TissueNexus

TissueNexus constructs tissue- and cell type-specific functional gene networks from a large compendium of RNA-seq datasets to support analysis of gene interactions underlying human physiology and disease.

Key Features:

• Large-scale RNA-seq compendium: Integrates 1,341 RNA-seq datasets comprising 52,087 samples.
• Functional Gene Networks (FGNs): Builds FGNs for 49 distinct human tissue and cell line combinations.
• Advanced machine learning integration: Employs an advanced machine learning approach that captures nonlinear relationships and is reported to surpass traditional Bayesian classifiers for gene network modeling.
• Extensive hand-curation of tissue labels: Performs intensive manual curation of tissue annotations to address inconsistent naming and missing labels in RNA-seq datasets.

Scientific Applications:

• Disease gene prioritization: Prioritizes candidate disease genes by leveraging tissue-specific FGNs to reveal relevant gene interactions.
• Tissue-specific functional analysis: Enables investigation of tissue- and cell type-specific gene interaction patterns to inform mechanisms of complex genetic diseases and potential therapeutic strategies.

Methodology:

Integration of a large RNA-seq compendium (1,341 datasets, 52,087 samples), intensive hand-curation of tissue labels, and construction of tissue- and cell line-specific FGNs using an advanced machine learning approach reported to outperform traditional Bayesian classifiers.