CaDrA is an R package that supports a heuristic search framework aimed towards the identification of candidate drivers of oncogenic activity. Given a binary genomic dataset (where the rows are 1/0 vectors indicating the presence/absence of genomic features such as somatic mutations or copy number alteration events), together with an associated sample ranking (where the samples are ranked by a certain phenotypic readout of interest such as protein expression, pathway activity etc.), CaDrA implements a step-wise search algorithm to determine a set of features that, together (based on their occurence union or 'logical OR'), is most-associated with the observed ranking, making it useful for finding mutually exclusive or largely non-overlapping anomalies that can lead to the same pathway phenotype.
0 commit comments