Metric Learning on Expression Data for Gene Function Prediction

Motivation: Co-expression of two genes across different conditions is indicative of their involvement in the same biological process. However, using RNA-Seq datasets with many experimental conditions from diverse sources introduces batch effects and other artefacts that might obscure the real co-expression signal. Moreover, only a subset of experimental conditions is expected to be relevant for finding genes related to a particular Gene Ontology (GO) term. Therefore, we hypothesize that when the purpose is to find similar functioning genes that the co-expression of genes should not be determined on all samples but only on those samples informative for the GO term of interest. Results: To address both types of effects, we developed MLC (Metric Learning for Co-xpression), a fast algorithm that assigns a GO-term-specific weight to each expression sample. The goal is to obtain a weighted co-expression measure that is more suitable than the unweighted Pearson correlation for applying Guilt-By-Association-based function predictions. More specifically, if two genes are annotated with a given GO term, MLC tries to maximize their weighted co-expression, and, in addition, if one of them is not annotated with that term, the weighted co-expression is minimized. Our experiments on publicly available Arabidopsis thaliana RNA-Seq data demonstrate that MLC outperforms standard Pearson correlation in term-centric performance. Availability: MLC is available as a Python package at www.github.com/stamakro/MLC .