Human-associated microbial communities have been implicated in a variety of chronic diseases,

including inflammatory bowel diseases, obesity, and autoimmune disorders like diabetes. Environmental communities are also important for bioconversion of waste products in biofuel

production. However, microbiomes are highly complex systems involving mutualism and competition between many constituent organisms, and a variety of fundamental and interesting theoretical challenges remain in the modeling of pathogenicity and community-wide response to perturbations. In this talk I will discuss several computational and statistical approaches to predictive modeling of microbiome behavior using high-throughput metagenomic and transcriptomic sequencing data, including models that leverage biological structures such as gene ontologies, phylogenies, and metabolic pathways to extract features and constrain model complexity.



Dan Knights is a Ph.D. candidate in Computer Science at the University of Colorado, with a certificate in Interdisciplinary Quantitative Biology and a focus on computational biology and machine learning. In general, he is interested in using machine learning and computational statistics to answer  important questions in biology, genomics, and engineering. His current research focus is the redictive modeling of complex microbial communities using multi-source high-throughput sequencing data, with applications in personalized medicine, early diagnosis of disease, and biofuel production. He has co-authored 16 journal publications during his graduate studies, including two papers in Science. His projects have received international media attention, and have been declared “must-read” by the Faculty of 1,000, listed in Cell Host & Microbe’s “Most Read Articles”, and selected runner-up for the Science “Breakthrough of the Year”. He has been the expert witness on National Public Radio’s “Wait Wait…Don’t Tell Me!”, and was the official 2003 Rubik’s Cube World Champion.


Faculty Host: Adam Siepel/Eva Tardos


B17 Upson Hall

Thursday, February 23, 2012

Refreshments at 3:45pm in the Upson 4th Floor Atrium


Computer Science


Predictive Modeling of


Dan Knights

(Univ of Colorado-Boulder)