It takes guts to learn: machine learning techniques for disease detection from the gut microbiome

Abstract

Associations between the human gut microbiome and expression of host illness have been noted in a variety of conditions ranging from gastrointestinal dysfunctions to neurological deficits. Machine learning (ML) methods have generated promising results for disease prediction from gut metagenomic information for diseases including liver cirrhosis and irritable bowel disease, but have lacked efficacy when predicting other illnesses. Here, we review current ML methods designed for disease classification from microbiome data. We highlight the computational challenges these methods have effectively overcome and discuss the biological components that have been overlooked to offer perspectives on future work in this area.

Publication
Emerging Topics in Life Sciences (2021). [DOI:10.1042/ETLS20210213]
Kristen Curry
Kristen Curry
PhD student

Kristen (2nd year PhD student) obtained a bachelor’s in computer science from the University of California, Berkeley. Prior to enrolling at Rice, she worked for a private biotech company focused on generating personalized health information from blood protein levels. Her primary research interests are microbial interactions and their impact on host health. Outside of the lab, she enjoys backpacking, running and practicing yoga.

Dr. Mike Nute
Dr. Mike Nute
Postdoctoral Scientist

Mike (Postdoctoral Scientist) received his Ph.D. in Statistics in 2019 from the University of Illinois at Urbana-Champaign where he was advised by Dr. Tandy Warnow in the Department of Computer Science and worked on algorithms related to multiple sequence alignment and phylogenetic tree estimation, in particular applying these methods to studying microbial communities. He was co-advised by Dr. Rebecca Stumpf in the Department of Anthropology where he and other lab members developed novel methods to compare the microbiomes of human and non-human primates. His research interest is in discovering a new applications for our understanding of microbial communities.

Todd J. Treangen
Todd J. Treangen
Assistant Professor of Computer Science

My research interests include algorithms and data structures for efficient analysis of microbial genomes and metagenomes

Next
Previous