Biology subjects

Benos, P. V.

Publications and source records attributed to Benos, P. V..

16 recordsLinked to original sources

bioRxiv2025

Alpha diversity analysis of hepatic transcriptome reveals novel pathways inalcohol-related hepatitis

Next generation sequencing can identify novel gene expression patterns in disease. Beyond differentially expressed genes analysis, we investigated the ability of within-population diversity (-diversity) of the transcriptome to reveal new biological information in alcohol-related liver disease (ALD), comparing Differential Shannon diversity (DSD) to transcriptome heterogeneity changes. RNA sequencing data from normal livers and patients with early silent ALD and severe AH were analyzed. -diversity indices and Percent Shannon Diversity of a gene, which refers to this genes contribution to total Shannon entropy were calculated. Ingenuity pathway analysis identified canonical pathways determined by differentially expressed genes (DEG) and DSD approaches. ALD significantly decreased hepatic transcriptome -diversity correlating with increased relative contribution of select genes. These changes were driven by lower abundance gene expression loss. DEG and DSD analyses showed overlapping genes and canonical pathways, but DSD also identified novel genes and pathways not highlighted by DEG. Importantly, DSD more effectively identified differences between preclinical ALD and AH severity stages. ALD decreases hepatic transcriptome heterogeneity, favoring pathways associated with organ damage or damage response. Preclinical and clinical ALD led to differential heterogeneity patterns that may provide new disease insights. DSD analysis identified enriched pathways missed by standard DEG analyses, potentially yielding novel insight into disease mechanisms and biomarkers. Graphical abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=53 SRC="FIGDIR/small/681263v1_ufig1.gif" ALT="Figure 1"> View larger version (16K): org.highwire.dtl.DTLVardef@d2d3a3org.highwire.dtl.DTLVardef@14496a8org.highwire.dtl.DTLVardef@1ff0dfeorg.highwire.dtl.DTLVardef@1efa8f3_HPS_FORMAT_FIGEXP M_FIG C_FIG

systems biology↗

bioRxiv2025

Frustratingly easy domain adaptation for cross-speciestranscription factor binding prediction

MotivationUnderstanding how DNA sequence encodes gene regulation remains a central challenge in genomics. While deep learning models can predict regulatory activity from sequence with high accuracy, their generalizability across species--and thus their ability to capture fundamental biological principles--remains limited. Cross-species prediction provides a powerful test of model robustness and offers a window into conserved regulatory logic, but effectively bridging species-specific genomic differences remains a major barrier. ResultsWe present MORALE, a novel and scalable domain adaptation framework that significantly advances cross-species prediction of transcription factor (TF) binding. By aligning statistical moments of sequence embeddings across species, MORALE enables deep learning models to learn species-invariant regulatory features without requiring adversarial training or complex architectures. Applied to multi-species TF ChIP-seq datasets, MORALE achieves state-of-the-art performance--outperforming both baseline and adversarial approaches across all TFs--while preserving model interpretability and recovering canonical motifs with greater precision. In the five-species transfer setting, MORALE not only improves human prediction accuracy beyond human-only training but also reveals regulatory features conserved across mammals. These results highlight the potential of simple yet powerful domain adaptation techniques to drive generalization and discovery in regulatory genomics. Crucially, MORALE is architecture-agnostic and can be seamlessly integrated into any embedding-based sequence model. AvailabilityCode is available at https://github.com/loudrxiv/frustrating.

genomics↗

bioRxiv2024

Altered AP-1, RUNX, and EGR chromatin dynamics drive fibrotic lung disease

Pulmonary fibrosis, including systemic sclerosis-associated interstitial lung disease (SSc-ILD), involves myofibroblasts and SPP1hi macrophages as drivers of fibrosis. Single-cell RNA sequencing has delineated fibroblast and macrophages transcriptomes, but limited insight into transcriptional control of profibrotic gene programs. To address this challenge, we analyzed multiomic snATAC/snRNA-seq on explanted SSc-ILD and donor control lungs. The neural network tool ChromBPNet inferred increased TF binding at single base pair resolution to profibrotic genes, including CTHRC1 and ADAM12, in fibroblasts and SPP1 and CCL18 in macrophages. The novel algorithm HALO confirmed AP-1, RUNX, and EGR TF activity controlling profibrotic gene programs and established TF-regulatory element-gene networks. This TF action atlas provides comprehensive insights into the transcriptional regulation of fibroblasts and macrophages in healthy and fibrotic human lungs.

systems biology↗

bioRxiv2024

Cellular and transcriptional profiles of peripheral blood mononuclear cells pre-vaccination predict immune response to preventative MUC1 vaccine

A single arm trial (NCT007773097) and a double-blind, placebo controlled randomized trial (NCT02134925) were conducted in individuals with a history of advanced colonic adenoma to test the safety and immunogenicity of the MUC1 tumor antigen vaccine and its potential to prevent new adenomas. These were the first two trials of a non-viral cancer vaccine administered in the absence of cancer. The vaccine was safe and strongly immunogenic in 43% (NCT007773097) and 25% (NCT02134925) of participants. The lack of response in a significant number of participants suggested, for the first time, that even in a premalignant setting, the immune system may have already been exposed to some level of suppression previously reported only in cancer. Analysis of single-cell RNA-sequencing (scRNA-seq) data from banked pre-vaccination peripheral blood mononuclear cells (PBMCs) (16 immune responders and 16 non-responders) identified specific cell types, genes, and pathways of a productive vaccine response. Responders had a significantly higher percentage of CD4+ naive T cells pre-vaccination, but a significantly lower percentage of CD8+ T effector memory (TEM) cells and CD16+ monocytes. Differential gene expression (DGE) and transcription factor inference analysis showed a higher level of expression of T cell activation genes, such as Fos and Jun, in CD4+ naive T cells. Pathway analysis showed enriched signaling activity in responders. Furthermore, Bayesian network analyses suggested that these genes were mechanistically related to response. Our analyses identified several immune mechanisms and candidate biomarkers which can be further validated as predictors of immune responses to a preventative cancer vaccine that could facilitate selection of individuals likely to benefit from a vaccine or be used in further research to improve vaccine responses. One Sentence SummarySingle-cell RNA sequencing reveals distinctive cell types, enriched biological pathways, and candidate biomarkers pre-vaccination that predict immune responses to the preventative MUC1 cancer vaccine.

cancer biology↗

bioRxiv2024

LaGrACE: Estimating gene program dysregulation using latent gene regulatory network for biomedical discovery

Gene expression programs that establish and maintain specific cellular states are orchestrated through a regulatory network composed of transcription factors, cofactors, and chromatin regulators. Dysregulation of this network can lead to a broad range of diseases by altering gene programs. This article presents LaGrACE, a novel method designed to estimate dysregulation of gene programs utilizing omics data with clinical information. This approach facilitates grouping of samples exhibiting similar patterns of gene program dysregulation, thereby enhancing the discovery of underlying molecular mechanisms. We rigorously evaluated LaGrACEs performance using synthetic data, bulk RNA-seq clinical datasets (breast cancer, chronic obstructive pulmonary disease (COPD)), and single-cell RNA-seq drug perturbation datasets. Our findings demonstrate that LaGrACE is exceptionally robust in identifying biologically meaningful and prognostic molecular subtypes. Additionally, it effectively discerns drug-response signals at a single-cell resolution. Moreover, the COPD analysis revealed a new association between LEF1 and COPD molecular mechanisms and mortality. Collectively, these results underscore the utility of LaGrACE as a valuable tool for elucidating underlying mechanisms of diseases.

bioinformatics↗

bioRxiv2023

A 50-gene high-risk profile predictive of COVID-19 and Idiopathic Pulmonary Fibrosis mortality originates from a molecular imbalance in monocyte and T-cell subsets that reverses in survivors with post-COVID-19 Interstitial Lung Disease

BackgroundWe aim to study the source of circulating immune cells expressing a 50-gene signature predictive of COVID-19 and IPF mortality. MethodsWhole blood and Peripheral Blood Mononuclear cells (PBMC) were obtained from 231 subjects with COVID-19, post-COVID-19-ILD, IPF and controls. We measured the 50-gene signature (nCounter, Nanostring), interleukin 6 (IL6), interferon {gamma}-induced protein (IP10), secreted phosphoprotein 1 (SPP1) and transforming growth factor beta (TGF-{beta}) by Luminex. PCR was used to validate COVID-19 endotypes. For single-cell RNA sequencing (scRNA-seq) we used Chromium Controller (10X Genomics). For analysis we used the Scoring Algorithm of Molecular Subphenotypes (SAMS), Cell Ranger, Seurat, Propeller, Kaplan-Meier curves, CoxPH models, Two-way ANOVA, T-test, and Fishers exact. ResultsWe identified three genomic risk profiles based on the 50-gene signature, and a subset of seven genes, associated with low, intermediate, or high-risk of mortality in COVID-19 with significant differences in IL6, IP10, SPP1 and TGF{beta}-1. scRNA-seq identified Monocytic-Myeloid-Derived Suppressive cells (M-MDSCs) expressing CD14+HLA DRlowCD163+ and high levels of the 7-gene signature (7Gene-M-MDSC) in COVID-19. These cells were not observed in post-COVID-19-ILD or IPF. The 43-gene signature was mostly expressed in CD4 T and CD8 T cell subsets. Increased expression of the 43 gene signature was seen in T cell subsets from survivors with post-COVID-19-ILD. The expression of these genes remained low in IPF. ConclusionA 50-gene, high-risk profile in COVID-19 is characterized by a genomic imbalance in monocyte and T-cell subsets that reverses in survivors with post-COVID-19 Interstitial Lung Disease

genomics↗

bioRxiv2023

LEF1 isoforms regulate cellular senescence and aging

BackgroundThe study of aging and its mechanisms, such as cellular senescence, has provided valuable insights into age-related pathologies, thus contributing to their prevention and treatment. The current abundance of high throughput data combined with the surge of robust analysis algorithms has facilitated novel ways of identifying underlying pathways that may drive these pathologies. MethodsWith the focus on identifying key regulators of lung aging, we performed comparative analyses of transcriptional profiles of aged versus young human subjects and mice, focusing on the common age-related changes in the transcriptional regulation in lung macrophages, T cells, and B immune cells. Importantly, we validated our findings in cell culture assays and human lung samples. ResultsWe identified Lymphoid Enhancer Binding Factor 1 (LEF1) as an important age-associated regulator of gene expression in all three cell types across different tissues and species. Follow-up experiments showed that the differential expression of long and short LEF1 isoforms is a key regulatory mechanism of cellular senescence. Further examination of lung tissue from patients with Idiopathic Pulmonary Fibrosis (IPF), an age-related disease with strong ties to cellular senescence, we demonstrated a stark dysregulation of LEF1. ConclusionsCollectively, our results suggest that the LEF1 is a key factor of aging, and its differential regulation is associated with human and murine cellular senescence.

genomics↗

bioRxiv2023

The plasma degradome reflects later development of NASH fibrosis after liver transplant

Although liver transplantation (LT) is an effective therapy for cirrhosis, the risk of post-LT NASH is alarmingly high and is associated with accelerated progression to fibrosis/cirrhosis, cardiovascular disease, and decreased survival. Lack of risk stratification strategies hamper liver undergoes significant remodeling during inflammatory injury. During such remodeling, degraded peptide fragments (i.e., degradome) of the ECM and other proteins increase in plasma, making it a useful diagnostic/prognostic tool in chronic liver disease. To investigate whether inflammatory liver injury caused by post-LT NASH would yield a unique degradome profile, predictive of severe post-LT NASH fibrosis, we performed a retrospective analysis of 22 biobanked samples from the Starzl Transplantation Institute (12 with post-LT NASH after 5 years and 10 without). Total plasma peptides were isolated and analyzed by 1D-LC-MS/MS analysis using a Proxeon EASY-nLC 1000 UHPLC and nanoelectrospray ionization into an Orbitrap Elite mass spectrometer. Qualitative and quantitative peptide features data were developed from MSn datasets using PEAKS Studio X (v10). LC-MS/MS yielded [~]2700 identifiable peptide features based on the results from Peaks Studio analysis. Several peptides were significantly altered in patients that later developed fibrosis and heatmap analysis of the top 25 most significantly-changed peptides, most of which were ECM-derived, clustered the 2 patient groups well. Supervised modeling of the dataset indicated that a fraction of the total peptide signal ([~]15%) could explain the differences between the groups, indicating a strong potential for representative biomarker selection. A similar degradome profile was observed when the plasma degradome patterns were compared being obesity sensitive (C57Bl6/J) and insensitive (AJ) mouse strains. Both The plasma degradome profile of post-LT patients yields stark difference based on later development of post-LT NASH fibrosis. This approach could yield new "fingerprints" that can serve as minimally-invasive biomarkers of negative outcomes post-LT.

pathology↗

bioRxiv2023

An intrinsically interpretable neural network architecture for sequence to function learning

MotivationSequence-based deep learning approaches have been shown to predict a multitude of functional genomic readouts, including regions of open chromatin and RNA expression of genes. However, a major limitation of current methods is that model interpretation relies on computationally demanding post hoc analyses, and even then, one can often not explain the internal mechanics of highly parameterized models. Here, we introduce a deep learning architecture called tiSFM (totally interpretable sequence to function model). tiSFM improves upon the performance of standard multi-layer convolutional models while using fewer parameters. Additionally, while tiSFM is itself technically a multi-layer neural network, internal model parameters are intrinsically interpretable in terms of relevant sequence motifs. ResultsWe analyze published open chromatin measurements across hematopoietic lineage cell-types and demonstrate that tiSFM outperforms a state-of-the-art convolutional neural network model custom-tailored to this dataset. We also show that it correctly identifies context specific activities of transcription factors with known roles in hematopoietic differentiation, including Pax5 and Ebf1 for B-cells, and Rorc for innate lymphoid cells. tiSFMs model parameters have biologically meaningful interpretations, and we show the utility of our approach on a complex task of predicting the change in epigenetic state as a function of developmental transition. Availability and implementationThe source code, including scripts for the analysis of key findings, can be found at https://github.com/boooooogey/ATAConv, implemented in Python. Contactatb44@pitt.edu

bioinformatics↗

bioRxiv2022

Towards Hierarchical Causal Representation Learning for Nonstationary Multi-Omics Data

Though open chromatin may promote active transcription, gene expression responses may not be directly coordinated with changes in chromatin accessibility. Most existing methods for single-cell multi-omics data focus only on learning stationary and shared information among these modalities, overlooking modality-specific information delineating cellular states and dynamics resulting from causal relations among modalities. To account for this, the epigenome and transcriptome relationship can be characterized in relation to time as "coupled" (changing dependently) or "decoupled" (changing independently). We propose the framework HALO, which adopts a causal approach to model these temporal causal relations on two levels. On the representation level, HALO factorizes these two modalities into both coupled and decoupled latent representations, identifying the dynamic interplay between chromatin accessibility and transcription through temporal modulations in the latent space. On the individual gene level, HALO matches gene-peak pairs and characterizes changing dynamics between gene expression and local peaks with time. HALO reveals bipotency in a subset of AT2 cells that exhibit different decisions in lineage specification between systemic sclerosis (SSc) and normal conditions. We demonstrate that using coupled and decoupled information, HALO discovers analogous biological functions between modalities, distinguishes epigenetic factors for lineage specification, and identifies temporal cis-regulation interactions relevant to cellular differentiation and complex human diseases.

bioinformatics↗

bioRxiv2022

Early events marking lung fibroblast transition to profibrotic state in idiopathic pulmonary fibrosis

BackgroundIdiopathic Pulmonary Fibrosis (IPF) is an age-associated progressive lung disease with accumulation of scar tissue impairing gas exchange. Previous high-throughput studies elucidated the role of cellular heterogeneity and molecular pathways in advanced disease. However, critical pathogenic pathways occurring in the transition of fibroblasts from normal to profibrotic have been largely overlooked. MethodsWe used single cell transcriptomics (scRNA-seq) from lungs of healthy controls and IPF patients (lower and upper lobes). We identified fibroblast subclusters, genes and pathways associated to early disease. Immunofluorescence assays validated the role of MOXD1 early in fibrosis. FindingsWe identified four distinct fibroblast subgroups, including one marking the normal-to-profibrotic state transition. Our results show for the first time that global downregulation of ribosomal proteins and significant upregulation of the majority of copper-binding proteins, including MOXD1, mark the IPF transition. We find no significant differences in gene expression in IPF upper and lower lobe samples, which were selected to have low and high degree of fibrosis, respectively. InterpretationEarly events during IPF onset in fibroblasts include dysregulation of ribosomal and copper-binding proteins. Fibroblasts in early stage IPF have already acquired a profibrotic phenotype while hallmarks of advanced disease, including fibroblast foci and honeycomb formation, are still not evident. The new transitional fibroblasts we discover could prove very important for studying the role of fibroblast plasticity in disease progression and help develop early diagnosis tools and therapeutic interventions targeting earlier disease states.

genomics↗

bioRxiv2021

Causal inference using deep-learning variable selection identifies and incorporates direct and indirect causalities in complex biological systems

In complex diseases, causal structure learning across biological variables is critical to identify modifiable triggers or potential therapeutic agents. A limitation of existing causal learning methods is that they cannot identify indirect causal relations, those that would interact through latent mediating variables. We developed the first computational method that identifies both direct and indirect causalities, causal inference using deep-learning variable-selection (causalDeepVASE). To accurately identify indirect causalities and incorporate them with direct causalities, causalDeepVASE develops a deep neural network approach and extends a flexible causal inference method. In simulated and biological data of various contexts, causalDeepVASE outperforms existing methods in identifying expected or validated causal relations. Further, causalDeepVASE facilitates a systematic understanding of complex diseases. For example, causalDeepVASE uniquely identified a possible causal relation between IFN{gamma} and creatinine suggested in a polymicrobial sepsis model. In future biomedical studies, causalDeepVASE can facilitate the identification of driver genes and therapeutic agents.

bioinformatics↗

bioRxiv2021

Myofibroblast transcriptome indicates SFRP2+ fibroblast progenitors in systemic sclerosis skin

Skin and lung fibrosis in systemic sclerosis (SSc) is driven by myofibroblasts, alpha-smooth muscle actin expressing cells that arise from a variety of cell types in murine fibrosis models. Utilizing single cell RNA-sequencing to examine the transcriptome changes, we show that SSc dermal myofibroblasts arise from an SFRP2/DPP4-expressing progenitor fibroblast population that globally upregulates expression of transcriptome markers, such as PRSS23 and THBS1. Only a fraction of SSc fibroblasts differentiate into myofibroblasts, as shown by expression of additional markers, SFRP4 and FNDC1. The myofibroblast transcriptome implicates upstream transcription factors that drive myofibroblast differentiation.

cell biology↗

bioRxiv2021

Bifurcated monocyte states are predictive of mortality in severe COVID-19

Coronavirus disease 2019 (COVID-19) caused by SARS-CoV-2 infection presents with varied clinical manifestations1, ranging from mild symptoms to acute respiratory distress syndrome (ARDS) with high mortality2,3. Despite extensive analyses, there remains an urgent need to delineate immune cell states that contribute to mortality in severe COVID-19. We performed high-dimensional cellular and molecular profiling of blood and respiratory samples from critically ill COVID-19 patients to define immune cell genomic states that are predictive of outcome in severe COVID-19 disease. Critically ill patients admitted to the intensive care unit (ICU) manifested increased frequencies of inflammatory monocytes and plasmablasts that were also associated with ARDS not due to COVID-19. Single-cell RNAseq (scRNAseq)-based deconvolution of genomic states of peripheral immune cells revealed distinct gene modules that were associated with COVID-19 outcome. Notably, monocytes exhibited bifurcated genomic states, with expression of a cytokine gene module exemplified by CCL4 (MIP-1{beta}) associated with survival and an interferon signaling module associated with death. These gene modules were correlated with higher levels of MIP-1{beta} and CXCL10 levels in plasma, respectively. Monocytes expressing genes reflective of these divergent modules were also detectable in endotracheal aspirates. Machine learning algorithms identified the distinctive monocyte modules as part of a multivariate peripheral immune system state that was predictive of COVID-19 mortality. Follow-up analysis of the monocyte modules on ICU day 5 was consistent with bifurcated states that correlated with distinct inflammatory cytokines. Our data suggests a pivotal role for monocytes and their specific inflammatory genomic states in contributing to mortality in life-threatening COVID-19 disease and may facilitate discovery of new diagnostics and therapeutics.

immunology↗

bioRxiv2020

Compensatory hepatic adaptation accompanies permanent absence of intrahepatic biliary network due to YAP1 loss in liver progenitors

YAP1 regulates cell plasticity during liver injury, regeneration and cancer, but its role in liver development is unknown. YAP1 activity was detected in biliary cells and in cells at the hepato-biliary bifurcation in single-cell RNA-sequencing analysis of developing livers. Hepatoblast deletion of Yap1 led to no impairment in Notch-driven SOX9+ ductal plate formation, but prevented the formation of the abutting second layer of SOX9+ ductal cells, blocking the formation of a patent intrahepatic biliary tree. Intriguingly, the mice survived for 8 months with severe cholestatic injury and without any hepatocyte-to-biliary transdifferentiation. Ductular reaction in the perihilar region suggested extrahepatic biliary proliferation likely seeking the missing intrahepatic biliary network. Long-term survival of these mice occurred through hepatocyte adaptation via reduced metabolic and synthetic function including altered bile acid metabolism and transport. Overall, we show YAP1 as a key regulator of bile duct development while highlighting a profound adaptive capability of hepatocytes.

pathology↗

bioRxiv2019

Causal network perturbations for instance-specific analysis of single cell and disease samples

Complex diseases involve perturbation in multiple pathways and a major challenge in clinical genomics is characterizing pathway perturbations in individual samples. This can lead to patient-specific identification of the underlying mechanism of disease thereby improving diagnosis and personalizing treatment. Existing methods rely on external databases to quantify pathway activity scores. This ignores the data dependencies and that pathways are incomplete or condition-specific.\n\nssNPA is a new approach for subtyping samples based on deregulation of their gene networks. ssNPA learns a causal graph directly from control data. Sample-specific network neighborhood deregulation is quantified via the error incurred in predicting the expression of each gene from its Markov blanket. We evaluate the performance of ssNPA on liver development single-cell RNAseq data, where the correct cell timing is recovered. In all analyses ssNPA consistently outperforms alternative methods, highlighting the advantage of network-based approaches.

bioinformatics↗