RAIG

RAIG identifies independent and recurrent somatic copy number aberrations (SCNAs) in cancer genomes using interval graphs and combinatorial optimization.

Key Features:

• Combinatorial Approach: Employs a combinatorial methodology to separate overlapping SCNA events across individuals into independent occurrences.
• Interval Graph Construction: Constructs an interval graph from overlaps between SCNAs to represent their pairwise relationships.
• Maximal Clique Derivation: Derives maximal cliques from the interval graph to represent groups of potentially independent and recurrent SCNAs.
• Dynamic Programming Enumeration: Utilizes a dynamic programming algorithm to enumerate cliques and select non-overlapping clique configurations.
• Objective Function Optimization: Optimizes a well-defined objective function rather than relying on heuristics to identify aberrations.
• Detection of Rare Aberrations: Enables identification of rare but potentially functional SCNAs that may be obscured by larger passenger aberrations.

Scientific Applications:

• Cancer genomics: Identification of recurrent and independent SCNAs to support distinction between driver and passenger mutations and study of tumorigenesis.
• Therapeutic target discovery: Prioritization of recurrent SCNAs that may represent candidate therapeutic targets.
• Benchmarking and evaluation: Demonstrated superior performance on simulated data and real-world cancer genome datasets from The Cancer Genome Atlas (TCGA).

Methodology:

Constructs an interval graph from SCNA overlaps, derives maximal cliques, and employs a dynamic programming algorithm to enumerate cliques and select an optimal set of non-overlapping cliques by optimizing a defined objective function.