Deciphering Information Encoded in the Dark Matter of the Human Genome

Among the three billion bases contained in the human genome, only 1.5% are well characterized, primarily in the form of protein-coding genes. Understanding all functional elements encoded in the remaining 98.5%, or the so-called dark matter, of the human genome is a fundamental challenge for human biology and medicine.

In this talk, I will describe computational methods for systematically dissecting the function of the genomic dark matter. Using pattern recognition, machine learning and comparative genomics, we have uncovered hundreds of novel regulatory motifs in these regions, creating a first systematic catalogue of both transcriptional and post-transcriptional regulatory elements in the human genome. Our systematic analyses also lead to a fundamental change of view on microRNA gene regulation, and the discovery of over 15,000 insulator sequences, which partition the human genome into domains of expression.

In a few years, genome sequences of over 50 mammals will become available. I will discuss how these data will empower our computational methods, and provide an opportunity to unravel all information encoded in the human genome.

BIO: Xiaohui Xie is a computational biologist working at the Broad Institute of MIT and Harvard. His recent research interests are in computational genomics and regulatory motif discovery in particular. Dr. Xie received his M.S. in Computer Science and Ph.D. in Computational Neuroscience, both from MIT, where he continued as a postdoc with Eric Lander.