eVIP2

eVIP2 predicts the functional impact of gene variants from gene expression data to classify variant effects and identify pathway-level consequences.

Key Features:

• Functional Impact Prediction: eVIP2 employs a decision tree-based algorithm on gene expression data comparing signatures induced by wild-type versus mutant open reading frames (ORFs) to classify variants as gain-of-function, loss-of-function, change-of-function, or neutral.
• Pathway Analysis Capability: The eVIP Pathways module performs pathway-level analysis to identify specific biological pathways altered by variants.
• RNA-seq Integration: The method supports RNA-seq (RNA sequencing) data as input for expression-based analyses.
• Application Example: eVIP2 was used to characterize recurrent RNF43 frameshift variants and predicted RNF43 G659fs as gain-of-function with alterations in TNF alpha via NFKB signaling, KRAS signaling, and hypoxia pathways.
• Validation of Predictions: Predictions from eVIP2 analyses were validated by reporter assays confirming activation of predicted pathways.

Scientific Applications:

• Precision Medicine: Supports interpretation of somatic mutations to inform targeted therapy decisions by elucidating variant functional effects.
• Cancer Genomics Research: Enables assessment of functional consequences of variants in cancer-associated genes (for example, RNF43) and investigation of pathway-level perturbations.

Methodology:

eVIP2 applies a decision tree-based algorithm to gene expression data by comparing signatures induced by wild-type versus mutant ORFs, classifies variants into gain-, loss-, change-of-function or neutral categories, and performs pathway-level analysis via the eVIP Pathways module using RNA-seq input.