Dataset Information

CNet: a multi-omics approach to detecting clinically associated, combinatory genomic signatures.

ABSTRACT: MOTIVATION:Genome-wide multi-omics profiling of complex diseases provides valuable resources and opportunities to discover associations between various measures of genes and diseases. Currently, a pressing challenge is how to effectively detect functional genes associated with or causing phenotypic outcomes. We developed CNet to identify groups of genomic signatures whose combinatory effect is significantly associated with clinical and phenotypical outcomes. RESULTS:CNet builds on a generalized sequential feedforward method, augmented by a down-sampling bootstrap strategy to reduce random hitchhiking signatures. It further applies a dynamic trimming procedure to remove relatively less informative signatures at every step. CNet can manage heterogeneous genomic signature profiles simultaneously and select the best signature to represent a specific gene. To deal with various forms of clinical and phenotypical measurements, we introduced four models to deal with continuous, categorical and censored data. We tested CNet using drug-response data, multidimensional cancer genomics data and genome-wide association study data for multiple traits. Our results demonstrated that in various scenarios, CNet could effectively identify signatures that are associated with the outcomes. In addition, we applied CNet to identify likely disease-causing chains involving somatic mutations, pathway activities and patient outcomes. With appropriate setting, CNet can be applied in many biological conditions. AVAILABILITY AND IMPLEMENTATION:CNet can be downloaded at https://github.com/bsml320/CNet. SUPPLEMENTARY INFORMATION:Supplementary data are available at Bioinformatics online.

SUBMITTER: Jia P

PROVIDER: S-EPMC6954662 | biostudies-literature | 2019 Dec

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

CNet: a multi-omics approach to detecting clinically associated, combinatory genomic signatures.

Jia Peilin P Pei Guangsheng G Zhao Zhongming Z

Bioinformatics (Oxford, England) 20191201 24

<h4>Motivation</h4>Genome-wide multi-omics profiling of complex diseases provides valuable resources and opportunities to discover associations between various measures of genes and diseases. Currently, a pressing challenge is how to effectively detect functional genes associated with or causing phenotypic outcomes. We developed CNet to identify groups of genomic signatures whose combinatory effect is significantly associated with clinical and phenotypical outcomes.<h4>Results</h4>CNet builds on ...[more]

PMID: 31141125

Similar Datasets

Project description:BackgroundProstate cancer, the second most prevalent malignancy among men, poses a significant threat to affected patients' well-being due to its poor prognosis. Novel biomarkers are required to enhance clinical outcomes and tailor personalized treatments. Herein, we describe our research to explore the prognostic value of long non-coding RNAs (lncRNAs) deregulated by copy number variations (CNVs) in prostate cancer.MethodsThe study employed an integrative multi-omics data analysis of the prostate cancer transcriptomic, CNV and methylation datasets to identify prognosis-related subtypes. Subtype-specific expression profiles of protein-coding genes (PCGs) and lncRNAs were determined. We analysed CNV patterns of lncRNAs across the genome to identify subtype-specific lncRNAs with CNV changes. LncRNAs exhibiting significant amplification or deletion and a positive correlation were designated CNV-deregulated lncRNAs. A prognostic risk score model was subsequently developed using these CNV-driven lncRNAs.ResultsSix molecular subtypes of prostate cancer were identified, demonstrating significant differences in prognosis (P = 0.034). The CNV profiles of subtype-specific lncRNAs were examined, revealing their correlation with CNV amplification or deletion. Six lncRNAs (CCAT2, LINC01593, LINC00276, GACAT2, LINC00457, LINC01343) were selected based on significant CNV amplifications or deletions using a rigorous univariate Cox proportional risk regression model. A robust risk score model was developed, stratifying patients into high-risk and low-risk categories. Notably, our prognostic model based on these six lncRNAs exhibited exceptional predictive capabilities for recurrence-free survival (RFS) in prostate cancer patients (P = 0.024).ConclusionsOur study successfully identified a prognostic risk score model comprising six CNV-driven lncRNAs that could potentially be prognostic biomarkers for prostate cancer. These lncRNA signatures are closely associated with RFS, providing promising prospects for improved patient prognostication and personalized therapeutic strategies for novel prostate cancer treatment.

Project description:ObjectivesCervical Squamous Cell Carcinoma (CESC) is one of the most fatal female malignancies, and the underlying molecular mechanisms governing this disease have not been fully explored. In this research, we planned to conduct the analysis of Gene Expression Omnibus (GEO) cervical squamous cell carcinoma microarray datasets by a detailed in silico approach and to explore some novel biomarkers of CESC.MethodsThe top commonly differentially expressed genes (DEGs) from the GSE138080 and GSE113942 datasets were analyzed by Limma package-based GEO2R tool. The protein-protein interaction (PPI) network of the DEGs was drawn through Search Tool for the Retrieval of Interacting Genes (STRING), and top 6 hub genes were obtained from Cytoscape. Expression analysis and validation of hub genes expression in CESC samples and cell lines were done using UALCAN, OncoDB, GENT2, and HPA. Additionally, cBioPortal, Gene set enrichment analysis (GSEA) tool, Kaplan-Meier (KM) plotter, ShinyGO, and DGIdb databases were also used to check some important values of hub genes in CESC.ResultsOut of 79 DEGs, the minichromosome maintenance complex component 4 (MCM4), nucleolar and spindle-associated protein 1 (NUSAP1), cell division cycle associated 5 (CDCA5), cell division cycle 45 (CDC45), denticleless E3 ubiquitin protein ligase homolog (DTL), and chromatin licensing and DNA replication factor 1 (CDT1) genes were regarded as hub genes in CESC. Further analysis revealed that the expressions of all these hub genes were significantly elevated in CESC cell lines and samples of diverse clinical attributes. In this study, we also documented some important correlations between hub genes and some other diverse measures, including DNA methylation, genetic alterations, and Overall Survival (OS). Last, we also identify hub genes associated ceRNA network and 31 important chemotherapeutic drugs.ConclusionThrough detailed in silico methodology, we identified 6 hub genes, including MCM4, NUSAP1, CDCA5, CDC45, DTL, and CDT1, which are likely to be associated with CESC development and diagnosis.

Project description:BackgroundRett syndrome (RTT) is a neurodevelopmental disorder mainly caused by mutations in the methyl-CpG-binding protein 2 gene (MECP2). MeCP2 is a multi-functional protein involved in many cellular processes, but the mechanisms by which its dysfunction causes disease are not fully understood. The duplication of the MECP2 gene causes a distinct disorder called MECP2 duplication syndrome (MDS), highlighting the importance of tightly regulating its dosage for proper cellular function. Additionally, some patients with mutations in genes other than MECP2 exhibit phenotypic similarities with RTT, indicating that these genes may also play a role in similar cellular functions. The purpose of this study was to characterise the molecular alterations in patients with RTT in order to identify potential biomarkers or therapeutic targets for this disorder.MethodsWe used a combination of transcriptomics (RNAseq) and proteomics (TMT mass spectrometry) to characterise the expression patterns in fibroblast cell lines from 22 patients with RTT and detected mutation in MECP2, 15 patients with MDS, 12 patients with RTT-like phenotypes and 13 healthy controls. Transcriptomics and proteomics data were used to identify differentially expressed genes at both RNA and protein levels, which were further inspected via enrichment and upstream regulator analyses and compared to find shared features in patients with RTT.ResultsWe identified molecular alterations in cellular functions and pathways that may contribute to the disease phenotype in patients with RTT, such as deregulated cytoskeletal components, vesicular transport elements, ribosomal subunits and mRNA processing machinery. We also compared RTT expression profiles with those of MDS seeking changes in opposite directions that could lead to the identification of MeCP2 direct targets. Some of the deregulated transcripts and proteins were consistently affected in patients with RTT-like phenotypes, revealing potentially relevant molecular processes in patients with overlapping traits and different genetic aetiology.ConclusionsThe integration of data in a multi-omics analysis has helped to interpret the molecular consequences of MECP2 dysfunction, contributing to the characterisation of the molecular landscape in patients with RTT. The comparison with MDS provides knowledge of MeCP2 direct targets, whilst the correlation with RTT-like phenotypes highlights processes potentially contributing to the pathomechanism leading these disorders.

Project description:BackgroundThe enrichment of Gram-negative bacteria of oral origin in the esophageal microbiome has been associated with the development of metaplasia. However, to date, no study has comprehensively assessed the relationships between the esophageal microbiome and the host.MethodsHere, we examine the esophageal microenvironment in gastro-esophageal reflux disease and metaplasia using multi-omics strategies targeting the microbiome and host transcriptome, followed by targeted culture, comparative genomics, and host-microbial interaction studies of bacterial signatures of interest.ResultsProfiling of the host transcriptome from esophageal mucosal biopsies revealed profound changes during metaplasia. Importantly, five biomarkers showed consistent longitudinal changes with disease progression from reflux disease to metaplasia. We showed for the first time that the esophageal microbiome is distinct from the salivary microbiome and the enrichment of Campylobacter species as a consistent signature in disease across two independent cohorts. Shape fitting and matrix correlation identified associations between the microbiome and host transcriptome profiles, with a novel co-exclusion relationship found between Campylobacter and napsin B aspartic peptidase. Targeted culture of Campylobacter species from the same cohort revealed a subset of isolates to have a higher capacity to survive within primary human macrophages. Comparative genomic analyses showed these isolates could be differentiated by specific genomic features, one of which was validated to be associated with intracellular fitness. Screening for these Campylobacter strain-specific signatures in shotgun metagenomics data from another cohort showed an increase in prevalence with disease progression. Comparative transcriptomic analyses of primary esophageal epithelial cells exposed to the Campylobacter isolates revealed expression changes within those infected with strains with high intracellular fitness that could explain the increased likelihood of disease progression.ConclusionsWe provide a comprehensive assessment of the esophageal microenvironment, identifying bacterial strain-specific signatures with high relevance to progression of metaplasia.

Dataset Information

CNet: a multi-omics approach to detecting clinically associated, combinatory genomic signatures.

Publications

CNet: a multi-omics approach to detecting clinically associated, combinatory genomic signatures.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets