aldro61/kover

Kover is an out-of-core implementation of the Set Covering Machine algorithm that has been tailored for genomic biomarker discovery. It produces highly interpretable models of phenotypes. The models are rule-based and rely on the presence/absence of k-mers. The identification of genomic biomarkers is a key step towards improving diagnostic tests and therapies. We present a new reference-free method for this task that relies on a k-mer representation of genomes and a machine learning algorithm that produces intelligible models.