Disease-specific Gene Expression Research Articles

Exploring the molecular mechanisms of asthma across multiple datasets.

Asthma, a prevalent chronic respiratory disorder, remains enigmatic, notwithstanding considerable advancements in our comprehension. Continuous efforts are crucial for discovering novel molecular targets and gaining a comprehensive understanding of its pathogenesis. In this study, we analyzed gene expression data from 212 individuals, including asthma patients and healthy controls, to identify 267 differentially expressed genes, among which C1orf64 and C7orf26 emerged as potential key genes in asthma pathogenesis. Various bioinformatics tools, including differential gene expression analysis, pathway enrichment, drug target prediction, and single-cell analysis, were employed to explore the potential roles of the genes. Quantitative PCR demonstrated differential expression of C1orf64 and C7orf26 in the asthmatic airway epithelial tissue, implying their potential involvement in asthma pathogenesis. GSEA enrichment analysis revealed significant enrichment of these genes in signaling pathways associated with asthma progression, such as ABC transporters, cell cycle, CAMs, DNA replication, and the Notch signaling pathway. Drug target prediction, based on upregulated and downregulated differential expression, highlighted potential asthma treatments, including Tyrphostin-AG-126, Cephalin, Verrucarin-a, and Emetine. The selection of these drugs was based on their significance in the analysis and their established anti-inflammatory and antiviral invasion properties. Utilizing Seurat and Celldex packages for single-cell sequencing analysis unveiled disease-specific gene expression patterns and cell types. Expression of C1orf64 and C7orf26 in T cells, NK cells, and B cells, instrumental in promoting hallmark features of asthma, was observed, suggesting their potential influence on asthma development and progression. This study uncovers novel genetic aspects of asthma, highlighting potential therapeutic pathways. It exemplifies the power of integrative bioinformatics in decoding complex disease patterns. However, these findings require further validation, and the precise roles of C1orf64 and C7orf26 in asthma warrant additional investigation to validate their therapeutic potential.

Abstract P3079: Integrating Existing Cardiac Single-cell RNA Sequencing Datasets To Uncover The Role Of B-cells In Heart Disease

Introduction: The role of B-cells in human heart failure remains unclear. To address this knowledge gap, we analyzed publicly available datasets of scRNAseq and snRNAseq from patients with various forms of cardiomyopathy. Methods and Results: We used two public datasets including single cells or single nuclei RNAseq data collected from the myocardium of patients with arrhythmogenic right ventricular cardiomyopathy (ARVC), dilated cardiomyopathy (DCM), idiopathic cardiac diseases (ICD), or healthy controls. The first dataset comprised 30 healthy controls and 22 non-ischemic DCM; the second contained 18 controls and 61 with DCM, ARVC, or ICD. Data were normalized and scaled, a UMAP plot was generated, and cells were classified using ScType. B-cells were merged into a single matrix, and the final dataset contained 1100 cells from 32 healthy controls, 59 DCM, and 8 ARVC patients. Four different B-cell clusters were identified. The top 100 biomarkers from each cluster were used to identify B-cell subtypes. Cluster 1 was identified as transitional splenic B-cells; Cluster 2 as plasma cells; Cluster 3 as follicular splenic B-cells; Cluster 4, present only in DCM, as immature B cells. Ingenuity Pathway Analysis (IPA) of the genes enriched in each cluster highlighted that different B cell clusters displayed distinct biological activity, ranging from phagosome formation to B cell receptor signaling and cytokine production. Differentially expressed gene analysis comparing healthy controls to DCM and ARVC showed both similar and disease-specific gene expression patterns in the various B cell clusters. For instance, cluster 2 showed upregulation of oxidative phosphorylation in both DCM and ARVC. On the other hand, DCM cluster 2 showed downregulation of CD40 signaling, while in ARVC the same cluster showed downregulation of IL-6 signaling (Z score >2, p<0.05). Discussion: Our findings suggest that human heart failure is associated with perturbations in B cell gene expression, with both shared and disease-specific features. This data supports the notion that B cells likely play a role in human cardiomyopathy. Further, this data suggests that B-cells play distinct roles in disparate forms of cardiac disease.

De novo drug design based on patient gene expression profiles via deep learning.

Computational de novo drug design is a challenging issue in medicine, and it is desirable to consider all of the relevant information of the biological systems in a disease state. Here, we propose a novel computational method to generate drug candidate molecular structures from patient gene expression profiles via deep learning, which we call DRAGONET. Our model can generate new molecules that are likely to counteract disease-specific gene expression patterns in patients, which is made possible by exploring the latent space constructed by a transformer-based variational autoencoder and integrating the substructures of disease-correlated molecules. We applied DRAGONET to generate drug candidate molecules for gastric cancer, atopic dermatitis, and Alzheimer's disease, and demonstrated that the newly generated molecules were chemically similar to registered drugs for each disease. This approach is applicable to diseases with unknown therapeutic target proteins and will make a significant contribution to the field of precision medicine.

Drug repositioning prediction for psoriasis using the adverse event reporting database.

Inverse signals produced from disproportional analyses using spontaneous drug adverse event reports can be used for drug repositioning purposes. The purpose of this study is to predict drug candidates using a computational method that integrates reported drug adverse event data, disease-specific gene expression profiles, and drug-induced gene expression profiles. Drug and adverse events from 2015 through 2020 were downloaded from the United States Food and Drug Administration Adverse Event Reporting System (FAERS). The reporting odds ratio (ROR), information component (IC) and empirical Bayes geometric mean (EBGM) were used to calculate the inverse signals. Psoriasis was selected as the target disease. Disease specific gene expression profiles were obtained by the meta-analysis of the Gene Expression Omnibus (GEO). The reverse gene expression scores were calculated using the Library of Integrated Network-based Cellular Signatures (LINCS) and their correlations with the inverse signals were obtained. Reversal genes and the candidate compounds were identified. Additionally, these correlations were validated using the relationship between the reverse gene expression scores and the half-maximal inhibitory concentration (IC50) values from the Chemical European Molecular Biology Laboratory (ChEMBL). Inverse signals produced from a disproportional analysis can be used for drug repositioning and to predict drug candidates against psoriasis.

Transcriptome Analysis of Dermal Fibroblasts Derived From Visceral Leishmaniasis and Post-Kala-Azar Dermal Leishmaniasis Patients Reveal Disease-Specific Gene Expression and Pathological Regulation.

Post-kala-azar dermal leishmaniasis (PKDL), a dermal form of the disease, occurs in some visceral leishmaniasis (VL) patients following treatment. The PKDL disease mechanism is not yet clearly understood. Here we have studied the role of dermal fibroblasts in VL and PKDL disease mechanism. Dermal fibroblasts were grown from skin biopsy explants collected from individual VL and PKDL patients and healthy controls. Fibroblasts from the third passage were subjected to RNA sequencing to analyze differentially expressed genes (DEGs). Significantly important genes were further validated by reverse transcription polymerase chain reaction (RT-PCR) and enzyme-linked immunosorbent assay (ELISA). Transcriptome analysis of PKDL versus VL identified 516 DEGs (263 were overrepresented and 253 were underrepresented in PKDL). Among the top hub genes, MMP2, IL1B, CXCL8, IFIH1, NFKB1A, IL6, ISG15, and EGFR were underexpressed and ACTB, HSP90AA1, RAB7A, and RPS27A were overexpressed in PKDL compared to VL. Our data indicate that PKDL fibroblasts may present antigens through the MHC I pathway activating CD8+ T-cell mediated response, while VL fibroblasts express nuclear factor-κB (NFκB)-mediated chemokines, IL1B, IL6, and IL8, resulting in the recruitment of natural killer (NK)-cells and monocytes to the site of infection, leading to the clearance of parasite from the skin and visceralization of the disease.

Shared and disease-specific glial gene expression changes in neurodegenerative diseases.

We found evolutionarily conserved astrocyte and microglia subpopulations shared across multiple brain regions. We reveal similarities and differences between Alzheimer's disease glia and Parkinson's disease glia, as well as regional variance linked to disease pathology and neurodegeneration.

Large-scale assessment of pros and cons of autopsy-derived or tumor-matched tissues as the norms for gene expression analysis in cancers

Normal tissues are essential for studying disease-specific differential gene expression. However, healthy human controls are typically available only in postmortal/autopsy settings. In cancer research, fragments of pathologically normal tissue adjacent to tumor site are frequently used as the controls. However, it is largely underexplored how cancers can systematically influence gene expression of the neighboring tissues. Here we performed a comprehensive pan-cancer comparison of molecular profiles of solid tumor-adjacent and autopsy-derived “healthy” normal tissues. We found a number of systemic molecular differences related to activation of the immune cells, intracellular transport and autophagy, cellular respiration, telomerase activation, p38 signaling, cytoskeleton remodeling, and reorganization of the extracellular matrix. The tumor-adjacent tissues were deficient in apoptotic signaling and negative regulation of cell growth including G2/M cell cycle transition checkpoint. We also detected an extensive rearrangement of the chemical perception network. Molecular targets of 32 and 37 cancer drugs were over- or underexpressed, respectively, in the tumor-adjacent norms. These processes may be driven by molecular events that are correlated between the paired cancer and adjacent normal tissues, that mostly relate to inflammation and regulation of intracellular molecular pathways such as the p38, MAPK, Notch, and IGF1 signaling. However, using a model of macaque postmortal tissues we showed that for the 30 min – 24-hour time frame at 4ºC, an RNA degradation pattern in lung biosamples resulted in an artifact “differential” expression profile for 1140 genes, although no differences could be detected in liver. Thus, such concerns should be addressed in practice.

Spatially resolved whole transcriptome profiling in human and mouse tissue using Digital Spatial Profiling.

Emerging spatial profiling technology has enabled high-plex molecular profiling in biological tissues, preserving the spatial and morphological context of gene expression. Here, we describe expanding the chemistry for the Digital Spatial Profiling platform to quantify whole transcriptomes in human and mouse tissues using a wide range of spatial profiling strategies and sample types. We designed multiplexed in situ hybridization probes targeting the protein-coding genes of the human and mouse transcriptomes, referred to as the human or mouse Whole Transcriptome Atlas (WTA). Human and mouse WTAs were validated in cell lines for concordance with orthogonal gene expression profiling methods in regions ranging from ∼10-500 cells. By benchmarking against bulk RNA-seq and fluorescence in situ hybridization, we show robust transcript detection down to ∼100 transcripts per region. To assess the performance of WTA across tissue and sample types, we applied WTA to biological questions in cancer, molecular pathology, and developmental biology. Spatial profiling with WTA detected expected gene expression differences between tumor and tumor microenvironment, identified disease-specific gene expression heterogeneity in histological structures of the human kidney, and comprehensively mapped transcriptional programs in anatomical substructures of nine organs in the developing mouse embryo. Digital Spatial Profiling technology with the WTA assays provides a flexible method for spatial whole transcriptome profiling applicable to diverse tissue types and biological contexts.

Deciphering the molecular mechanism and crosstalk between Parkinson's disease and breast cancer through multi-omics and drug repurposing approach

Epidemiological studies indicate a higher occurrence of breast cancer (BRCA) in patients with Parkinson's disease. However, the exact molecular mechanism is still not precise. Herein, we tested the hypothesis that this inverse comorbidity result from shared genetic and molecular processes. We conducted an integrated omics analysis to identify the common gene signatures associated with PD and BRCA. Secondly, several dysregulated biological processes in both indications were analyzed by functional enrichment methods, and significant overlapping processes were identified. To establish common regulatory mechanisms, information about transcription factors and miRNAs associated with both the disorders was extracted. Finally, disease-specific gene expression signatures were compared through LINCS L1000 analysis to identify potential repurposing drugs for PD. The potential repurposed drug candidates were then correlated with PD-specific gene signatures by Cmap analysis. In conclusion, this study highlights the shared genes, biological pathways and regulatory signatures associated with PD and BRCA with an improved understanding of crosstalk involved. Additionally, the role of therapeutics was investigated in context with their comorbid associations. These findings could help to explain the complex molecular patterns of associations between PD and BRCA.

A New Approach to Drug Repurposing with Two-Stage Prediction, Machine Learning, and Unsupervised Clustering of Gene Expression.

Drug repurposing has broad importance in planetary health for therapeutics innovation in infectious diseases as well as common or rare chronic human diseases. Drug repurposing has also proved important to develop interventions against the COVID-19 pandemic. We propose a new approach for drug repurposing involving two-stage prediction and machine learning. First, diseases are clustered by gene expression on the premise that similar patterns of altered gene expression imply critical pathways shared in different disease conditions. Next, drug efficacy is assessed by the reversibility of abnormal gene expression, and results are clustered to identify repurposing targets. To cluster similar diseases, gene expression data from 262 cases of 31 diseases and 268 controls were analyzed by Uniform Manifold Approximation and Projection for Dimension Reduction followed by k-means to optimize the number of clusters. For evaluation, we examined disease-specific gene expression data for inclusion, body myositis, polymyositis, and dermatomyositis (DM), and used LINCS L1000 characteristic direction signatures search engine (L1000CDS2) to obtain lists of small-molecule compounds that reversed the expression patterns of these specifically altered genes as candidates for drug repurposing. Finally, the functions of affected genes were analyzed by Gene Set Enrichment Analysis to examine consistency with expected drug efficacy. Consequently, we found disease-specific gene expression, and importantly, identified 20 drugs such as BMS-387032, phorbol-12-myristate-13-acetate, mitoxantrone, alvocidib, and vorinostat as candidates for repurposing. These were previously noted to be effective against two of the three diseases, and have a high probability of being effective against the other. That is, inclusion body myositis and DM. The two-stage prediction approach to drug repurposing presented here offers innovation to inform future drug discovery and clinical trials in a variety of human diseases.

Single-cell transcriptomics provides insights into hypertrophic cardiomyopathy.

Hypertrophic cardiomyopathy (HCM) is a genetic heart disease that is characterized by unexplained segmental hypertrophy that is usually most pronounced in the septum. While sarcomeric gene mutations are often the genetic basis for HCM, the mechanistic origin for the heterogeneous remodeling remains largely unknown. A better understanding of the gene networks driving the cardiomyocyte (CM) hypertrophy is required to improve therapeutic strategies. Patients suffering from HCM often receive a septal myectomy surgery to relieve outflow tract obstruction due to hypertrophy. Using single-cell RNA sequencing (scRNA-seq) on septal myectomy samples from patients with HCM, we identify functional links between genes, transcription factors, and cell size relevant for HCM. The data show the utility of using scRNA-seq on the human hypertrophic heart, highlight CM heterogeneity, and provide a wealth of insights into molecular events involved in HCM that can eventually contribute to the development of enhanced therapies.

Random forest-integrated analysis in AD and LATE brain transcriptome-wide data to identify disease-specific gene expression.

Alzheimer’s disease (AD) is a complex neurodegenerative disorder that affects thinking, memory, and behavior. Limbic-predominant age-related TDP-43 encephalopathy (LATE) is a recently identified common neurodegenerative disease that mimics the clinical symptoms of AD. The development of drugs to prevent or treat these neurodegenerative diseases has been slow, partly because the genes associated with these diseases are incompletely understood. A notable hindrance from data analysis perspective is that, usually, the clinical samples for patients and controls are highly imbalanced, thus rendering it challenging to apply most existing machine learning algorithms to directly analyze such datasets. Meeting this data analysis challenge is critical, as more specific disease-associated gene identification may enable new insights into underlying disease-driving mechanisms and help find biomarkers and, in turn, improve prospects for effective treatment strategies. In order to detect disease-associated genes based on imbalanced transcriptome-wide data, we proposed an integrated multiple random forests (IMRF) algorithm. IMRF is effective in differentiating putative genes associated with subjects having LATE and/or AD from controls based on transcriptome-wide data, thereby enabling effective discrimination between these samples. Various forms of validations, such as cross-domain verification of our method over other datasets, improved and competitive classification performance by using identified genes, effectiveness of testing data with a classifier that is completely independent from decision trees and random forests, and relationships with prior AD and LATE studies on the genes linked to neurodegeneration, all testify to the effectiveness of IMRF in identifying genes with altered expression in LATE and/or AD. We conclude that IMRF, as an effective feature selection algorithm for imbalanced data, is promising to facilitate the development of new gene biomarkers as well as targets for effective strategies of disease prevention and treatment.

In Vitro andpre-Clinical In Vivo evidence Support MDMX/MDM2 As Common Vulnerabilities across TP53 -Wild-Type T-Cell Lymphomas That Are Targetable with the α-Helical p53 Stapled Peptide Alrn-6924

T-cell lymphomas (TCL) are a heterogeneous group of non-Hodgkin lymphomas with poor outcomes. Biological heterogeneity across TCLs has hampered the development of targeted therapies. Thus, we sought to identify common vulnerabilities that could be targeted across multiple TCL subtypes. We performed transcriptome and exome sequencing of 21 TCL cell lines and 8 TCL patient-derived xenografts (PDX), and a genome-wide CRISPR-Cas9 pooled screening in 7 TCL lines. Sequencing identified multiple genomic alterations of clinical significance and disease-specific recurrence, like activating JAK1 and STAT3 mutations in cell lines and a PDX-model of anaplastic large cell lymphoma (ALCL) or mutations of TET2, DNMT3A and RHOA in PDX-models of angioimmunoblastic TCL (AITL). Unsupervised hierarchical clustering of transcriptomes identified disease-specific gene-expression profiles across cell lines and PDX models. Ranking the genetic vulnerabilities of the CRISPR-Cas9 screen by z-score and recurrence across multiple cell lines, we identified both previously defined (e.g. IRF4) and novel vulnerabilities (e.g. PTPN2) within multiple TCL subtypes. The most consistent dependence in TP53- wild-type (WT) lines was the E3 ubiquitin ligase MDM2, which polyubiquitinates and targets p53 for degradation. Multiple TCL lines harbored copy number gains or amplifications of MDM2 and/or MDMX that correlated with high transcript levels. To address the therapeutic potential for targeting MDMX/MDM2 in TCL we tested ALRN-6924, a p53-derived stapled peptide that inhibits both MDMX and MDM2, and that is currently in multicenter phase I/Ib and phase II trials. ALRN-6924 was recently shown to induce objective responses in a variety of cancers as a single agent in phase 1 dose-escalation (Meric-Bernstam, F. et al, ASCO 2017). Our results showed that ALRN-6924 potently induced G1 cell cycle arrest and apoptotic cell death in WT TP53 TCL lines (n=8; IC50 100 nm-1 µM) while inducing expression of p21 and Annexin V. ALRN-6924 was inactive in TP53 -mutant lines (n=11; IC50 > 5 µM). Next, we performed a phase II-like trial in eight WT TP53 PDXs: hepatosplenic (HS-TCL), cutaneous (CTCL), NK/T, AITL (n=2), T-cell prolymphocytic leukemia (TPLL) and Alk+ ALCL (n=2). At the time of extensive lymphoma engraftment (e.g. >20% splenic involvement), mice were randomized to treatment with vehicle, ALRN-6924 or the 2nd line standard-of-care agent Romidepsin (n=3/arm/model) on days 1, 4 and 7, and sacrificed on day 8. ALRN-6924 had a favorable safety profile and significantly reduced tumor burden compared to vehicle in all 8 models, with superior efficacy to Romidepsin in the majority of models (Figure). Residual tumor cells collected from ALRN-6924-treated mice consistently had increased AAD/Annexin V staining and p21 induction, consistent with on-target p53 activation and apoptosis.In summary, we identified MDM2/MDMX as a targetable vulnerability that is effectively targeted by the stapled peptide ALRN-6924 in TP53 WT TCLs with promising activity across multiple subtypes, supporting further clinical development of ALRN-6924 in TCL for patients requiring 2nd or frontline treatment. A Phase IIa clinical trial of ALRN-6924 in PTCL patients is currently enrolling (NCT02264613). [Display omitted] DisclosuresHorwitz:Millenium/Takeda: Consultancy, Research Funding; HUYA: Consultancy; Forty-Seven: Consultancy, Research Funding; BMS: Consultancy; Seattle Genetics: Consultancy, Research Funding; Kyowa-Hakka-Kirin: Consultancy, Research Funding; Celgene: Consultancy, Research Funding; Mundipharma: Consultancy; ADCT Therapeutics: Research Funding; Aileron Therapeutics: Research Funding; Infinity/Verastem: Consultancy, Research Funding. Santiago:Aileron: Employment. Ren:Aileron: Employment. Guerlavais:Aileron: Employment. Annis:Aileron: Employment. Aivado:Aileron: Employment.

Transcriptomic and Genetic Associations between Alzheimer's Disease, Parkinson's Disease, and Cancer.

Simple SummaryEpidemiological studies have identified a link between neurodegenerative disorders and a reduced risk of overall cancer. Increases and decreases in the risk of site-specific cancers have also been reported. However, it is still unknown whether these associations arise due to shared genetic and molecular factors or are explained by other phenomena (e.g., biases in epidemiological studies or the use of medication). In this study, we aimed to investigate the potential molecular, genetic, and pharmacological links between Alzheimer’s and Parkinson’s diseases and a large panel of 22 cancer types. To examine the overlapping involvement of genes and pathways, we obtained differential gene expression profiles through meta-analyses of post-mortem brain tissues from Alzheimer’s and Parkinson’s disease patients, primary tumors, and tissue-matched controls, and compared them. Genetic similarities were assessed through network-based methods and the computation of genetic correlations. Finally, the potential impact of drugs indicated for each disorder in the identified associations was evaluated using transcriptomic methods. Our research extends previous work in the field by identifying new significant patterns of transcriptomic associations (direct and inverse) between Alzheimer’s disease, Parkinson’s disease, and different site-specific cancers. The results reveal significant genetic correlations between Parkinson’s disease, prostate cancer, and melanoma. In addition, to our knowledge, this is the first time that the role of drugs indicated for the treatment of both sets of disorders has been investigated in the context of their comorbid associations using transcriptomic methods.Alzheimer’s (AD) and Parkinson’s diseases (PD) are the two most prevalent neurodegenerative disorders in human populations. Epidemiological studies have shown that patients suffering from either condition present a reduced overall risk of cancer than controls (i.e., inverse comorbidity), suggesting that neurodegeneration provides a protective effect against cancer. Reduced risks of several site-specific tumors, including colorectal, lung, and prostate cancers, have also been observed in AD and PD. By contrast, an increased risk of melanoma has been described in PD patients (i.e., direct comorbidity). Therefore, a fundamental question to address is whether these associations are due to shared genetic and molecular factors or are explained by other phenomena, such as flaws in epidemiological studies, exposure to shared risk factors, or the effect of medications. To this end, we first evaluated the transcriptomes of AD and PD post-mortem brain tissues derived from the hippocampus and the substantia nigra and analyzed their similarities to those of a large panel of 22 site-specific cancers, which were obtained through differential gene expression meta-analyses of array-based studies available in public repositories. Genes and pathways that were deregulated in both disorders in each analyzed pair were examined. Second, we assessed potential genetic links between AD, PD, and the selected cancers by establishing interactome-based overlaps of genes previously linked to each disorder. Then, their genetic correlations were computed using cross-trait LD score regression and GWAS summary statistics data. Finally, the potential role of medications in the reported comorbidities was assessed by comparing disease-specific differential gene expression profiles to an extensive collection of differential gene expression signatures generated by exposing cell lines to drugs indicated for AD, PD, and cancer treatment (LINCS L1000). We identified significant inverse associations of transcriptomic deregulation between AD hippocampal tissues and breast, lung, liver, and prostate cancers, and between PD substantia nigra tissues and breast, lung, and prostate cancers. Moreover, significant direct (same direction) associations of deregulation were observed between AD and PD and brain and thyroid cancers, as well as between PD and kidney cancer. Several biological processes, including the immune system, oxidative phosphorylation, PI3K/AKT/mTOR signaling, and the cell cycle, were found to be deregulated in both cancer and neurodegenerative disorders. Significant genetic correlations were found between PD and melanoma and prostate cancers. Several drugs indicated for the treatment of neurodegenerative disorders and cancer, such as galantamine, selegiline, exemestane, and estradiol, were identified as potential modulators of the comorbidities observed between neurodegeneration and cancer.

CorGO: An Integrated Method for Clustering Functionally Similar Genes.

Identification of groups of co-expressed or co-regulated genes is critical for exploring the underlying mechanism behind a particular disease like cancer. Condition-specific (disease-specific) gene-expression profiles acquired from different platforms are widely utilized by researchers to get insight into the regulatory mechanism of the disease. Several clustering algorithms are developed using gene expression profiles to identify the group of similar genes. These algorithms are computationally efficient but are not able to capture the functional similarity present between the genes, which is very important from a biological perspective. In this study, an algorithm named CorGO is introduced, that specifically deals with the identification of functionally similar gene-clusters. Two types of relationships are calculated for this purpose. Firstly, the Correlation (Cor) between the genes are captured from the gene-expression data, which helps in deciphering the relationship between genes based on its expression across several diseased samples. Secondly, Gene Ontology (GO)-based semantic similarity information available for the genes is utilized, that helps in adding up biological relevance to the identified gene-clusters. A similarity measure is defined by integrating these two components that help in the identification of homogeneous and functionally similar groups of genes. CorGO is applied to four different types of gene expression profiles of different types of cancer. Gene-clusters identified by CorGO, are further validated by pathway enrichment, disease enrichment, and network analysis. These biological analyses demonstrated significant connectivity and functional relatedness within the genes of the same cluster. A comparative study with commonly used clustering algorithms is also performed to show the efficacy of the proposed method.

Induced Pluripotent Stem Cell–derived Hepatocytes From Patients With Nonalcoholic Fatty Liver Disease Display a Disease-specific Gene Expression Profile

Nonalcoholic fatty liver disease (NAFLD) is a leading cause of liver disease worldwide. Animal models are widely used to investigate the mechanisms of fatty liver disease, but they do not faithfully represent NAFLD in humans.1 Thus, there is strong interest in studying NAFLD pathogenesis directly in humans whenever possible. Some researchers have used embryonic or induced pluripotent stem cells (iPSCs) derived from individual human subjects to generate complex liver platforms that model an NAFLD-like state in culture.

Ileal Derived Organoids From Crohn’s Disease Patients Show Unique Transcriptomic and Secretomic Signatures

Background & AimsWe used patient-derived organoids (PDOs) to study the epithelial-specific transcriptional and secretome signatures of the ileum during Crohn’s disease (CD) with varying phenotypes to screen for disease profiles and potential druggable targets.MethodsRNA sequencing was performed on isolated intestinal crypts and 3-week-old PDOs derived from ileal biopsies of CD patients (n = 8 B1, inflammatory; n = 8 B2, stricturing disease) and non-inflammatory bowel disease (IBD) controls (n = 13). Differentially expressed (DE) genes were identified by comparing CD vs control, B1 vs B2, and inflamed vs non-inflamed. DE genes were used for computational screening to find candidate small molecules that could potentially reverse B1and B2 gene signatures. The secretome of a second cohort (n = 6 non-IBD controls, n = 7 CD, 5 non-inflamed, 2 inflamed) was tested by Luminex using cultured organoid conditioned medium.ResultsWe found 90% similarity in both the identity and abundance of protein coding genes between PDOs and intestinal crypts (15,554 transcripts of 19,900 genes). DE analysis identified 814 genes among disease group (CD vs non-IBD control), 470 genes different between the CD phenotypes, and 5 false discovery rate correction significant genes between inflamed and non-inflamed CD. The PDOs showed both similarity and diversity in the levels and types of soluble cytokines and growth factors they released. Perturbagen analysis revealed potential candidate compounds to reverse B2 disease phenotype to B1 in PDOs.ConclusionsPDOs are similar at the transcriptome level with the in vivo epithelium and retain disease-specific gene expression for which we have identified secretome products, druggable targets, and corresponding pharmacologic agents. Targeting the epithelium could reverse a stricturing phenotype and improve outcomes.

Identification of a disease-specific gene expression profile of children with acute asthma by weighted gene co-expression network analysis.

Asthma is one of the most common diseases, with a high prevalence among children. To date, systemic co-expression analysis for this disease has not been undertaken to explain its pathogenesis. Here we identified differentially expressed genes (DEGs) in 87 samples, and then constructed co-expression modules via weighted gene co-expression network analysis (WGCNA) and investigated the functional enrichment of co-expressed genes in terms of Gene Ontology and Kyoto Encyclopedia of Genes and Genomes (KEGG). Meanwhile, protein-protein interaction (PPI) network and miRNA-transcription factor-target (miRNA-TF-target) regulatory network analyses were performed to screen hub genes. As a result, 3,469 DEGs were identified in this study, of which 1,860 genes were up-regulated and 1,609 genes were down-regulated. Using WGCNA, we identified two key modules, named MEbrown and MEblue, that may play important roles in asthma. Functional enrichment analysis revealed that MEbrown was enriched in 37 KEGG pathways and 472 biological processes (BPs), while MEblue was enriched in 16 KEGG pathways and 449 BPs. From PPI and miRNA-TF-target regulatory network analysis, a total of 31 TFs, seven miRNAs and 28 nodes were identified. Our findings should provide a framework of therapeutic targets for treating children with acute asthma.

GuiltyTargets: Prioritization of Novel Therapeutic Targets With Network Representation Learning.

The majority of clinical trials fail due to low efficacy of investigated drugs, often resulting from a poor choice of target protein. Existing computational approaches aim to support target selection either via genetic evidence or by putting potential targets into the context of a disease specific network reconstruction. The purpose of this work was to investigate whether network representation learning techniques could be used to allow for a machine learning based prioritization of putative targets. We propose a novel target prioritization approach, GuiltyTargets, which relies on attributed network representation learning of a genome-wide protein-protein interaction network annotated with disease-specific differential gene expression and uses positive-unlabeled (PU) machine learning for candidate ranking. We evaluated our approach on 12 datasets from six diseases of different type (cancer, metabolic, neurodegenerative) within a 10 times repeated 5-fold stratified cross-validation and achieved AUROC values between 0.92 - 0.97, significantly outperforming previous approaches that relied on manually engineered topological features. Moreover, we showed that GuiltyTargets allows for target repositioning across related disease areas. An application of GuiltyTargets to Alzheimer's disease resulted in a number of highly ranked candidates that are currently discussed as targets in the literature. Interestingly, one (COMT) is also the target of an approved drug (Tolcapone) for Parkinson's disease, highlighting the potential for target repositioning with our method. The GuiltyTargets Python package is available on PyPI and all code used for analysis can be found under the MIT License at https://github.com/GuiltyTargets. Attributed network representation learning techniques provide an interesting approach to effectively leverage the existing knowledge about the molecular mechanisms in different diseases. In this work, the combination with positive-unlabeled learning for target prioritization demonstrated a clear superiority compared to classical feature engineering approaches. Our work highlights the potential of attributed network representation learning for target prioritization. Given the overarching relevance of networks in computational biology we believe that attributed network representation learning techniques could have a broader impact in the future.

RNA sequencing of corneas from two keratoconus patient groups identifies potential biomarkers and decreased NRF2-antioxidant responses

Keratoconus is a highly prevalent (1 in 2000), genetically complex and multifactorial, degenerative disease of the cornea whose pathogenesis and underlying transcriptomic changes are poorly understood. To identify disease-specific changes and gene expression networks, we performed next generation RNA sequencing from individual corneas of two distinct patient populations - one from the Middle East, as keratoconus is particularly severe in this group, and the second from an African American population in the United States. We conducted a case: control RNA sequencing study of 7 African American, 12 Middle Eastern subjects, and 7 controls. A Principal Component Analysis of all expressed genes was used to ascertain differences between samples. Differentially expressed genes were identified using Cuffdiff and DESeq2 analyses, and identification of over-represented signaling pathways by Ingenuity Pathway Analysis. Although separated by geography and ancestry, key commonalities in the two patient transcriptomes speak of disease - intrinsic gene expression networks. We identified an overwhelming decrease in the expression of anti-oxidant genes regulated by NRF2 and those of the acute phase and tissue injury response pathways, in both patient groups. Concordantly, NRF2 immunofluorescence staining was decreased in patient corneas, while KEAP1, which helps to degrade NRF2, was increased. Diminished NRF2 signaling raises the possibility of NRF2 activators as future treatment strategies in keratoconus. The African American patient group showed increases in extracellular matrix transcripts that may be due to underlying profibrogenic changes in this group. Transcripts increased across all patient samples include Thrombospondin 2 (THBS2), encoding a matricellular protein, and cellular proteins, GAS1, CASR and OTOP2, and are promising biomarker candidates. Our approach of analyzing transcriptomic data from different populations and patient groups will help to develop signatures and biomarkers for keratoconus subtypes. Further, RNA sequence data on individual patients obtained from multiple studies may lead to a core keratoconus signature of deregulated genes and a better understanding of its pathogenesis.