Ira M. Hall
Primary AppointmentAssociate Professor, Biochemistry and Molecular Genetics
Bioinformatics; Genome Variation; Genomic Instability; Cancer Genome Evolution
Research in my laboratory is focused on understanding the mechanistic origins and functional consequences of genome variation in mammals. We are primarily interested in structural variation (SV), which includes duplications, deletions, inversions, insertions, translocations and other forms of genomic rearrangement. These large scale genetic differences comprise a significant fraction of inherited human genetic variation, are increasingly found to underlie sporadic disease, and are a driving force in the evolution of cancer genomes. Somatically-acquired structural variants may also contribute to phenotypes that emerge during development and aging.
Our understanding of structural variation has been, and remains to be, limited by technology. Due to the profound experimental and computational challenges associated with performing unbiased genome-wide experiments, many fundamental questions regarding the origin and impact of structural variation remain unanswered.
We are currently pursuing three general lines of investigation. First, we are developing methods to map and interpret genome variation from next-generation sequence data. These include algorithms to map structural variation by paired-end mapping, split-read mapping and depth of coverage analysis, as well as a suite of software tools for comparing raw sequence data, annotations and variant calls. We are also developing computational methods for optimal split-read alignment, haplotype-aware local assembly, and breakpoint visualization. Finally, we are working on experimental methods to map SV in small populations of cells, which will allow for a direct assessment of genetic diversity in somatic lineages.
Second, we are using these methods to tackle basic questions regarding structural variation in normal germline and somatic genomes. While our initial work on these topics has used the laboratory mouse as model, the recent explosion of human genome sequence data has allowed us to transition most of our work to human. Questions that we are interested in include: what are the mechanisms that generate new SV? To what extent are these mechanisms under cellular control, or affected by stress, and thus variable depending on genetic background or environmental conditions? What is the contribution of structurally unstable and/or hyper-variable loci to natural variation? How prevalent is de novo SV in different somatic lineages and single somatic cells, and to what extent can this phenomenon account for traits that emerge during the course of development or aging?
Finally, we are studying the process of tumor genome evolution by comparing the patterns and mechanisms of genomic rearrangement in cancer cells relative to those found in normal cells. We are mapping and assembling structural variant breakpoints from hundreds of human whole-genome sequence datasets, comparing the breakpoint properties of germline versus somatically-acquired structural variation, and attempting to understand the mutational forces that shape highly rearranged tumor genomes. It is our hope that these experiments will lead to new insights into tumor genome evolution and perhaps suggest novel genome-based diagnostic tests that correlate with malignancy.