GSEApy: a comprehensive package for performing gene set enrichment analysis in Python

scGPT: toward building a foundation model for single-cell multi-omics using generative AI

Generative pretrained models have achieved remarkable success in various domains such as language and computer vision. Specifically, the combination of large-scale diverse datasets and pretrained transformers has emerged as a promising approach for developing foundation models. Drawing parallels between language and cellular biology (in which texts comprise words; similarly, cells are defined by genes), our study probes the applicability of foundation models to advance cellular biology and genetic research. Using burgeoning single-cell sequencing data, we have constructed a foundation model for single-cell biology, scGPT, based on a generative pretrained transformer across a repository of over 33 million cells. Our findings illustrate that scGPT effectively distills critical biological insights concerning genes and cells. Through further adaptation of transfer learning, scGPT can be optimized to achieve superior performance across diverse downstream applications. This includes tasks such as cell type annotation, multi-batch integration, multi-omic integration, perturbation response prediction and gene network inference.

Cohesin composition and dosage independently affect early development in zebrafish

Cohesin, a chromatin-associated protein complex with four core subunits (Smc1a, Smc3, Rad21 and either Stag1 or 2), has a central role in cell proliferation and gene expression in metazoans. Human developmental disorders termed 'cohesinopathies' are characterized by germline variants of cohesin or its regulators that do not entirely eliminate cohesin function. However, it is not clear whether mutations in individual cohesin subunits have independent developmental consequences. Here, we show that zebrafish rad21 or stag2b mutants independently influence embryonic tailbud development. Both mutants have altered mesoderm induction, but only homozygous or heterozygous rad21 mutation affects cell cycle gene expression. stag2b mutants have narrower notochords and reduced Wnt signaling in neuromesodermal progenitors as revealed by single-cell RNA sequencing. Stimulation of Wnt signaling rescues transcription and morphology in stag2b, but not rad21, mutants. Our results suggest that mutations altering the quantity versus composition of cohesin have independent developmental consequences, with implications for the understanding and management of cohesinopathies.

In vivo single-cell CRISPR uncovers distinct TNF programmes in tumour evolution

The tumour evolution model posits that malignant transformation is preceded by randomly distributed driver mutations in cancer genes, which cause clonal expansions in phenotypically normal tissues. Although clonal expansions can remodel entire tissues1-3, the mechanisms that result in only a small number of clones transforming into malignant tumours remain unknown. Here we develop an in vivo single-cell CRISPR strategy to systematically investigate tissue-wide clonal dynamics of the 150 most frequently mutated squamous cell carcinoma genes. We couple ultrasound-guided in utero lentiviral microinjections, single-cell RNA sequencing and guide capture to longitudinally monitor clonal expansions and document their underlying gene programmes at single-cell transcriptomic resolution. We uncover a tumour necrosis factor (TNF) signalling module, which is dependent on TNF receptor 1 and involving macrophages, that acts as a generalizable driver of clonal expansions in epithelial tissues. Conversely, during tumorigenesis, the TNF signalling module is downregulated. Instead, we identify a subpopulation of invasive cancer cells that switch to an autocrine TNF gene programme associated with epithelial-mesenchymal transition. Finally, we provide in vivo evidence that the autocrine TNF gene programme is sufficient to mediate invasive properties and show that the TNF signature correlates with shorter overall survival of patients with squamous cell carcinoma. Collectively, our study demonstrates the power of applying in vivo single-cell CRISPR screening to mammalian tissues, unveils distinct TNF programmes in tumour evolution and highlights the importance of understanding the relationship between clonal expansions in epithelia and tumorigenesis.

SAFER: sub-hypergraph attention-based neural network for predicting effective responses to dose combinations

BACKGROUND

The potential benefits of drug combination synergy in cancer medicine are significant, yet the risks must be carefully managed due to the possibility of increased toxicity. Although artificial intelligence applications have demonstrated notable success in predicting drug combination synergy, several key challenges persist: (1) Existing models often predict average synergy values across a restricted range of testing dosages, neglecting crucial dose amounts and the mechanisms of action of the drugs involved. (2) Many graph-based models rely on static protein-protein interactions, failing to adapt to dynamic and higher-order relationships. These limitations constrain the applicability of current methods.

RESULTS

We introduce SAFER, a Sub-hypergraph Attention-based graph model, addressing these issues by incorporating complex relationships among biological knowledge networks and considering dosing effects on subject-specific networks. SAFER outperformed previous models on the benchmark and the independent test set. The analysis of subgraph attention weight for the lung cancer cell line highlighted JAK-STAT signaling pathway, PRDM12, ZNF781, and CDC5L that have been implicated in lung fibrosis.

CONCLUSIONS

SAFER presents an interpretable framework designed to identify drug-responsive signals. Tailored for comprehending dose effects on subject-specific molecular contexts, our model uniquely captures dose-level drug combination responses. This capability unlocks previously inaccessible avenues of investigation compared to earlier models. Furthermore, the SAFER framework can be leveraged by future inquiries to investigate molecular networks that uniquely characterize individual patients and can be applied to prioritize personalized effective treatment based on safe dose combinations.

Comparison of uridine and N1-methylpseudouridine mRNA platforms in development of an Andes virus vaccine

The rodent-borne Andes virus (ANDV) causes a severe disease in humans. We developed an ANDV mRNA vaccine based on the M segment of the viral genome, either with regular uridine (U-mRNA) or N1-methylpseudouridine (m1Ψ-mRNA). Female mice immunized by m1Ψ-mRNA developed slightly greater germinal center (GC) responses than U-mRNA-immunized mice. Single cell RNA and BCR sequencing of the GC B cells revealed similar levels of activation, except an additional cluster of cells exhibiting interferon response in animals vaccinated with U-mRNA but not m1Ψ-mRNA. Similar immunoglobulin class-switching and somatic hypermutations were observed in response to the vaccines. Female Syrian hamsters were immunized via a prime-boost regimen with two doses of each vaccine. The titers of glycoprotein-binding antibodies were greater for U-mRNA construct than for m1Ψ-mRNA construct; however, the titers of ANDV-neutralizing antibodies were similar. Vaccinated animals were challenged with a lethal dose of ANDV, along with a naïve control group. All control animals and two animals vaccinated with a lower dose of m1Ψ-mRNA succumbed to infection whereas other vaccinated animals survived without evidence of virus replication. The data demonstrate the development of a protective vaccine against ANDV and the lack of a substantial effect of m1Ψ modification on immunogenicity and protection in rodents.

Whole brain alignment of spatial transcriptomics between humans and mice with BrainAlign

The increasing utilization of mouse models in human neuroscience research places higher demands on computational methods to translate findings from the mouse brain to the human one. In this study, we develop BrainAlign, a self-supervised learning approach, for the whole brain alignment of spatial transcriptomics (ST) between humans and mice. BrainAlign encodes spots and genes simultaneously in two separated shared embedding spaces by a heterogeneous graph neural network. We demonstrate that BrainAlign could integrate cross-species spots into the embedding space and reveal the conserved brain regions supported by ST information, which facilitates the detection of homologous regions between humans and mice. Genomic analysis further presents gene expression connections between humans and mice and reveals similar expression patterns for marker genes. Moreover, BrainAlign can accurately map spatially similar homologous regions or clusters onto a unified spatial structural domain while preserving their relative positions.

ADP-ribosylome analysis reveals homogeneous DNA-damage-induced serine ADP-ribosylation across wild-type and BRCA-mutant breast cancer cell lines

ADP-ribosylation (ADPr) signaling plays a crucial role in DNA damage response. Inhibitors against the main enzyme catalyzing ADPr after DNA damage, poly(ADP-ribose) polymerase 1 (PARP1), are used to treat patients with breast cancer harboring BRCA1/2 mutations. However, resistance to PARP inhibitors (PARPi) is a major obstacle in treating patients. To understand the role of ADPr in PARPi sensitivity, we use liquid chromatography-tandem mass spectrometry (LC-MS/MS) to analyze ADPr in six breast cancer cell lines exhibiting different PARPi sensitivities. We identify 1,632 sites on 777 proteins across all cell lines, primarily on serine residues, with site-specific overlap of targeted residues across DNA-damage-related proteins across all cell lines, demonstrating high conservation of serine ADPr-signaling networks upon DNA damage. Furthermore, we observe site-specific differences in ADPr intensities in PARPi-sensitive BRCA mutants and unique ADPr sites in PARPi-resistant BRCA-mutant HCC1937 cells, which have low poly(ADP-ribose) glycohydrolase (PARG) levels and longer ADPr chains on PARP1.

The metabolic role of vitamin D in children's neurodevelopment: a network study

Neurodevelopmental disorders are rapidly increasing in prevalence and have been linked to various environmental risk factors. Mounting evidence suggests a potential role of vitamin D in child neurodevelopment, though the causal mechanisms remain largely unknown. Here, we investigate how vitamin D deficiency affects children's communication development, particularly in relation to Autism Spectrum Disorder (ASD). We do so by developing an integrative network approach that combines metabolomic profiles, clinical traits, and neurodevelopmental data from a pediatric cohort. Our results show that low levels of vitamin D are associated with changes in the metabolic networks of tryptophan, linoleic, and fatty acid metabolism. These changes correlate with distinct ASD-related phenotypes, including delayed communication skills and respiratory dysfunctions. Additionally, our analysis suggests the kynurenine and serotonin sub-pathways may mediate the effect of vitamin D on early life communication development. Altogether, our findings provide metabolome-wide insights into the potential of vitamin D as a therapeutic option for ASD and other communication disorders.

A subpopulation of human bone marrow erythroid cells displays a myeloid gene expression signature similar to that of classic monocytes

Erythroid cells, serving as progenitors and precursors to erythrocytes responsible for oxygen transport, were shown to exhibit an immunosuppressive and immunoregulatory phenotype. Previous investigations from our research group have revealed an antimicrobial gene expression profile within murine bone marrow erythroid cells which suggested a role for erythroid cells in innate immunity. In the present study, we focused on elucidating the characteristics of human bone marrow erythroid cells through comprehensive analyses, including NanoString gene expression profiling utilizing the Immune Response V2 panel, a BioPlex examination of chemokine and TGF-beta family proteins secretion, and analysis of publicly available single-cell RNA-seq data. Our findings demonstrate that an erythroid cell subpopulation manifests a myeloid-like gene expression signature comprised of antibacterial immunity and neutrophil chemotaxis genes which suggests an involvement of human erythroid cells in the innate immunity. Furthermore, we found that human erythroid cells secreted CCL22, CCL24, CXCL5, CXCL8, and MIF chemokines. The ability of human erythroid cells to express these chemokines might facilitate the restriction of immune cells in the bone marrow under normal conditions or contribute to the ability of erythroid cells to induce local immunosuppression by recruiting immune cells in their immediate vicinity in case of extramedullary hematopoiesis.

Spatiotemporal transcriptomic profiling and modeling of mouse brain at single-cell resolution reveals cell proximity effects of aging and rejuvenation

Old age is associated with a decline in cognitive function and an increase in neurodegenerative disease risk1. Brain aging is complex and accompanied by many cellular changes2-20. However, the influence that aged cells have on neighboring cells and how this contributes to tissue decline is unknown. More generally, the tools to systematically address this question in aging tissues have not yet been developed. Here, we generate spatiotemporal data at single-cell resolution for the mouse brain across lifespan, and we develop the first machine learning models based on spatial transcriptomics ('spatial aging clocks') to reveal cell proximity effects during brain aging and rejuvenation. We collect a single-cell spatial transcriptomics brain atlas of 4.2 million cells from 20 distinct ages and across two rejuvenating interventions-exercise and partial reprogramming. We identify spatial and cell type-specific transcriptomic fingerprints of aging, rejuvenation, and disease, including for rare cell types. Using spatial aging clocks and deep learning models, we find that T cells, which infiltrate the brain with age, have a striking pro-aging proximity effect on neighboring cells. Surprisingly, neural stem cells have a strong pro-rejuvenating effect on neighboring cells. By developing computational tools to identify mediators of these proximity effects, we find that pro-aging T cells trigger a local inflammatory response likely via interferon-γ whereas pro-rejuvenating neural stem cells impact the metabolism of neighboring cells possibly via growth factors (e.g. vascular endothelial growth factor) and extracellular vesicles, and we experimentally validate some of these predictions. These results suggest that rare cells can have a drastic influence on their neighbors and could be targeted to counter tissue aging. We anticipate that these spatial aging clocks will not only allow scalable assessment of the efficacy of interventions for aging and disease but also represent a new tool for studying cell-cell interactions in many spatial contexts.

Increased prevalence of hybrid epithelial/mesenchymal state and enhanced phenotypic heterogeneity in basal breast cancer

Intra-tumoral phenotypic heterogeneity promotes tumor relapse and therapeutic resistance and remains an unsolved clinical challenge. Decoding the interconnections among different biological axes of plasticity is crucial to understand the molecular origins of phenotypic heterogeneity. Here, we use multi-modal transcriptomic data-bulk, single-cell, and spatial transcriptomics-from breast cancer cell lines and primary tumor samples, to identify associations between epithelial-mesenchymal transition (EMT) and luminal-basal plasticity-two key processes that enable heterogeneity. We show that luminal breast cancer strongly associates with an epithelial cell state, but basal breast cancer is associated with hybrid epithelial/mesenchymal phenotype(s) and higher phenotypic heterogeneity. Mathematical modeling of core underlying gene regulatory networks representative of the crosstalk between the luminal-basal and epithelial-mesenchymal axes elucidate mechanistic underpinnings of the observed associations from transcriptomic data. Our systems-based approach integrating multi-modal data analysis with mechanism-based modeling offers a predictive framework to characterize intra-tumor heterogeneity and identify interventions to restrict it.

scSemiProfiler: Advancing large-scale single-cell studies through semi-profiling with deep generative models and active learning

Single-cell sequencing is a crucial tool for dissecting the cellular intricacies of complex diseases. Its prohibitive cost, however, hampers its application in expansive biomedical studies. Traditional cellular deconvolution approaches can infer cell type proportions from more affordable bulk sequencing data, yet they fall short in providing the detailed resolution required for single-cell-level analyses. To overcome this challenge, we introduce "scSemiProfiler", an innovative computational framework that marries deep generative models with active learning strategies. This method adeptly infers single-cell profiles across large cohorts by fusing bulk sequencing data with targeted single-cell sequencing from a few rigorously chosen representatives. Extensive validation across heterogeneous datasets verifies the precision of our semi-profiling approach, aligning closely with true single-cell profiling data and empowering refined cellular analyses. Originally developed for extensive disease cohorts, "scSemiProfiler" is adaptable for broad applications. It provides a scalable, cost-effective solution for single-cell profiling, facilitating in-depth cellular investigation in various biological domains.

OmicVerse: a framework for bridging and deepening insights across bulk and single-cell sequencing

Single-cell sequencing is frequently affected by "omission" due to limitations in sequencing throughput, yet bulk RNA-seq may contain these ostensibly "omitted" cells. Here, we introduce the single cell trajectory blending from Bulk RNA-seq (BulkTrajBlend) algorithm, a component of the OmicVerse suite that leverages a Beta-Variational AutoEncoder for data deconvolution and graph neural networks for the discovery of overlapping communities. This approach effectively interpolates and restores the continuity of "omitted" cells within single-cell RNA sequencing datasets. Furthermore, OmicVerse provides an extensive toolkit for both bulk and single cell RNA-seq analysis, offering seamless access to diverse methodologies, streamlining computational processes, fostering exquisite data visualization, and facilitating the extraction of significant biological insights to advance scientific research.

Fatecode enables cell fate regulator prediction using classification-supervised autoencoder perturbation

Cell reprogramming, which guides the conversion between cell states, is a promising technology for tissue repair and regeneration, with the ultimate goal of accelerating recovery from diseases or injuries. To accomplish this, regulators must be identified and manipulated to control cell fate. We propose Fatecode, a computational method that predicts cell fate regulators based only on single-cell RNA sequencing (scRNA-seq) data. Fatecode learns a latent representation of the scRNA-seq data using a deep learning-based classification-supervised autoencoder and then performs in silico perturbation experiments on the latent representation to predict genes that, when perturbed, would alter the original cell type distribution to increase or decrease the population size of a cell type of interest. We assessed Fatecode's performance using simulations from a mechanistic gene-regulatory network model and scRNA-seq data mapping blood and brain development of different organisms. Our results suggest that Fatecode can detect known cell fate regulators from single-cell transcriptomics datasets.

Mutant IDH1 inhibition induces dsDNA sensing to activate tumor immunity

Isocitrate dehydrogenase 1 (IDH1) is the most commonly mutated metabolic gene across human cancers. Mutant IDH1 (mIDH1) generates the oncometabolite (R)-2-hydroxyglutarate, disrupting enzymes involved in epigenetics and other processes. A hallmark of IDH1-mutant solid tumors is T cell exclusion, whereas mIDH1 inhibition in preclinical models restores antitumor immunity. Here, we define a cell-autonomous mechanism of mIDH1-driven immune evasion. IDH1-mutant solid tumors show selective hypermethylation and silencing of the cytoplasmic double-stranded DNA (dsDNA) sensor CGAS, compromising innate immune signaling. mIDH1 inhibition restores DNA demethylation, derepressing CGAS and transposable element (TE) subclasses. dsDNA produced by TE-reverse transcriptase (TE-RT) activates cGAS, triggering viral mimicry and stimulating antitumor immunity. In summary, we demonstrate that mIDH1 epigenetically suppresses innate immunity and link endogenous RT activity to the mechanism of action of a US Food and Drug Administration-approved oncology drug.

Generation of a biliary tract cancer cell line atlas reveals molecular subtypes and therapeutic targets

Biliary tract cancers (BTCs) are a group of deadly malignancies encompassing intrahepatic and extrahepatic cholangiocarcinoma, gallbladder carcinoma, and ampullary carcinoma. Here, we present the integrative analysis of 63 BTC cell lines via multi-omics clustering and genome- scale CRISPR screens, providing a platform to illuminate BTC biology and inform therapeutic development. We identify dependencies broadly enriched in BTC compared to other cancers as well as dependencies selective to the anatomic subtypes. Notably, cholangiocarcinoma cell lines are stratified into distinct lineage subtypes based on biliary or dual biliary/hepatocyte marker signatures, associated with dependency on specific lineage survival factors. Transcriptional analysis of patient specimens demonstrates the prognostic significance of these lineage subtypes. Additionally, we delineate strategies to enhance targeted therapies or to overcome resistance in cell lines with key driver gene mutations. Furthermore, clustering based on dependencies and proteomics data elucidates unexpected functional relationships, including a BTC subgroup with partial squamous differentiation. Thus, this cell line atlas reveals potential therapeutic targets in molecularly defined BTCs, unveils biologically distinct disease subtypes, and offers a vital resource for BTC research.

GPS-SUMO 2.0: an updated online service for the prediction of SUMOylation sites and SUMO-interacting motifs

Small ubiquitin-like modifiers (SUMOs) are tiny but important protein regulators involved in orchestrating a broad spectrum of biological processes, either by covalently modifying protein substrates or by noncovalently interacting with other proteins. Here, we report an updated server, GPS-SUMO 2.0, for the prediction of SUMOylation sites and SUMO-interacting motifs (SIMs). For predictor training, we adopted three machine learning algorithms, penalized logistic regression (PLR), a deep neural network (DNN), and a transformer, and used 52 404 nonredundant SUMOylation sites in 8262 proteins and 163 SIMs in 102 proteins. To further increase the accuracy of predicting SUMOylation sites, a pretraining model was first constructed using 145 545 protein lysine modification sites, followed by transfer learning to fine-tune the model. GPS-SUMO 2.0 exhibited greater accuracy in predicting SUMOylation sites than did other existing tools. For users, one or multiple protein sequences or identifiers can be input, and the prediction results are shown in a tabular list. In addition to the basic statistics, we integrated knowledge from 35 public resources to annotate SUMOylation sites or SIMs. The GPS-SUMO 2.0 server is freely available at https://sumo.biocuckoo.cn/. We believe that GPS-SUMO 2.0 can serve as a useful tool for further analysis of SUMOylation and SUMO interactions.

Targeting IL-1 controls refractory pityriasis rubra pilaris

Pityriasis rubra pilaris (PRP) is a rare inflammatory skin disease with a poorly understood pathogenesis. Through a molecularly driven precision medicine approach and an extensive mechanistic pathway analysis in PRP skin samples, compared to psoriasis, atopic dermatitis, healed PRP, and healthy controls, we identified IL-1β as a key mediator, orchestrating an NF-κB-mediated IL-1β-CCL20 axis, including activation of CARD14 and NOD2. Treatment of three patients with the IL-1 antagonists anakinra and canakinumab resulted in rapid clinical improvement and reversal of the PRP-associated molecular signature with a 50% improvement in skin lesions after 2 to 3 weeks. This transcriptional signature was consistent with in vitro stimulation of keratinocytes with IL-1β. With the central role of IL-1β underscoring its potential as a therapeutic target, our findings propose a redefinition of PRP as an autoinflammatory keratinization disorder. Further clinical trials are needed to validate the efficacy of IL-1β antagonists in PRP.

EstroGene2.0: A multi-omic database of response to estrogens, ER-modulators, and resistance to endocrine therapies in breast cancer

Endocrine therapies targeting the estrogen receptor (ER/ESR1) are the cornerstone to treat ER-positive breast cancers patients, but resistance often limits their effectiveness. Understanding the molecular mechanisms is thus key to optimize the existing drugs and to develop new ER-modulators. Notable progress has been made although the fragmented way data is reported has reduced their potential impact. Here, we introduce EstroGene2.0, an expanded database of its precursor 1.0 version. EstroGene2.0 focusses on response and resistance to endocrine therapies in breast cancer models. Incorporating multi-omic profiling of 361 experiments from 212 studies across 28 cell lines, a user-friendly browser offers comprehensive data visualization and metadata mining capabilities (https://estrogeneii.web.app/). Taking advantage of the harmonized data collection, our follow-up meta-analysis revealed substantial diversity in response to different classes of ER-modulators including SERMs, SERDs, SERCA and LDD/PROTAC. Notably, endocrine resistant models exhibit a spectrum of transcriptomic alterations including a contra-directional shift in ER and interferon signaling, which is recapitulated clinically. Furthermore, dissecting multiple ESR1-mutant cell models revealed the different clinical relevance of genome-edited versus ectopic overexpression model engineering and identified high-confidence mutant-ER targets, such as NPY1R. These examples demonstrate how EstroGene2.0 helps investigate breast cancer's response to endocrine therapies and explore resistance mechanisms.

aPEAch: Automated Pipeline for End-to-End Analysis of Epigenomic and Transcriptomic Data

With the advent of next-generation sequencing (NGS), experimental techniques that capture the biological significance of DNA loci or RNA molecules have emerged as fundamental tools for studying the epigenome and transcriptional regulation on a genome-wide scale. The volume of the generated data and the underlying complexity regarding their analysis highlight the need for robust and easy-to-use computational analytic methods that can streamline the process and provide valuable biological insights. Our solution, aPEAch, is an automated pipeline that facilitates the end-to-end analysis of both DNA- and RNA-sequencing assays, including small RNA sequencing, from assessing the quality of the input sample files to answering meaningful biological questions by exploiting the rich information embedded in biological data. Our method is implemented in Python, based on a modular approach that enables users to choose the path and extent of the analysis and the representations of the results. The pipeline can process samples with single or multiple replicates in batches, allowing the ease of use and reproducibility of the analysis across all samples. aPEAch provides a variety of sample metrics such as quality control reports, fragment size distribution plots, and all intermediate output files, enabling the pipeline to be re-executed with different parameters or algorithms, along with the publication-ready visualization of the results. Furthermore, aPEAch seamlessly incorporates advanced unsupervised learning analyses by automating clustering optimization and visualization, thus providing invaluable insight into the underlying biological mechanisms.

Atf4 protects islet β-cell identity and function under acute glucose-induced stress but promotes β-cell failure in the presence of free fatty acid

Glucolipotoxicity, caused by combined hyperglycemia and hyperlipidemia, results in β-cell failure and type 2 diabetes (T2D) via cellular stress-related mechanisms. Activating transcription factor 4 (Atf4) is an essential effector of stress response. We show here that Atf4 expression in β-cells is dispensable for glucose homeostasis in young mice, but it is required for β-cell function during aging and under obesity-related metabolic stress. Henceforth, aged Atf4- deficient β-cells display compromised secretory function under acute hyperglycemia. In contrast, they are resistant to acute free fatty acid-induced loss-of identity and dysfunction. At molecular level, Atf4 -deficient β-cells down-regulate genes involved in protein translation, reducing β-cell identity gene products under high glucose. They also upregulate several genes involved in lipid metabolism or signaling, likely contributing to their resistance to free fatty acid-induced dysfunction. These results suggest that Atf4 activation is required for β-cell identity and function under high glucose, but this paradoxically induces β-cell failure in the presence of high levels of free fatty acids. Different branches of Atf4 activity could be manipulated for protecting β-cells from metabolic stress-induced failure.

Highlights

Atf4 is dispensable in β-cells in young miceAtf4 protects β-cells under high glucoseAtf4 exacerbate fatty acid-induced β-cell defectsAtf4 activates translation but depresses lipid-metabolism.

Supervised discovery of interpretable gene programs from single-cell data

Factor analysis decomposes single-cell gene expression data into a minimal set of gene programs that correspond to processes executed by cells in a sample. However, matrix factorization methods are prone to technical artifacts and poor factor interpretability. We address these concerns with Spectra, an algorithm that combines user-provided gene programs with the detection of novel programs that together best explain expression covariation. Spectra incorporates existing gene sets and cell-type labels as prior biological information, explicitly models cell type and represents input gene sets as a gene-gene knowledge graph using a penalty function to guide factorization toward the input graph. We show that Spectra outperforms existing approaches in challenging tumor immune contexts, as it finds factors that change under immune checkpoint therapy, disentangles the highly correlated features of CD8+ T cell tumor reactivity and exhaustion, finds a program that explains continuous macrophage state changes under therapy and identifies cell-type-specific immune metabolic programs.

Paradoxical Activation of Oncogenic Signaling as a Cancer Treatment Strategy

Cancer homeostasis depends on a balance between activated oncogenic pathways driving tumorigenesis and engagement of stress response programs that counteract the inherent toxicity of such aberrant signaling. Although inhibition of oncogenic signaling pathways has been explored extensively, there is increasing evidence that overactivation of the same pathways can also disrupt cancer homeostasis and cause lethality. We show here that inhibition of protein phosphatase 2A (PP2A) hyperactivates multiple oncogenic pathways and engages stress responses in colon cancer cells. Genetic and compound screens identify combined inhibition of PP2A and WEE1 as synergistic in multiple cancer models by collapsing DNA replication and triggering premature mitosis followed by cell death. This combination also suppressed the growth of patient-derived tumors in vivo. Remarkably, acquired resistance to this drug combination suppressed the ability of colon cancer cells to form tumors in vivo. Our data suggest that paradoxical activation of oncogenic signaling can result in tumor-suppressive resistance. Significance: A therapy consisting of deliberate hyperactivation of oncogenic signaling combined with perturbation of the stress responses that result from this is very effective in animal models of colon cancer. Resistance to this therapy is associated with loss of oncogenic signaling and reduced oncogenic capacity, indicative of tumor-suppressive drug resistance.

Interaction between MED12 and ΔNp63 activates basal identity in pancreatic ductal adenocarcinoma

The presence of basal lineage characteristics signifies hyperaggressive human adenocarcinomas of the breast, bladder and pancreas. However, the biochemical mechanisms that maintain this aberrant cell state are poorly understood. Here we performed marker-based genetic screens in search of factors needed to maintain basal identity in pancreatic ductal adenocarcinoma (PDAC). This approach revealed MED12 as a powerful regulator of the basal cell state in this disease. Using biochemical reconstitution and epigenomics, we show that MED12 carries out this function by bridging the transcription factor ΔNp63, a known master regulator of the basal lineage, with the Mediator complex to activate lineage-specific enhancer elements. Consistent with this finding, the growth of basal-like PDAC is hypersensitive to MED12 loss when compared to PDAC cells lacking basal characteristics. Taken together, our genetic screens have revealed a biochemical interaction that sustains basal identity in human cancer, which could serve as a target for tumor lineage-directed therapeutics.

DeepGSEA: explainable deep gene set enrichment analysis for single-cell transcriptomic data

MOTIVATION

Gene set enrichment (GSE) analysis allows for an interpretation of gene expression through pre-defined gene set databases and is a critical step in understanding different phenotypes. With the rapid development of single-cell RNA sequencing (scRNA-seq) technology, GSE analysis can be performed on fine-grained gene expression data to gain a nuanced understanding of phenotypes of interest. However, with the cellular heterogeneity in single-cell gene profiles, current statistical GSE analysis methods sometimes fail to identify enriched gene sets. Meanwhile, deep learning has gained traction in applications like clustering and trajectory inference in single-cell studies due to its prowess in capturing complex data patterns. However, its use in GSE analysis remains limited, due to interpretability challenges.

RESULTS

In this paper, we present DeepGSEA, an explainable deep gene set enrichment analysis approach which leverages the expressiveness of interpretable, prototype-based neural networks to provide an in-depth analysis of GSE. DeepGSEA learns the ability to capture GSE information through our designed classification tasks, and significance tests can be performed on each gene set, enabling the identification of enriched sets. The underlying distribution of a gene set learned by DeepGSEA can be explicitly visualized using the encoded cell and cellular prototype embeddings. We demonstrate the performance of DeepGSEA over commonly used GSE analysis methods by examining their sensitivity and specificity with four simulation studies. In addition, we test our model on three real scRNA-seq datasets and illustrate the interpretability of DeepGSEA by showing how its results can be explained.

AVAILABILITY AND IMPLEMENTATION

https://github.com/Teddy-XiongGZ/DeepGSEA.

Cell type mapping of inflammatory muscle diseases highlights selective myofiber vulnerability in inclusion body myositis

Inclusion body myositis (IBM) is the most prevalent inflammatory muscle disease in older adults with no effective therapy available. In contrast to other inflammatory myopathies such as subacute, immune-mediated necrotizing myopathy (IMNM), IBM follows a chronic disease course with both inflammatory and degenerative features of pathology. Moreover, causal factors and molecular drivers of IBM progression are largely unknown. Therefore, we paired single-nucleus RNA sequencing with spatial transcriptomics from patient muscle biopsies to map cell-type-specific drivers underlying IBM pathogenesis compared with IMNM muscles and noninflammatory skeletal muscle samples. In IBM muscles, we observed a selective loss of type 2 myonuclei paralleled by increased levels of cytotoxic T and conventional type 1 dendritic cells. IBM myofibers were characterized by either upregulation of cell stress markers featuring GADD45A and NORAD or protein degradation markers including RNF7 associated with p62 aggregates. GADD45A upregulation was preferentially seen in type 2A myofibers associated with severe tissue inflammation. We also noted IBM-specific upregulation of ACHE encoding acetylcholinesterase, which can be regulated by NORAD activity and result in functional denervation of myofibers. Our results provide promising insights into possible mechanisms of myofiber degeneration in IBM and suggest a selective type 2 fiber vulnerability linked to genomic stress and denervation pathways.

A deep learning model of tumor cell architecture elucidates response and resistance to CDK4/6 inhibitors

Cyclin-dependent kinase 4 and 6 inhibitors (CDK4/6is) have revolutionized breast cancer therapy. However, <50% of patients have an objective response, and nearly all patients develop resistance during therapy. To elucidate the underlying mechanisms, we constructed an interpretable deep learning model of the response to palbociclib, a CDK4/6i, based on a reference map of multiprotein assemblies in cancer. The model identifies eight core assemblies that integrate rare and common alterations across 90 genes to stratify palbociclib-sensitive versus palbociclib-resistant cell lines. Predictions translate to patients and patient-derived xenografts, whereas single-gene biomarkers do not. Most predictive assemblies can be shown by CRISPR-Cas9 genetic disruption to regulate the CDK4/6i response. Validated assemblies relate to cell-cycle control, growth factor signaling and a histone regulatory complex that we show promotes S-phase entry through the activation of the histone modifiers KAT6A and TBL1XR1 and the transcription factor RUNX1. This study enables an integrated assessment of how a tumor's genetic profile modulates CDK4/6i resistance.

Prediction of DNA methylation-based tumor types from histopathology in central nervous system tumors with deep learning

Precision in the diagnosis of diverse central nervous system (CNS) tumor types is crucial for optimal treatment. DNA methylation profiles, which capture the methylation status of thousands of individual CpG sites, are state-of-the-art data-driven means to enhance diagnostic accuracy but are also time consuming and not widely available. Here, to address these limitations, we developed Deep lEarning from histoPathoLOgy and methYlation (DEPLOY), a deep learning model that classifies CNS tumors to ten major categories from histopathology. DEPLOY integrates three distinct components: the first classifies CNS tumors directly from slide images ('direct model'), the second initially generates predictions for DNA methylation beta values, which are subsequently used for tumor classification ('indirect model'), and the third classifies tumor types directly from routinely available patient demographics. First, we find that DEPLOY accurately predicts beta values from histopathology images. Second, using a ten-class model trained on an internal dataset of 1,796 patients, we predict the tumor categories in three independent external test datasets including 2,156 patients, achieving an overall accuracy of 95% and balanced accuracy of 91% on samples that are predicted with high confidence. These results showcase the potential future use of DEPLOY to assist pathologists in diagnosing CNS tumors within a clinically relevant short time frame.

Aging limits stemness and tumorigenesis in the lung by reprogramming iron homeostasis

Aging is associated with a decline in the number and fitness of adult stem cells 1-4 . Aging-associated loss of stemness is posited to suppress tumorigenesis 5,6 , but this hypothesis has not been tested in vivo . Here, using physiologically aged autochthonous genetically engineered mouse models and primary cells 7,8 , we demonstrate aging suppresses lung cancer initiation and progression by degrading stemness of the alveolar cell of origin. This phenotype is underpinned by aging-associated induction of the transcription factor NUPR1 and its downstream target lipocalin-2 in the cell of origin in mice and humans, leading to a functional iron insufficiency in the aged cells. Genetic inactivation of the NUPR1-lipocalin-2 axis or iron supplementation rescue stemness and promote tumorigenic potential of aged alveolar cells. Conversely, targeting the NUPR1- lipocalin-2 axis is detrimental to young alveolar cells via induction of ferroptosis. We find that aging-associated DNA hypomethylation at specific enhancer sites associates with elevated NUPR1 expression, which is recapitulated in young alveolar cells by inhibition of DNA methylation. We uncover that aging drives a functional iron insufficiency, which leads to loss of stemness and tumorigenesis, but promotes resistance to ferroptosis. These findings have significant implications for the therapeutic modulation of cellular iron homeostasis in regenerative medicine and in cancer prevention. Furthermore, our findings are consistent with a model whereby most human cancers initiate in young individuals, revealing a critical window for such cancer prevention efforts.

Integrating patients in time series clinical transcriptomics data

MOTIVATION

Analysis of time series transcriptomics data from clinical trials is challenging. Such studies usually profile very few time points from several individuals with varying response patterns and dynamics. Current methods for these datasets are mainly based on linear, global orderings using visit times which do not account for the varying response rates and subgroups within a patient cohort.

RESULTS

We developed a new method that utilizes multi-commodity flow algorithms for trajectory inference in large scale clinical studies. Recovered trajectories satisfy individual-based timing restrictions while integrating data from multiple patients. Testing the method on multiple drug datasets demonstrated an improved performance compared to prior approaches suggested for this task, while identifying novel disease subtypes that correspond to heterogeneous patient response patterns.

AVAILABILITY AND IMPLEMENTATION

The source code and instructions to download the data have been deposited on GitHub at https://github.com/euxhenh/Truffle.

Single-cell morphodynamical trajectories enable prediction of gene expression accompanying cell state change

Extracellular signals induce changes to molecular programs that modulate multiple cellular phenotypes, including proliferation, motility, and differentiation status. The connection between dynamically adapting phenotypic states and the molecular programs that define them is not well understood. Here we develop data-driven models of single-cell phenotypic responses to extracellular stimuli by linking gene transcription levels to "morphodynamics" - changes in cell morphology and motility observable in time-lapse image data. We adopt a dynamics-first view of cell state by grouping single-cell trajectories into states with shared morphodynamic responses. The single-cell trajectories enable development of a first-of-its-kind computational approach to map live-cell dynamics to snapshot gene transcript levels, which we term MMIST, Molecular and Morphodynamics-Integrated Single-cell Trajectories. The key conceptual advance of MMIST is that cell behavior can be quantified based on dynamically defined states and that extracellular signals alter the overall distribution of cell states by altering rates of switching between states. We find a cell state landscape that is bound by epithelial and mesenchymal endpoints, with distinct sequences of epithelial to mesenchymal transition (EMT) and mesenchymal to epithelial transition (MET) intermediates. The analysis yields predictions for gene expression changes consistent with curated EMT gene sets and provides a prediction of thousands of RNA transcripts through extracellular signal-induced EMT and MET with near-continuous time resolution. The MMIST framework leverages true single-cell dynamical behavior to generate molecular-level omics inferences and is broadly applicable to other biological domains, time-lapse imaging approaches and molecular snapshot data.

Single-cell-resolved interspecies comparison shows a shared inflammatory axis and a dominant neutrophil-endothelial program in severe COVID-19

A key issue for research on COVID-19 pathogenesis is the lack of biopsies from patients and of samples at the onset of infection. To overcome these hurdles, hamsters were shown to be useful models for studying this disease. Here, we further leverage the model to molecularly survey the disease progression from time-resolved single-cell RNA sequencing data collected from healthy and severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)-infected Syrian and Roborovski hamster lungs. We compare our data to human COVID-19 studies, including bronchoalveolar lavage, nasal swab, and postmortem lung tissue, and identify a shared axis of inflammation dominated by macrophages, neutrophils, and endothelial cells, which we show to be transient in Syrian and terminal in Roborovski hamsters. Our data suggest that, following SARS-CoV-2 infection, commitment to a type 1- or type 3-biased immunity determines moderate versus severe COVID-19 outcomes, respectively.

An epigenetically distinct HSC subset supports thymic reconstitution

Hematopoietic stem cells (HSCs) with multilineage potential are critical for effective T cell reconstitution and restoration of the adaptive immune system after allogeneic Hematopoietic Cell Transplantation (allo-HCT). The Kit lo subset of HSCs is enriched for multipotential precursors, 1, 2 but their T-cell lineage potential has not been well-characterized. We therefore studied the thymic reconstituting and T-cell potential of Kit lo HSCs. Using a preclinical allo-HCT model, we demonstrate that Kit lo HSCs support better thymic recovery, and T-cell reconstitution resulting in improved T cell responses to infection post-HCT. Furthermore, Kit lo HSCs with augmented BM lymphopoiesis mitigate age-associated thymic alterations, thus enhancing T-cell recovery in middle-aged hosts. We find the frequency of the Kit lo subset declines with age, providing one explanation for the reduced frequency of T-competent HSCs and reduced T-lymphopoietic potential in BM precursors of aged mice. 3, 4, 5 Chromatin profiling revealed that Kit lo HSCs exhibit higher activity of lymphoid-specifying transcription factors (TFs), including Zbtb1 . Deletion of Zbtb1 in Kit lo HSCs diminished their T-cell potential, while reinstating Zbtb1 in megakaryocytic-biased Kit hi HSCs rescued T-cell potential, in vitro and in vivo . Finally, we discover an analogous Kit lo HSC subset with enhanced lymphoid potential in human bone marrow. Our results demonstrate that Kit lo HSCs with enhanced lymphoid potential have a distinct underlying epigenetic program.

GOAT: efficient and robust identification of gene set enrichment

Gene set enrichment analysis is foundational to the interpretation of high throughput biology. Identifying enriched Gene Ontology (GO) terms or disease-associated gene sets within a list of gene effect sizes that represent experimental outcomes is an everyday task in life science that crucially depends on robust and sensitive statistical tools. We here present GOAT, a parameter-free algorithm for gene set enrichment analysis of preranked gene lists. The algorithm can precompute null distributions from standardized gene scores, enabling enrichment testing of the GO database in one second. Validations using synthetic data show that estimated gene set p-values are well calibrated under the null hypothesis and invariant to gene list length and gene set size. Application to various real-world proteomics and gene expression studies demonstrates that GOAT identifies more significant GO terms as compared to current methods. GOAT is freely available as an R package and user-friendly online tool for gene set enrichment analyses that includes interactive data visualizations: https://ftwkoopmans.github.io/goat .

node2vec2rank: Large Scale and Stable Graph Differential Analysis via Multi-Layer Node Embeddings and Ranking

Computational methods in biology can infer large molecular interaction networks from multiple data sources and at different resolutions, creating unprecedented opportunities to explore the mechanisms driving complex biological phenomena. Networks can be built to represent distinct conditions and compared to uncover graph-level differences-such as when comparing patterns of gene-gene interactions that change between biological states. Given the importance of the graph comparison problem, there is a clear and growing need for robust and scalable methods that can identify meaningful differences. We introduce node2vec2rank (n2v2r), a method for graph differential analysis that ranks nodes according to the disparities of their representations in joint latent embedding spaces. Improving upon previous bag-of-features approaches, we take advantage of recent advances in machine learning and statistics to compare graphs in higher-order structures and in a data-driven manner. Formulated as a multi-layer spectral embedding algorithm, n2v2r is computationally efficient, incorporates stability as a key feature, and can provably identify the correct ranking of differences between graphs in an overall procedure that adheres to veridical data science principles. By better adapting to the data, node2vec2rank clearly outperformed the commonly used node degree in finding complex differences in simulated data. In the real-world applications of breast cancer subtype characterization, analysis of cell cycle in single-cell data, and searching for sex differences in lung adenocarcinoma, node2vec2rank found meaningful biological differences enabling the hypothesis generation for therapeutic candidates. Software and analysis pipelines implementing n2v2r and used for the analyses presented here are publicly available.

Microbiota succession influences nematode physiology in a beetle microcosm ecosystem

Unravelling the multifaceted and bidirectional interactions between microbiota and host physiology represents a major scientific challenge. Here, we utilise the nematode model, Pristionchus pacificus, coupled to a laboratory-simulated decay process of its insect host, to mimic natural microbiota succession and investigate associated tripartite interactions. Metagenomics reveal that during initial decay stages, the population of vitamin B-producing bacteria diminishes, potentially due to a preferential selection by nematodes. As decay progresses to nutrient-depleted stages, bacteria with smaller genomes producing less nutrients become more prevalent. Lipid utilisation and dauer formation, representing key nematode survival strategies, are influenced by microbiota changes. Additionally, horizontally acquired cellulases extend the nematodes' reproductive phase due to more efficient foraging. Lastly, the expressions of Pristionchus species-specific genes are more responsive to natural microbiota compared to conserved genes, suggesting their importance in the organisms' adaptation to its ecological niche. In summary, we show the importance of microbial successions and their reciprocal interaction with nematodes for insect decay in semi-artificial ecosystems.

Integrative analysis of single-cell RNA-seq and gut microbiome metabarcoding data elucidates macrophage dysfunction in mice with DSS-induced ulcerative colitis

Ulcerative colitis (UC) is a significant inflammatory bowel disease caused by an abnormal immune response to gut microbes. However, there are still gaps in our understanding of how immune and metabolic changes specifically contribute to this disease. Our research aims to address this gap by examining mouse colons after inducing ulcerative colitis-like symptoms. Employing single-cell RNA-seq and 16 s rRNA amplicon sequencing to analyze distinct cell clusters and microbiomes in the mouse colon at different time points after induction with dextran sodium sulfate. We observe a significant reduction in epithelial populations during acute colitis, indicating tissue damage, with a partial recovery observed in chronic inflammation. Analyses of cell-cell interactions demonstrate shifts in networking patterns among different cell types during disease progression. Notably, macrophage phenotypes exhibit diversity, with a pronounced polarization towards the pro-inflammatory M1 phenotype in chronic conditions, suggesting the role of macrophage heterogeneity in disease severity. Increased expression of Nampt and NOX2 complex subunits in chronic UC macrophages contributes to the inflammatory processes. The chronic UC microbiome exhibits reduced taxonomic diversity compared to healthy conditions and acute UC. The study also highlights the role of T cell differentiation in the context of dysbiosis and its implications in colitis progression, emphasizing the need for targeted interventions to modulate the inflammatory response and immune balance in colitis.

An interactive atlas of genomic, proteomic, and metabolomic biomarkers promotes the potential of proteins to predict complex diseases

Multiomics analyses have identified multiple potential biomarkers of the incidence and prevalence of complex diseases. However, it is not known which type of biomarker is optimal for clinical purposes. Here, we make a systematic comparison of 90 million genetic variants, 1453 proteins, and 325 metabolites from 500,000 individuals with complex diseases from the UK Biobank. A machine learning pipeline consisting of data cleaning, data imputation, feature selection, and model training using cross-validation and comparison of the results on holdout test sets showed that proteins were most predictive, followed by metabolites, and genetic variants. Only five proteins per disease resulted in median (min-max) areas under the receiver operating characteristic curves for incidence of 0.79 (0.65-0.86) and 0.84 (0.70-0.91) for prevalence. In summary, our work suggests the potential of predicting complex diseases based on a limited number of proteins. We provide an interactive atlas (macd.shinyapps.io/ShinyApp/) to find genomic, proteomic, or metabolomic biomarkers for different complex diseases.

Spatial transition tensor of single cells

Spatial transcriptomics and messenger RNA splicing encode extensive spatiotemporal information for cell states and transitions. The current lineage-inference methods either lack spatial dynamics for state transition or cannot capture different dynamics associated with multiple cell states and transition paths. Here we present spatial transition tensor (STT), a method that uses messenger RNA splicing and spatial transcriptomes through a multiscale dynamical model to characterize multistability in space. By learning a four-dimensional transition tensor and spatial-constrained random walk, STT reconstructs cell-state-specific dynamics and spatial state transitions via both short-time local tensor streamlines between cells and long-time transition paths among attractors. Benchmarking and applications of STT on several transcriptome datasets via multiple technologies on epithelial-mesenchymal transitions, blood development, spatially resolved mouse brain and chicken heart development, indicate STT's capability in recovering cell-state-specific dynamics and their associated genes not seen using existing methods. Overall, STT provides a consistent multiscale description of single-cell transcriptome data across multiple spatiotemporal scales.

Ablation of Vitamin D Signaling in Cardiomyocytes Leads to Functional Impairment and Stimulation of Pro-Inflammatory and Pro-Fibrotic Gene Regulatory Networks in a Left Ventricular Hypertrophy Model in Mice

The association between vitamin D deficiency and cardiovascular disease remains a controversial issue. This study aimed to further elucidate the role of vitamin D signaling in the development of left ventricular (LV) hypertrophy and dysfunction. To ablate the vitamin D receptor (VDR) specifically in cardiomyocytes, VDRfl/fl mice were crossed with Mlcv2-Cre mice. To induce LV hypertrophy experimentally by increasing cardiac afterload, transverse aortic constriction (TAC) was employed. Sham or TAC surgery was performed in 4-month-old, male, wild-type, VDRfl/fl, Mlcv2-Cre, and cardiomyocyte-specific VDR knockout (VDRCM-KO) mice. As expected, TAC induced profound LV hypertrophy and dysfunction, evidenced by echocardiography, aortic and cardiac catheterization, cardiac histology, and LV expression profiling 4 weeks post-surgery. Sham-operated mice showed no differences between genotypes. However, TAC VDRCM-KO mice, while having comparable cardiomyocyte size and LV fibrosis to TAC VDRfl/fl controls, exhibited reduced fractional shortening and ejection fraction as measured by echocardiography. Spatial transcriptomics of heart cryosections revealed more pronounced pro-inflammatory and pro-fibrotic gene regulatory networks in the stressed cardiac tissue niches of TAC VDRCM-KO compared to VDRfl/fl mice. Hence, our study supports the notion that vitamin D signaling in cardiomyocytes plays a protective role in the stressed heart.

A multi-view graph contrastive learning framework for deciphering spatially resolved transcriptomics data

Spatially resolved transcriptomics data are being used in a revolutionary way to decipher the spatial pattern of gene expression and the spatial architecture of cell types. Much work has been done to exploit the genomic spatial architectures of cells. Such work is based on the common assumption that gene expression profiles of spatially adjacent spots are more similar than those of more distant spots. However, related work might not consider the nonlocal spatial co-expression dependency, which can better characterize the tissue architectures. Therefore, we propose MuCoST, a Multi-view graph Contrastive learning framework for deciphering complex Spatially resolved Transcriptomic architectures with dual scale structural dependency. To achieve this, we employ spot dependency augmentation by fusing gene expression correlation and spatial location proximity, thereby enabling MuCoST to model both nonlocal spatial co-expression dependency and spatially adjacent dependency. We benchmark MuCoST on four datasets, and we compare it with other state-of-the-art spatial domain identification methods. We demonstrate that MuCoST achieves the highest accuracy on spatial domain identification from various datasets. In particular, MuCoST accurately deciphers subtle biological textures and elaborates the variation of spatially functional patterns.

Acquired resistance to immunotherapy and chemoradiation in MYC amplified head and neck cancer

The proto-oncogene MYC encodes a nuclear transcription factor that has an important role in a variety of cellular processes, such as cell cycle progression, proliferation, metabolism, adhesion, apoptosis, and therapeutic resistance. MYC amplification is consistently observed in aggressive forms of several solid malignancies and correlates with poor prognosis and distant metastases. While the tumorigenic effects of MYC in patients with head and neck squamous cell carcinoma (HNSCC) are well known, the molecular mechanisms by which the amplification of this gene may confer treatment resistance, especially to immune checkpoint inhibitors, remains under-investigated. Here we present a unique case of a patient with recurrent/metastatic (R/M) HNSCC who, despite initial response to nivolumab-based treatment, developed rapidly progressive metastatic disease after the acquisition of MYC amplification. We conducted comparative transcriptomic analysis of this patient's tumor at baseline and upon progression to interrogate potential molecular processes through which MYC may confer resistance to immunotherapy and/or chemoradiation and used TCGA-HNSC dataset and an institutional cohort to further explore clinicopathologic features and key molecular networks associated with MYC amplification in HNSCC. This study highlights MYC amplification as a potential mechanism of immune checkpoint inhibitor resistance and suggest its use as a predictive biomarker and potential therapeutic target in R/M HNSCC.

Transcriptomic and spatial dissection of human ex vivo right atrial tissue reveals proinflammatory microvascular changes in ischemic heart disease

Cardiovascular disease plays a central role in the electrical and structural remodeling of the right atrium, predisposing to arrhythmias, heart failure, and sudden death. Here, we dissect with single-nuclei RNA sequencing (snRNA-seq) and spatial transcriptomics the gene expression changes in the human ex vivo right atrial tissue and pericardial fluid in ischemic heart disease, myocardial infarction, and ischemic and non-ischemic heart failure using asymptomatic patients with valvular disease who undergo preventive surgery as the control group. We reveal substantial differences in disease-associated gene expression in all cell types, collectively suggesting inflammatory microvascular dysfunction and changes in the right atrial tissue composition as the valvular and vascular diseases progress into heart failure. The data collectively suggest that investigation of human cardiovascular disease should expand to all functionally important parts of the heart, which may help us to identify mechanisms promoting more severe types of the disease.

Transcriptomic signatures of classical monocytes reveal pro-inflammatory modules and heterogeneity in polyarticular juvenile idiopathic arthritis

Introduction

Polyarticular juvenile idiopathic arthritis (pJIA) is a childhood-onset autoimmune disease. Immune cells contribute to persistent inflammation observed in pJIA. Despite the crucial role of monocytes in arthritis, the precise involvement of classical monocytes in the pathogenesis of pJIA remains uncertain. Here, we aimed to uncover the transcriptomic patterns of classical monocytes in pJIA, focusing on their involvement in disease mechanism and heterogeneity.

Methods

A total of 17 healthy subjects and 18 premenopausal women with pJIA according to ILAR criteria were included. Classical monocytes were isolated, and RNA sequencing was performed. Differential expression analysis was used to compare pJIA patients and healthy control group. Differentially expressed genes (DEGs) were identified, and gene set enrichment analysis (GSEA) was performed. Using unsupervised learning approach, patients were clustered in two groups based on their similarities at transcriptomic level. Subsequently, these clusters underwent a comparative analysis to reveal differences at the transcriptomic level.

Results

We identified 440 DEGs in pJIA patients of which 360 were upregulated and 80 downregulated. GSEA highlighted TNF-α and IFN-γ response. Importantly, this analysis not only detected genes targeted by pJIA therapy but also identified new modulators of immuno-inflammation. PLAUR, IL1B, IL6, CDKN1A, PIM1, and ICAM1 were pointed as drivers of chronic hyperinflammation. Unsupervised learning approach revealed two clusters within pJIA, each exhibiting varying inflammation levels.

Conclusion

These findings indicate the pivotal role of immuno-inflammation driven by classical monocytes in pJIA and reveals the existence of two subclusters within pJIA, regardless the positivity of rheumatoid factor and anti-CCP, paving the way to precision medicine.

Optimal CXCR5 Expression during Tfh Maturation Involves the Bhlhe40-Pou2af1 Axis

The pair of transcription factors Bcl6-Blimp1 is well-known for follicular T helper (Tfh) cell fate determination, however, the mechanism(s) for Bcl6-independent regulation of CXCR5 during Tfh migration into germinal center (GC) is still unclear. In this study, we uncovered another pair of transcription factors, Bhlhe40-Pou2af1, that regulates CXCR5 expression. Pou2af1 was specifically expressed in Tfh cells whereas Bhlhe40 expression was found high in non-Tfh cells. Pou2af1 promoted Tfh formation and migration into GC by upregulating CXCR5 but not Bcl6, while Bhlhe40 repressed this process by inhibiting Pou2af1 expression. RNA-Seq analysis of antigen-specific Tfh cells generated in vivo confirmed the role of Bhlhe40-Pou2af1 axis in regulating optimal CXCR5 expression. Thus, the regulation of CXCR5 expression and migration of Tfh cells into GC involves a transcriptional regulatory circuit consisting of Bhlhe40 and Pou2af1, which operates independent of the Bcl6-Blimp1 circuit that determines the Tfh cell fate.

Reference-free inferring of transcriptomic events in cancer cells on single-cell data

BACKGROUND

Cancerous cells' identity is determined via a mixture of multiple factors such as genomic variations, epigenetics, and the regulatory variations that are involved in transcription. The differences in transcriptome expression as well as abnormal structures in peptides determine phenotypical differences. Thus, bulk RNA-seq and more recent single-cell RNA-seq data (scRNA-seq) are important to identify pathogenic differences. In this case, we rely on k-mer decomposition of sequences to identify pathogenic variations in detail which does not need a reference, so it outperforms more traditional Next-Generation Sequencing (NGS) analysis techniques depending on the alignment of the sequences to a reference.

RESULTS

Via our alignment-free analysis, over esophageal and glioblastoma cancer patients, high-frequency variations over multiple different locations (repeats, intergenic regions, exons, introns) as well as multiple different forms (fusion, polyadenylation, splicing, etc.) could be discovered. Additionally, we have analyzed the importance of less-focused events systematically in a classic transcriptome analysis pipeline where these events are considered as indicators for tumor prognosis, tumor prediction, tumor neoantigen inference, as well as their connection with respect to the immune microenvironment.

CONCLUSIONS

Our results suggest that esophageal cancer (ESCA) and glioblastoma processes can be explained via pathogenic microbial RNA, repeated sequences, novel splicing variants, and long intergenic non-coding RNAs (lincRNAs). We expect our application of reference-free process and analysis to be helpful in tumor and normal samples differential scRNA-seq analysis, which in turn offers a more comprehensive scheme for major cancer-associated events.

Supervised latent factor modeling isolates cell-type-specific transcriptomic modules that underlie Alzheimer's disease progression

Late onset Alzheimer's disease (AD) is a progressive neurodegenerative disease, with brain changes beginning years before symptoms surface. AD is characterized by neuronal loss, the classic feature of the disease that underlies brain atrophy. However, GWAS reports and recent single-nucleus RNA sequencing (snRNA-seq) efforts have highlighted that glial cells, particularly microglia, claim a central role in AD pathophysiology. Here, we tailor pattern-learning algorithms to explore distinct gene programs by integrating the entire transcriptome, yielding distributed AD-predictive modules within the brain's major cell-types. We show that these learned modules are biologically meaningful through the identification of new and relevant enriched signaling cascades. The predictive nature of our modules, especially in microglia, allows us to infer each subject's progression along a disease pseudo-trajectory, confirmed by post-mortem pathological brain tissue markers. Additionally, we quantify the interplay between pairs of cell-type modules in the AD brain, and localized known AD risk genes to enriched module gene programs. Our collective findings advocate for a transition from cell-type-specificity to gene modules specificity to unlock the potential of unique gene programs, recasting the roles of recently reported genome-wide AD risk loci.

Novel cell states arise in embryonic cells devoid of key reprogramming factors

The capacity for embryonic cells to differentiate relies on a large-scale reprogramming of the oocyte and sperm nucleus into a transient totipotent state. In zebrafish, this reprogramming step is achieved by the pioneer factors Nanog, Pou5f3, and Sox19b (NPS). Yet, it remains unclear whether cells lacking this reprogramming step are directed towards wild type states or towards novel developmental canals in the Waddington landscape of embryonic development. Here we investigate the developmental fate of embryonic cells mutant for NPS by analyzing their single-cell gene expression profiles. We find that cells lacking the first developmental reprogramming steps can acquire distinct cell states. These states are manifested by gene expression modules that result from a failure of nuclear reprogramming, the persistence of the maternal program, and the activation of somatic compensatory programs. As a result, most mutant cells follow new developmental canals and acquire new mixed cell states in development. In contrast, a group of mutant cells acquire primordial germ cell-like states, suggesting that NPS-dependent reprogramming is dispensable for these cell states. Together, these results demonstrate that developmental reprogramming after fertilization is required to differentiate most canonical developmental programs, and loss of the transient totipotent state canalizes embryonic cells into new developmental states in vivo.

drGAT: Attention-Guided Gene Assessment of Drug Response Utilizing a Drug-Cell-Gene Heterogeneous Network

Drug development is a lengthy process with a high failure rate. Increasingly, machine learning is utilized to facilitate the drug development processes. These models aim to enhance our understanding of drug characteristics, including their activity in biological contexts. However, a major challenge in drug response (DR) prediction is model interpretability as it aids in the validation of findings. This is important in biomedicine, where models need to be understandable in comparison with established knowledge of drug interactions with proteins. drGAT, a graph deep learning model, leverages a heterogeneous graph composed of relationships between proteins, cell lines, and drugs. drGAT is designed with two objectives: DR prediction as a binary sensitivity prediction and elucidation of drug mechanism from attention coefficients. drGAT has demonstrated superior performance over existing models, achieving 78% accuracy (and precision), and 76% F1 score for 269 DNA-damaging compounds of the NCI60 drug response dataset. To assess the model's interpretability, we conducted a review of drug-gene co-occurrences in Pubmed abstracts in comparison to the top 5 genes with the highest attention coefficients for each drug. We also examined whether known relationships were retained in the model by inspecting the neighborhoods of topoisomerase-related drugs. For example, our model retained TOP1 as a highly weighted predictive feature for irinotecan and topotecan, in addition to other genes that could potentially be regulators of the drugs. Our method can be used to accurately predict sensitivity to drugs and may be useful in the identification of biomarkers relating to the treatment of cancer patients.

Enhancing pathological complete response prediction in breast cancer: the role of dynamic characterization of DCE-MRI and its association with tumor heterogeneity

BACKGROUND

Early prediction of pathological complete response (pCR) is important for deciding appropriate treatment strategies for patients. In this study, we aimed to quantify the dynamic characteristics of dynamic contrast-enhanced magnetic resonance images (DCE-MRI) and investigate its value to improve pCR prediction as well as its association with tumor heterogeneity in breast cancer patients.

METHODS

The DCE-MRI, clinicopathologic record, and full transcriptomic data of 785 breast cancer patients receiving neoadjuvant chemotherapy were retrospectively included from a public dataset. Dynamic features of DCE-MRI were computed from extracted phase-varying radiomic feature series using 22 CAnonical Time-sereis CHaracteristics. Dynamic model and radiomic model were developed by logistic regression using dynamic features and traditional radiomic features respectively. Various combined models with clinical factors were also developed to find the optimal combination and the significance of each components was evaluated. All the models were evaluated in independent test set in terms of area under receiver operating characteristic curve (AUC). To explore the potential underlying biological mechanisms, radiogenomic analysis was implemented on patient subgroups stratified by dynamic model to identify differentially expressed genes (DEGs) and enriched pathways.

RESULTS

A 10-feature dynamic model and a 4-feature radiomic model were developed (AUC = 0.688, 95%CI: 0.635-0.741 and AUC = 0.650, 95%CI: 0.595-0.705) and tested (AUC = 0.686, 95%CI: 0.594-0.778 and AUC = 0.626, 95%CI: 0.529-0.722), with the dynamic model showing slightly higher AUC (train p = 0.181, test p = 0.222). The combined model of clinical, radiomic, and dynamic achieved the highest AUC in pCR prediction (train: 0.769, 95%CI: 0.722-0.816 and test: 0.762, 95%CI: 0.679-0.845). Compared with clinical-radiomic combined model (train AUC = 0.716, 95%CI: 0.665-0.767 and test AUC = 0.695, 95%CI: 0.656-0.714), adding the dynamic component brought significant improvement in model performance (train p < 0.001 and test p = 0.005). Radiogenomic analysis identified 297 DEGs, including CXCL9, CCL18, and HLA-DPB1 which are known to be associated with breast cancer prognosis or angiogenesis. Gene set enrichment analysis further revealed enrichment of gene ontology terms and pathways related to immune system.

CONCLUSION

Dynamic characteristics of DCE-MRI were quantified and used to develop dynamic model for improving pCR prediction in breast cancer patients. The dynamic model was associated with tumor heterogeniety in prognostic-related gene expression and immune-related pathways.