With the advent of sequencing-based genomics assays, we now have thousands of measurements of the biological activity at each genomic position. My research focuses on the development of machine learning methods that leverage these data sets to improve our understanding of the human genome. I will present reference genome annotations of 164 human cell types, a study of chromatin domains through joint analysis of chromatin state and chromatin conformation, and an optimization-based approach for determining the most informative panel of genomics assays.
Understanding human genome regulation through unsupervised machine learning
Friday, March 17, 2017 - 12:00pm
Carl Icahn Lab 101
Lewis-Sigler Institute for Integrative Genomics