GMSCA

GMSCA predicts gene multifunctionality by leveraging secondary co-expression networks to assign functions and cell-type contexts to genes from bulk-tissue brain gene expression data.

Key Features:

• Secondary co-expression networks: Uses secondary co-expression networks to expand the functional annotations associated with genes.
• Enhanced gene function prediction: Increases the number of predicted triplets (gene, function, cell type) for individual genes.
• Functional coherence and cell-type specificity: Assigns functional coherence and cell-type specificity to a larger fraction of genes used in co-expression network construction.
• Bulk-tissue expression input: Operates on co-expression networks derived from bulk-tissue gene expression profiling.
• Application to brain tissues: Applied to 27 co-expression networks from various brain tissues, including neurons and glial cells (microglia, astrocytes, oligodendrocytes).
• Novel cell-type assignments: Identifies additional roles for genes traditionally linked to specific cell types, exemplified by predicting SNCA function in oligodendrocytes.

Scientific Applications:

• Disease research: Uncovers multifunctional gene roles across cell types to inform disease mechanisms and potential therapeutic targets.
• Neuroscience studies: Enables investigation of complex cellular interactions in the brain by providing cell-type-resolved functional annotations from bulk-tissue data.

Methodology:

Builds upon existing co-expression networks derived from bulk-tissue gene expression profiling and systematically identifies and assigns additional functions and cell-type contexts to genes using secondary co-expression network analysis.