Relabeling metabolic pathway data with groups to improve prediction outcomes

Abstract

We present reMap (relabeling multi-label pathway data based on bag approach), a simple, and yet, generic framework, that performs relabeling examples to a different set of labels, characterized as bags. A bag is comprised of a subset of correlated pathways, and a pathway is allowed to be mixed over bags, constituting an overlapping pathway over a subset of bags. Bag based approach was followed to overcome low sensitivity scores of triUMPF for the pathway prediction task. The relabeling process in reMap is achieved by alternating between 1) assigning bags to each sample and 2) updating reMap’s parameters. reMap’s effectiveness was evaluated on metabolic pathway prediction where resulting performance metrics equaled or exceeded other prediction methods on organismal genomes with improved sensitivity score.Availability and implementationThe software package is published on github.com/reMap.Contactshallam@mail.ubc.ca