CaDrA

CaDrA performs stepwise and backward heuristic search to identify candidate (epi)genetic drivers from a binary multi-omics dataset based on the union occurrence (i.e. logical OR) of events that are significantly skewed or enriched relative to a given vector of continuous scores.

Key Features:

• Heuristic search algorithms: Performs both stepwise and backward heuristic search strategies to identify candidate driver features.
• Union occurrence scoring: Evaluates feature combinations based on the logical OR of binary event occurrences across samples.
• Multi-omics data support: Analyzes binary multi-omics datasets to identify (epi)genetic drivers.
• Continuous score integration: Identifies features significantly skewed or enriched relative to sample-specific continuous scores such as protein expression or pathway activity.

Scientific Applications:

• Candidate driver identification: Discovers (epi)genetic driver features associated with phenotypic readouts of interest.
• Phenotype association analysis: Links multi-omics alterations to continuous phenotypic scores such as protein expression levels or pathway activity measurements.

Methodology:

CaDrA employs stepwise and backward heuristic search algorithms on binary multi-omics data, scoring feature combinations by their union occurrence (logical OR) and evaluating enrichment against a vector of continuous sample-specific scores.