Bioinformatics and Genomic Networks

The paper identifies DNA damage-associated protein co-expression modules in cardiomyocytes and shows their relationship to genetic variation tolerance and CVD risk.

GREA improves gene set enrichment analysis by incorporating gene interactions, outperforming existing tools in detecting pathways related to respiratory viral infections.

This paper introduces SGTCCA-Net, a framework for multi-omics network inference that addresses limitations of existing methods by incorporating higher-order correlations and improving computational efficiency.

This paper introduces a computational framework for predicting Alzheimer's protein interactions using graph-based learning, spiking neural networks, and hyperdimensional encoding.

This paper introduces a method to improve protein-protein interaction prediction by using topology-driven negative sampling and an ML pipeline that enhances generalizability and interpretability.

This paper introduces GGOT to detect critical transitions and identify trigger molecules in disease progression by analyzing data distributions and transport maps across disease stages.

RNAchat integrates machine learning with clinical and multi-omics data to identify metapathways and cellular interactions, improving understanding of disease mechanisms like RA and HF.

RNAcompare integrates clinical and multi-omics data using machine learning to identify phenotypic similarities in rheumatoid arthritis and heart failure, revealing shared molecular pathways and age-related subgroups.

GeneCOCOA identifies context-specific functions of individual genes using co-expression data and curated functional gene sets, outperforming existing methods in gene-disease association recall.

miss-SNF integrates multimodal patient data with missing sources by recovering patient similarities and identifying clinically relevant subgroups.

The paper consolidates RNA-seq and microarray data to infer gene-pathway associations in Tetrahymena thermophila, identifying known and novel functionally related genes through co-expression clustering.

The paper evaluates sequence and structural similarity metrics for predicting shared paralog functions using synthetic lethality and protein-protein interactions.

The paper systematically compares pathway databases by querying them with genes, pathways, and conditions, finding that PathDIP and Reactome perform best depending on the query type.

This paper introduces 3D IntelliGenes, a tool for visualizing AI/ML results on multi-omics data to discover biomarkers and predict diseases through 3D clustering and feature analysis.

This study uses gene set embeddings to uncover biological function associations across species by integrating node2vec with MSigDB data.

The paper constructs a comprehensive liver-specific gene interaction network using transcriptomic data from multiple species to enable unbiased analysis of liver disease pathways.

This study uses a single-cell network approach to analyze metabolic regulation in gynecologic and breast cancers, identifying key transcriptional regulators and metabolic pathways involved in tumor heterogeneity.

The paper develops scNET to infer cell-type-specific co-expression networks from single-cell data, enhancing the human interactome for improved disease gene prediction.

SEHI-PPI predicts human-influenza protein-protein interactions using a double-view deep learning framework with adaptive negative sampling, outperforming existing methods and demonstrating strong generalizability.

EasyPubPlot is a user-friendly Shiny web app and R package that simplifies omics data visualization, generating publishable plots with minimal steps.

The paper identifies pathway redistribution across cellular states using a framework combining co-expression data and deep learning, revealing shared signaling modules and context-dependent regulatory networks in RNA-binding proteins.

This paper identifies a flaw in PPV and ACC calculations for protein complex detection and proposes revised metrics, demonstrating improved reliability through benchmark experiments.

The paper predicts pathway involvement of unannotated compounds using Reactome data, achieving higher accuracy than KEGG-based models.

The paper introduces CoPPIs, an algorithm that identifies differentially correlated functional modules by combining PPIs and co-expression networks, applied to proteomic data from Parkinson's disease and GBA1 mutation cases.

MOSim generates synthetic bulk and single-cell multi-omics datasets for benchmarking methods that analyze regulatory networks across various omics layers.

The paper introduces a motif mining approach to identify higher-order biomarkers from expression data, showing improved performance in classification and clustering compared to traditional methods in a COVID-19 study.

The paper identifies common gene regulatory signatures across multiple inflammatory skin diseases using an integrative systems biology framework and proposes potential drug targets.

ORCO is an unsupervised network-based tool that computes Ollivier-Ricci curvature to analyze biological network robustness and identify therapeutic targets.

PETS simulates drug-gene interactions using multi-scale network propagation to identify and rank drug candidates for reversing disease-specific gene expression signatures.

evoGO minimizes redundancy in GO enrichment analysis by considering the GO hierarchy, improving specificity while preserving relevant biological information.