13 genes and ECC. The Bonferroni corrected-log10 P value of amelogenesis gene Cluster was 2.25 (0.05/88 = 5.6 × 10^-4). Search Tool for Retrieval of Interacting Genes and Proteins plot constructed to comprehend the protein-protein interaction revealed the presence of four functional clusters. Gene function prediction using Multiple Association Network Integration Algorithm revealed that physical interaction between these genes was 69.3%.

Conclusion

Polymorphisms of genes regulating amelogenesis can influence the susceptibility to ECC. AA genotype of rs12640848 may increase the susceptibility to ECC. Gene-based analysis revealed a significant association between multiple polymorphisms of genes regulating amelogenesis and ECC.

Application of Machine Learning in Spatial Proteomics

Spatial proteomics is an interdisciplinary field that investigates the localization and dynamics of proteins, and it has gained extensive attention in recent years, especially the subcellular proteomics. Numerous evidence indicate that the subcellular localization of proteins is associated with various cellular processes and disease progression. Mass spectrometry (MS)-based and imaging-based experimental approaches have been developed to acquire large-scale spatial proteomic data. To allow the reliable analysis of increasingly complex spatial proteomics data, machine learning (ML) methods have been widely used in both MS-based and imaging-based spatial proteomic data analysis pipelines. Here, we comprehensively survey the applications of ML in spatial proteomics from following aspects: (1) data resources for spatial proteome are comprehensively introduced; (2) the roles of different ML algorithms in data analysis pipelines are elaborated; (3) successful applications of spatial proteomics and several analytical tools integrating ML methods are presented; (4) challenges existing in modern ML-based spatial proteomics studies are discussed. This review provides guidelines for researchers seeking to apply ML methods to analyze spatial proteomic data and can facilitate insightful understanding of cell biology as well as the future research in medical and drug discovery communities.

Assisted clustering of gene expression data using regulatory data from partially overlapping sets of individuals

BACKGROUND

As omics measurements profiled on different molecular layers are interconnected, integrative approaches that incorporate the regulatory effect from multi-level omics data are needed. When the multi-level omics data are from the same individuals, gene expression (GE) clusters can be identified using information from regulators like genetic variants and DNA methylation. When the multi-level omics data are from different individuals, the choice of integration approaches is limited.

METHODS

We developed an approach to improve GE clustering from microarray data by integrating regulatory data from different but partially overlapping sets of individuals. We achieve this through (1) decomposing gene expression into the regulated component and the other component that is not regulated by measured factors, (2) optimizing the clustering goodness-of-fit objective function. We do not require the availability of different omics measurements on all individuals. A certain amount of individual overlap between GE data and the regulatory data is adequate for modeling the regulation, thus improving GE clustering.

RESULTS

A simulation study shows that the performance of the proposed approach depends on the strength of the GE-regulator relationship, degree of missingness, data dimensionality, sample size, and the number of clusters. Across the various simulation settings, the proposed method shows competitive performance in terms of accuracy compared to the alternative K-means clustering method, especially when the clustering structure is due mostly to the regulated component, rather than the unregulated component. We further validate the approach with an application to 8,902 Framingham Heart Study participants with data on up to 17,873 genes and regulation information of DNA methylation and genotype from different but partially overlapping sets of participants. We identify clustering structures of genes associated with pulmonary function while incorporating the predicted regulation effect from the measured regulators. We further investigate the over-representation of these GE clusters in pathways of other diseases that may be related to lung function and respiratory health.

CONCLUSION

We propose a novel approach for clustering GE with the assistance of regulatory data that allowed for different but partially overlapping sets of individuals to be included in different omics data.

COMBING: Clustering in Oncology for Mathematical and Biological Identification of Novel Gene Signatures

Precision medicine is a paradigm shift in healthcare relying heavily on genomics data. However, the complexity of biological interactions, the large number of genes as well as the lack of comparisons on the analysis of data, remain a tremendous bottleneck regarding clinical adoption. In this paper, we introduce a novel, automatic and unsupervised framework to discover low-dimensional gene biomarkers. Our method is based on the LP-Stability algorithm, a high dimensional center-based unsupervised clustering algorithm. It offers modularity as concerns metric functions and scalability, while being able to automatically determine the best number of clusters. Our evaluation includes both mathematical and biological criteria to define a quantitative metric. The recovered signature is applied to a variety of biological tasks, including screening of biological pathways and functions, and characterization relevance on tumor types and subtypes. Quantitative comparisons among different distance metrics, commonly used clustering methods and a referential gene signature used in the literature, confirm state of the art performance of our approach. In particular, our signature, based on 27 genes, reports at least 30 times better mathematical significance (average Dunn's Index) and 25% better biological significance (average Enrichment in Protein-Protein Interaction) than those produced by other referential clustering methods. Finally, our signature reports promising results on distinguishing immune inflammatory and immune desert tumors, while reporting a high balanced accuracy of 92% on tumor types classification and averaged balanced accuracy of 68% on tumor subtypes classification, which represents, respectively 7% and 9% higher performance compared to the referential signature.

A critical review of datasets and computational suites for improving cancer theranostics and biomarker discovery

Cancer has been constantly evolving and so is the research pertaining to cancer diagnosis and therapeutic regimens. Early detection and specific therapeutics are the key features of modern cancer therapy. These requirements can only be fulfilled with the integration of diverse high-throughput technologies. Integration of advanced omics methodology involving genomics, epigenomics, proteomics, and transcriptomics provide a clear understanding of multi-faceted cancer. In the past few years, tremendous high-throughput data have been generated from cancer genomics and epigenomic analyses, which on further methodological analyses can yield better biological insights. The major epigenetic alterations reported in cancer are DNA methylation levels, histone post-translational modifications, and epi-miRNA regulating the oncogenes and tumor suppressor genes. While the genomic analyses like gene expression profiling, cancer gene prediction, and genome annotation divulge the genetic alterations in oncogenes or tumor suppressor genes. Also, systems biology approach using biological networks is being extensively used to identify novel cancer biomarkers. Therefore, integration of these multi-dimensional approaches will help to identify potential diagnostic and therapeutic biomarkers. Here, we reviewed the critical databases and tools dedicated to various epigenomic and genomic alterations in cancer. The review further focuses on the multi-omics resources available for further validating the identified cancer biomarkers. We also highlighted the tools for cancer biomarker discovery using a systems biology approach utilizing genomic and epigenomic data. Biomarkers predicted using such integrative approaches are shown to be more clinically relevant.

Performance Analysis of Ovarian Cancer Detection and Classification for Microarray Gene Data

The most common gynecologic cancer, behind cervical and uterine, is ovarian cancer. Ovarian cancer is a severe concern for women. Abnormal cells form and spread throughout the body. Ovarian cancer microarray data can diagnose and prognosis. Typically, ovarian cancer microarray data contains tens of thousands of genes. In order to reduce computational complexity, selecting the most critical genes or attributes in the entire dataset is necessary. Because microarray datasets have limited samples and many characteristics, classifier detection lags. So, dimensionality reduction measures are essential to protect disease classification genes. In this research, initially the ANOVA method is used for gene selection and then two clustering-based and three transform-based feature extraction methods, namely, Fuzzy C Means, Softmax Discriminant Algorithm (SDA), Hilbert Transform, Fast Fourier Transform (FFT), and Discrete Cosine Transform (DCT), respectively, are used to select relevant genes further. Six classifiers further classify the features as normal and abnormal. The NLR classifier gives the highest accuracy for SDA features at 92%, and KNN gives the lowest accuracy of 55% for SDA, Hilbert, and DCT features. With correlation distance feature selection, the NLR classifier attains the lowest accuracy of 53%, and the highest accuracy of 88% is obtained by the GMM classifier.

Mapping the Complex Transcriptional Landscape of the Phytopathogenic Bacterium Dickeya dadantii

Dickeya dadantii is a phytopathogenic bacterium that causes soft rot in a wide range of plant hosts worldwide and a model organism for studying virulence gene regulation. The present study provides a comprehensive and annotated transcriptomic map of D. dadantii obtained by a computational method combining five independent transcriptomic data sets: (i) paired-end RNA sequencing (RNA-seq) data for a precise reconstruction of the RNA landscape; (ii) DNA microarray data providing transcriptional responses to a broad variety of environmental conditions; (iii) long-read Nanopore native RNA-seq data for isoform-level transcriptome validation and determination of transcription termination sites; (iv) differential RNA sequencing (dRNA-seq) data for the precise mapping of transcription start sites; (v) in planta DNA microarray data for a comparison of gene expression profiles between in vitro experiments and the early stages of plant infection. Our results show that transcription units sometimes coincide with predicted operons but are generally longer, most of them comprising internal promoters and terminators that generate alternative transcripts of variable gene composition. We characterize the occurrence of transcriptional read-through at terminators, which might play a basal regulation role and explain the extent of transcription beyond the scale of operons. We finally highlight the presence of noncontiguous operons and excludons in the D. dadantii genome, novel genomic arrangements that might contribute to the basal coordination of transcription. The highlighted transcriptional organization may allow D. dadantii to finely adjust its gene expression program for a rapid adaptation to fast-changing environments.

Predicting sepsis severity at first clinical presentation: The role of endotypes and mechanistic signatures

BACKGROUND

Inter-individual variability during sepsis limits appropriate triage of patients. Identifying, at first clinical presentation, gene expression signatures that predict subsequent severity will allow clinicians to identify the most at-risk groups of patients and enable appropriate antibiotic use.

METHODS

Blood RNA-Seq and clinical data were collected from 348 patients in four emergency rooms (ER) and one intensive-care-unit (ICU), and 44 healthy controls. Gene expression profiles were analyzed using machine learning and data mining to identify clinically relevant gene signatures reflecting disease severity, organ dysfunction, mortality, and specific endotypes/mechanisms.

FINDINGS

Gene expression signatures were obtained that predicted severity/organ dysfunction and mortality in both ER and ICU patients with accuracy/AUC of 77-80%. Network analysis revealed these signatures formed a coherent biological program, with specific but overlapping mechanisms/pathways. Given the heterogeneity of sepsis, we asked if patients could be assorted into discrete groups with distinct mechanisms (endotypes) and varying severity. Patients with early sepsis could be stratified into five distinct and novel mechanistic endotypes, named Neutrophilic-Suppressive/NPS, Inflammatory/INF, Innate-Host-Defense/IHD, Interferon/IFN, and Adaptive/ADA, each based on ∼200 unique gene expression differences, and distinct pathways/mechanisms (e.g., IL6/STAT3 in NPS). Endotypes had varying overall severity with two severe (NPS/INF) and one relatively benign (ADA) groupings, consistent with reanalysis of previous endotype studies. A 40 gene-classification tool (accuracy=96%) and several gene-pairs (accuracy=89-97%) accurately predicted endotype status in both ER and ICU validation cohorts.

INTERPRETATION

The severity and endotype signatures indicate that distinct immune signatures precede the onset of severe sepsis and lethality, providing a method to triage early sepsis patients.

Age-Related Changes of Gene Expression Profiles in Drosophila

An individual’s gene expression profile changes throughout their life. This change in gene expression is shaped by differences in physiological needs and functions between the younger and older organism. Despite intensive studies, the aging process is not fully understood, and several genes involved in this process may remain to be identified. Here we report a transcriptomic analysis of Drosophila melanogaster using microarrays. We compared the expression profiles of two-day-old female adult flies with those of 45-day-old flies. We identified 1184 genes with pronounced differences in expression level between young and old age groups. Most genes involved in muscle development/maintenance that display different levels of expression with age were downregulated in older flies. Many of these genes contributed to sarcomere formation and function. Several of these genes were functionally related to direct and indirect flight muscles; some of them were exclusively expressed in these muscles. Conversely, several genes involved in apoptosis processes were upregulated in aging flies. In addition, several genes involved in resistance to toxic chemicals were upregulated in aging flies, which is consistent with a global upregulation of the defense response system in aging flies. Finally, we randomly selected 12 genes among 232 genes with unknown function and generated transgenic flies expressing recombinant proteins fused with GFP protein to determine their subcellular expression. We also found that the knockdown of some of those 12 genes can affect the lifespan of flies.

Combining heterogeneous subgroups with graph-structured variable selection priors for Cox regression

Background

Important objectives in cancer research are the prediction of a patient’s risk based on molecular measurements such as gene expression data and the identification of new prognostic biomarkers (e.g. genes). In clinical practice, this is often challenging because patient cohorts are typically small and can be heterogeneous. In classical subgroup analysis, a separate prediction model is fitted using only the data of one specific cohort. However, this can lead to a loss of power when the sample size is small. Simple pooling of all cohorts, on the other hand, can lead to biased results, especially when the cohorts are heterogeneous.

Results

We propose a new Bayesian approach suitable for continuous molecular measurements and survival outcome that identifies the important predictors and provides a separate risk prediction model for each cohort. It allows sharing information between cohorts to increase power by assuming a graph linking predictors within and across different cohorts. The graph helps to identify pathways of functionally related genes and genes that are simultaneously prognostic in different cohorts.

Conclusions

Results demonstrate that our proposed approach is superior to the standard approaches in terms of prediction performance and increased power in variable selection when the sample size is small.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12859-021-04483-z.

Weighted Cox regression for the prediction of heterogeneous patient subgroups

BACKGROUND

An important task in clinical medicine is the construction of risk prediction models for specific subgroups of patients based on high-dimensional molecular measurements such as gene expression data. Major objectives in modeling high-dimensional data are good prediction performance and feature selection to find a subset of predictors that are truly associated with a clinical outcome such as a time-to-event endpoint. In clinical practice, this task is challenging since patient cohorts are typically small and can be heterogeneous with regard to their relationship between predictors and outcome. When data of several subgroups of patients with the same or similar disease are available, it is tempting to combine them to increase sample size, such as in multicenter studies. However, heterogeneity between subgroups can lead to biased results and subgroup-specific effects may remain undetected.

METHODS

For this situation, we propose a penalized Cox regression model with a weighted version of the Cox partial likelihood that includes patients of all subgroups but assigns them individual weights based on their subgroup affiliation. The weights are estimated from the data such that patients who are likely to belong to the subgroup of interest obtain higher weights in the subgroup-specific model.

RESULTS

Our proposed approach is evaluated through simulations and application to real lung cancer cohorts, and compared to existing approaches. Simulation results demonstrate that our proposed model is superior to standard approaches in terms of prediction performance and variable selection accuracy when the sample size is small.

CONCLUSIONS

The results suggest that sharing information between subgroups by incorporating appropriate weights into the likelihood can increase power to identify the prognostic covariates and improve risk prediction.

Tissue-specific transcriptional patterns underlie seasonal phenotypes in honey bees (Apis mellifera)

Faced with adverse conditions, such as winter in temperate regions or hot and dry conditions in tropical regions, many insect species enter a state of diapause, a period of dormancy associated with a reduction or arrest of physical activity, development and reproduction. Changes in common physiological pathways underlie diapause phenotypes in different insect species. However, most transcriptomic studies of diapause have not simultaneously evaluated and compared expression patterns in different tissues. Honey bees (Apis mellifera) represent a unique model system to study the mechanisms underpinning diapause-related phenotypes. In winter, honey bees exhibit a classic diapause phenotype, with reduced metabolic activity, increased physiological nutritional resources and altered hormonal profiles. However, winter bees actively heat their colony by vibrating their wing muscles; thus, this tissue is not quiescent. Here, we evaluated the transcriptional profiles of flight muscle tissue and fat body tissue (involved in nutrient storage, metabolism and immune function) of winter bees. We also evaluated two behavioural phenotypes of summer bees: nurses, which exhibit high nutritional stores and low flight activity, and foragers, which exhibit low nutritional stores and high flight activity. We found winter bees and nurses have similar fat body transcriptional profiles, whereas winter bees and foragers have similar flight muscle transcriptional profiles. Additionally, differentially expressed genes were enriched in diapause-related gene ontology terms. Thus, honey bees exhibit tissue-specific transcriptional profiles associated with seasonal phenotypes, laying the groundwork for future studies evaluating the mechanisms, evolution and consequences of this tissue-specific regulation.

AI applications in functional genomics

We review the current applications of artificial intelligence (AI) in functional genomics. The recent explosion of AI follows the remarkable achievements made possible by "deep learning", along with a burst of "big data" that can meet its hunger. Biology is about to overthrow astronomy as the paradigmatic representative of big data producer. This has been made possible by huge advancements in the field of high throughput technologies, applied to determine how the individual components of a biological system work together to accomplish different processes. The disciplines contributing to this bulk of data are collectively known as functional genomics. They consist in studies of: i) the information contained in the DNA (genomics); ii) the modifications that DNA can reversibly undergo (epigenomics); iii) the RNA transcripts originated by a genome (transcriptomics); iv) the ensemble of chemical modifications decorating different types of RNA transcripts (epitranscriptomics); v) the products of protein-coding transcripts (proteomics); and vi) the small molecules produced from cell metabolism (metabolomics) present in an organism or system at a given time, in physiological or pathological conditions. After reviewing main applications of AI in functional genomics, we discuss important accompanying issues, including ethical, legal and economic issues and the importance of explainability.

Gene Expression Analysis through Parallel Non-Negative Matrix Factorization

Genetic expression analysis is a principal tool to explain the behavior of genes in an organism when exposed to different experimental conditions. In the state of art, many clustering algorithms have been proposed. It is overwhelming the amount of biological data whose high-dimensional structure exceeds mostly current computational architectures. The computational time and memory consumption optimization actually become decisive factors in choosing clustering algorithms. We propose a clustering algorithm based on Non-negative Matrix Factorization and K-means to reduce data dimensionality but whilst preserving the biological context and prioritizing gene selection, and it is implemented within parallel GPU-based environments through the CUDA library. A well-known dataset is used in our tests and the quality of the results is measured through the Rand and Accuracy Index. The results show an increase in the acceleration of 6.22× compared to the sequential version. The algorithm is competitive in the biological datasets analysis and it is invariant with respect to the classes number and the size of the gene expression matrix.

Comparative Analysis of Machine Learning Algorithms on Surface Enhanced Raman Spectra of Clinical Staphylococcus Species

Raman spectroscopy (RS) is a widely used analytical technique based on the detection of molecular vibrations in a defined system, which generates Raman spectra that contain unique and highly resolved fingerprints of the system. However, the low intensity of normal Raman scattering effect greatly hinders its application. Recently, the newly emerged surface enhanced Raman spectroscopy (SERS) technique overcomes the problem by mixing metal nanoparticles such as gold and silver with samples, which greatly enhances signal intensity of Raman effects by orders of magnitudes when compared with regular RS. In clinical and research laboratories, SERS provides a great potential for fast, sensitive, label-free, and non-destructive microbial detection and identification with the assistance of appropriate machine learning (ML) algorithms. However, choosing an appropriate algorithm for a specific group of bacterial species remains challenging, because with the large volumes of data generated during SERS analysis not all algorithms could achieve a relatively high accuracy. In this study, we compared three unsupervised machine learning methods and 10 supervised machine learning methods, respectively, on 2,752 SERS spectra from 117 Staphylococcus strains belonging to nine clinically important Staphylococcus species in order to test the capacity of different machine learning methods for bacterial rapid differentiation and accurate prediction. According to the results, density-based spatial clustering of applications with noise (DBSCAN) showed the best clustering capacity (Rand index 0.9733) while convolutional neural network (CNN) topped all other supervised machine learning methods as the best model for predicting Staphylococcus species via SERS spectra (ACC 98.21%, AUC 99.93%). Taken together, this study shows that machine learning methods are capable of distinguishing closely related Staphylococcus species and therefore have great application potentials for bacterial pathogen diagnosis in clinical settings.

CGPE: an integrated online server for Cancer Gene and Pathway Exploration

SUMMARY

Cancer Gene and Pathway Explorer (CGPE) is developed to guide biological and clinical researchers, especially those with limited informatics and programming skills, performing preliminary cancer-related biomedical research using transcriptional data and publications. CGPE enables three user-friendly online analytical and visualization modules without requiring any local deployment. The GenePub HotIndex applies natural language processing, statistics and association discovery to provide analytical results on gene-specific PubMed publications, including gene-specific research trends, cancer types correlations, top-related genes and the WordCloud of publication profiles. The OnlineGSEA enables Gene Set Enrichment Analysis (GSEA) and results visualizations through an easy-to-follow interface for public or in-house transcriptional datasets, integrating the GSEA algorithm and preprocessed public TCGA and GEO datasets. The preprocessed datasets ensure gene sets analysis with appropriate pathway alternation and gene signatures. The CellLine Search presents evidence-based guidance for cell line selections with combined information on cell line dependency, gene expressions and pathway activity maps, which are valuable knowledge to have before conducting gene-related experiments. In a nutshell, the CGPE webserver provides a user-friendly, visual, intuitive and informative bioinformatics tool that allows biomedical researchers to perform efficient analyses and preliminary studies on in-house and publicly available bioinformatics data.

AVAILABILITY AND IMPLEMENTATION

The webserver is freely available online at https://cgpe.soic.iupui.edu.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

The transcriptome of rabbit conjunctiva in dry eye disease: Large-scale changes and similarity to the human dry eye

The pathophysiology of dry eye disease (DED) remains largely unknown, accounting in part for the lack of successful treatments. We explored the pathophysiology of DED using a rabbit model of chronic DED induced with 3 weekly injections of Concanavalin A into the periorbital lacrimal glands. The transcriptome of full-thickness's conjunctival tissue from rabbits with DED and from normal controls was determined using microarrays and, as needed, confirmatory real-time polymerase chain reactions. Results were subjected to bioinformatic analysis. DED induced large-scale changes in gene transcription involving 5,184 genes (22% of the total). Differentially expressed genes could be segregated into: functional modules and clusters; altered pathways; functionally linked genes; and groups of individual genes of known or suspected pathophysiological relevance to DED. A common feature of these subgroups is the breadth and magnitude of the changes that encompass ocular immunology and essentially all aspects of cell biology. Prominent changes concerned innate and adaptive immune responses; ocular surface inflammation; at least 25 significantly altered signaling pathways; a large number of chemokines; cell cycle; and apoptosis. Comparison of our findings to the limited extant transcriptomic data from DED patients associated with either Sjogren's syndrome or non-Sjogren's etiologies revealed a significant correlation between human and rabbit DED transcriptomes. Our data, establishing the large-scale transcriptomic changes of DED and their potential similarity to the human, underscore the enormous complexity of DED; establish a robust animal model of DED; will help expand our understanding of its pathophysiology; and could guide the development of successful therapeutic strategies.

Identification of miR-203a, mir-10a, and miR-194 as predictors for risk of lymphovascular invasion in head and neck cancers

Lymphovascular invasion (LVI) is an important prognostic indicator of lymph node metastasis and disease aggressiveness but clear molecular mechanisms mediating this in head and neck cancers (HNSC) remain undefined. To identify important microRNAs (miRNAs) in HNSC that associate with and are also predictive of increased risk of LVI, we used a combination of clustering algorithms, multiple regression analyses and machine learning approaches and analyzed miRNA expression profiles in the TCGA HNSC database. As the first step, we identified miRNAs with increased association with LVI as a binary variable. In order to determine whether the identified miRNAs would show functional clusters that are also indicative of increased risk for LVI, we carried out unsupervised as well as supervised clustering. Our results identified distinct clusters of miRNAs that are predictive of increased LVI. We further refined these findings using a Random forest approach, and miR-203a-3p, mir-10a-5p, and miR-194-5p to be most strongly associated with LVI. Pathway enrichment analysis showed these miRNAs targeted genes involved in Hippo signaling and fatty acid oxidation pathways that are mediators of lymph node metastasis. Specific association was also identified between the miRNAs associated with LVI and expression of several lymphangiogenic genes that could be critical for determination of therapeutic strategies.

Bioinformatic Analysis of Temporal and Spatial Proteome Alternations During Infections

Microbial pathogens have evolved numerous mechanisms to hijack host's systems, thus causing disease. This is mediated by alterations in the combined host-pathogen proteome in time and space. Mass spectrometry-based proteomics approaches have been developed and tailored to map disease progression. The result is complex multidimensional data that pose numerous analytic challenges for downstream interpretation. However, a systematic review of approaches for the downstream analysis of such data has been lacking in the field. In this review, we detail the steps of a typical temporal and spatial analysis, including data pre-processing steps (i.e., quality control, data normalization, the imputation of missing values, and dimensionality reduction), different statistical and machine learning approaches, validation, interpretation, and the extraction of biological information from mass spectrometry data. We also discuss current best practices for these steps based on a collection of independent studies to guide users in selecting the most suitable strategies for their dataset and analysis objectives. Moreover, we also compiled the list of commonly used R software packages for each step of the analysis. These could be easily integrated into one's analysis pipeline. Furthermore, we guide readers through various analysis steps by applying these workflows to mock and host-pathogen interaction data from public datasets. The workflows presented in this review will serve as an introduction for data analysis novices, while also helping established users update their data analysis pipelines. We conclude the review by discussing future directions and developments in temporal and spatial proteomics and data analysis approaches. Data analysis codes, prepared for this review are available from https://github.com/BabuLab-UofR/TempSpac, where guidelines and sample datasets are also offered for testing purposes.

Gene expression data classification using topology and machine learning models

Background

Interpretation of high-throughput gene expression data continues to require mathematical tools in data analysis that recognizes the shape of the data in high dimensions. Topological data analysis (TDA) has recently been successful in extracting robust features in several applications dealing with high dimensional constructs. In this work, we utilize some recent developments in TDA to curate gene expression data. Our work differs from the predecessors in two aspects: (1) Traditional TDA pipelines use topological signatures called barcodes to enhance feature vectors which are used for classification. In contrast, this work involves curating relevant features to obtain somewhat better representatives with the help of TDA. This representatives of the entire data facilitates better comprehension of the phenotype labels. (2) Most of the earlier works employ barcodes obtained using topological summaries as fingerprints for the data. Even though they are stable signatures, there exists no direct mapping between the data and said barcodes.

Results

The topology relevant curated data that we obtain provides an improvement in shallow learning as well as deep learning based supervised classifications. We further show that the representative cycles we compute have an unsupervised inclination towards phenotype labels. This work thus shows that topological signatures are able to comprehend gene expression levels and classify cohorts accordingly.

Conclusions

In this work, we engender representative persistent cycles to discern the gene expression data. These cycles allow us to directly procure genes entailed in similar processes.

Matrix adhesion and remodeling diversifies modes of cancer invasion across spatial scales

The metastasis of malignant epithelial tumors begins with the egress of transformed cells from the confines of their basement membrane (BM) to their surrounding collagen-rich stroma. Invasion can be morphologically diverse: when breast cancer cells are separately cultured within BM-like matrix, collagen I (Coll I), or a combination of both, they exhibit collective-, dispersed mesenchymal-, and a mixed collective-dispersed (multimodal)- invasion, respectively. In this paper, we asked how distinct these invasive modes are with respect to the cellular and microenvironmental cues that drive them. A rigorous computational exploration of invasion was performed within an experimentally motivated Cellular Potts-based modeling environment. The model comprised of adhesive interactions between cancer cells, BM- and Coll I-like extracellular matrix (ECM), and reaction-diffusion-based remodeling of ECM. The model outputs were parameters cognate to dispersed- and collective- invasion. A clustering analysis of the output distribution curated through a careful examination of subsumed phenotypes suggested at least four distinct invasive states: dispersed, papillary-collective, bulk-collective, and multimodal, in addition to an indolent/non-invasive state. Mapping input values to specific output clusters suggested that each of these invasive states are specified by distinct input signatures of proliferation, adhesion and ECM remodeling. In addition, specific input perturbations allowed transitions between the clusters and revealed the variation in the robustness between the invasive states. Our systems-level approach proffers quantitative insights into how the diversity in ECM microenvironments may steer invasion into diverse phenotypic modes during early dissemination of breast cancer and contributes to tumor heterogeneity.

Identification of transcriptional subtypes in lung adenocarcinoma and squamous cell carcinoma through integrative analysis of microarray and RNA sequencing data

Classification of tumors into subtypes can inform personalized approaches to treatment including the choice of targeted therapies. The two most common lung cancer histological subtypes, lung adenocarcinoma and lung squamous cell carcinoma, have been previously divided into transcriptional subtypes using microarray data, and corresponding signatures were subsequently used to classify RNA-seq data. Cross-platform unsupervised classification facilitates the identification of robust transcriptional subtypes by combining vast amounts of publicly available microarray and RNA-seq data. However, cross-platform classification is challenging because of intrinsic differences in data generated using the two gene expression profiling technologies. In this report, we show that robust gene expression subtypes can be identified in integrated data representing over 3500 normal and tumor lung samples profiled using two widely used platforms, Affymetrix HG-U133 Plus 2.0 Array and Illumina HiSeq RNA sequencing. We tested and analyzed consensus clustering for 384 combinations of data processing methods. The agreement between subtypes identified in single-platform and cross-platform normalized data was then evaluated using a variety of statistics. Results show that unsupervised learning can be achieved with combined microarray and RNA-seq data using selected preprocessing, cross-platform normalization, and unsupervised feature selection methods. Our analysis confirmed three lung adenocarcinoma transcriptional subtypes, but only two consistent subtypes in squamous cell carcinoma, as opposed to four subtypes previously identified. Further analysis showed that tumor subtypes were associated with distinct patterns of genomic alterations in genes coding for therapeutic targets. Importantly, by integrating quantitative proteomics data, we were able to identify tumor subtype biomarkers that effectively classify samples on the basis of both gene and protein expression. This study provides the basis for further integrative data analysis across gene and protein expression profiling platforms.

Enhancing semantic belief function to handle decision conflicts in SoS using k-means clustering

BACKGROUND

The endeavouring to offer complex special functions from individual systems gave rise to what is known as the System of Systems (SoS). SoS co-integrating systems together while allowing for absorbing more systems in the future. SoS as an integrated system simplifies operations, reduces costs, and ensures efficiency. However, conflict may result while co-integrating systems, violating the main benefits of SoS. This paper is concerned with enhancing the time required to detect and solve such conflicts.

METHODS

We adopted the k-means clustering technique to enhance the detection and solving of conflict resulting while co-integrating new systems into an existing SoS. Instead of dealing with SoS as a single entity, we partition it into clusters. Each cluster contains nearby systems according to pre-specified criteria. We can consider each cluster a Sub SoS (S-SoS). By doing so, the conflict that may arise while co-integrating new systems can be detected and solved in a shorter time. We propose the Smart Semantic Belief Function Clustered System of Systems (SSBFCSoS), which is an enhancement of the Ontology Belief Function System of Systems (OBFSoS).

RESULTS

The proposed method proved the ability to rapidly detect and resolve conflicts. It showed the ability to accommodate more systems as well, therefore achieving the objectives of SoS. In order to test the applicability of the SSBFCSoS and compare its performance with other approaches, two datasets were employed. They are (Glest & StarCraft Brood War). With each dataset, 15 test cases were examined. We achieved, on average, 89% in solving the conflict compared to 77% for other approaches. Moreover, it showed an acceleration of up to proportionality over previous approaches for about 16% in solving conflicts as well. Besides, it reduced the frequency of the same conflicts by approximately 23% better than the other method, not only in the same cluster but even while combining different clusters.

Deep learning-based clustering approaches for bioinformatics

Clustering is central to many data-driven bioinformatics research and serves a powerful computational method. In particular, clustering helps at analyzing unstructured and high-dimensional data in the form of sequences, expressions, texts and images. Further, clustering is used to gain insights into biological processes in the genomics level, e.g. clustering of gene expressions provides insights on the natural structure inherent in the data, understanding gene functions, cellular processes, subtypes of cells and understanding gene regulations. Subsequently, clustering approaches, including hierarchical, centroid-based, distribution-based, density-based and self-organizing maps, have long been studied and used in classical machine learning settings. In contrast, deep learning (DL)-based representation and feature learning for clustering have not been reviewed and employed extensively. Since the quality of clustering is not only dependent on the distribution of data points but also on the learned representation, deep neural networks can be effective means to transform mappings from a high-dimensional data space into a lower-dimensional feature space, leading to improved clustering results. In this paper, we review state-of-the-art DL-based approaches for cluster analysis that are based on representation learning, which we hope to be useful, particularly for bioinformatics research. Further, we explore in detail the training procedures of DL-based clustering algorithms, point out different clustering quality metrics and evaluate several DL-based approaches on three bioinformatics use cases, including bioimaging, cancer genomics and biomedical text mining. We believe this review and the evaluation results will provide valuable insights and serve a starting point for researchers wanting to apply DL-based unsupervised methods to solve emerging bioinformatics research problems.

mbkmeans: Fast clustering for single cell data using mini-batch k-means

Single-cell RNA-Sequencing (scRNA-seq) is the most widely used high-throughput technology to measure genome-wide gene expression at the single-cell level. One of the most common analyses of scRNA-seq data detects distinct subpopulations of cells through the use of unsupervised clustering algorithms. However, recent advances in scRNA-seq technologies result in current datasets ranging from thousands to millions of cells. Popular clustering algorithms, such as k-means, typically require the data to be loaded entirely into memory and therefore can be slow or impossible to run with large datasets. To address this problem, we developed the mbkmeans R/Bioconductor package, an open-source implementation of the mini-batch k-means algorithm. Our package allows for on-disk data representations, such as the common HDF5 file format widely used for single-cell data, that do not require all the data to be loaded into memory at one time. We demonstrate the performance of the mbkmeans package using large datasets, including one with 1.3 million cells. We also highlight and compare the computing performance of mbkmeans against the standard implementation of k-means and other popular single-cell clustering methods. Our software package is available in Bioconductor at https://bioconductor.org/packages/mbkmeans.

Comparative Transcriptomics of the Bovine Apicomplexan Parasite Theileria parva Developmental Stages Reveals Massive Gene Expression Variation and Potential Vaccine Antigens

Theileria parva is a protozoan parasite that causes East Coast fever (ECF), an economically important disease of cattle in Africa. It is transmitted mainly by the tick Rhipicephalus appendiculatus. Research efforts to develop a subunit vaccine based on parasite neutralizing antibodies and cytotoxic T-lymphocytes have met with limited success. The molecular mechanisms underlying T. parva life cycle stages in the tick vector and bovine host are poorly understood, thus limiting progress toward an effective and efficient control of ECF. Transcriptomics has been used to identify candidate vaccine antigens or markers associated with virulence and disease pathology. Therefore, characterization of gene expression throughout the parasite's life cycle should shed light on host-pathogen interactions in ECF and identify genes underlying differences in parasite stages as well as potential, novel therapeutic targets. Recently, the first gene expression profiling of T. parva was conducted for the sporoblast, sporozoite, and schizont stages. The sporozoite is infective to cattle, whereas the schizont is the major pathogenic form of the parasite. The schizont can differentiate into piroplasm, which is infective to the tick vector. The present study was designed to extend the T. parva gene expression profiling to the piroplasm stage with reference to the schizont. Pairwise comparison revealed that 3,279 of a possible 4,084 protein coding genes were differentially expressed, with 1,623 (49%) genes upregulated and 1,656 (51%) downregulated in the piroplasm relative to the schizont. In addition, over 200 genes were stage-specific. In general, there were more molecular functions, biological processes, subcellular localizations, and pathways significantly enriched in the piroplasm than in the schizont. Using known antigens as benchmarks, we identified several new potential vaccine antigens, including TP04_0076 and TP04_0640, which were highly immunogenic in naturally T. parva-infected cattle. All the candidate vaccine antigens identified have yet to be investigated for their capacity to induce protective immune response against ECF.