RT Journal Article
SR Electronic
T1 Predictive Computational Phenotyping and Biomarker Discovery Using Reference-Free Genome Comparisons
JF bioRxiv
FD Cold Spring Harbor Laboratory
SP 045153
DO 10.1101/045153
A1 Alexandre Drouin
A1 Sébastien Giguère
A1 Maxime Déraspe
A1 Mario Marchand
A1 Michael Tyers
A1 Vivian G. Loo
A1 Anne-Marie Bourgault
A1 François Laviolette
A1 Jacques Corbeil
YR 2016
UL http://biorxiv.org/content/early/2016/03/27/045153.abstract
AB 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. The method is computationally scalable and well-suited for whole genome sequencing studies. The method was validated by generating models that predict the antibiotic resistance of C. difficile, M. tuberculosis, P. aeruginosa and S. pneumoniae. We show that the obtained models are accurate and that they highlight biologically relevant biomarkers, while providing insight into the process of antibiotic resistance acquisition.