Public Health Sciences
Genomic analysis of colorectal cancer risk
My research focuses on unravelling the complex genetic mechanisms underlying risk of colorectal cancer in order to enhance disease surveillance leading to improved health outcome. Our approach integrates large-scale human genetic association data with epigenetic and transcriptomic profiling, and incorporates genome editing and in vitro human 3D normal colon organoid models to better understand causal mechanisms of disease.
1. From risk association to causal genetic variant:
Our ongoing genome wide association studies of colorectal cancer have led to the identification of a growing number of inherited risk associated genetic variants. Our lab is interested in discovering the gene targets of these associations as well as the mechanisms underlying gene regulation, leading to a better understanding of the biology of disease initiation. The challenge is that few of these risk variants are themselves functional - the causal variants are in linkage disequilibrium with the associated risk variants and most map to enhancer elements and other regulatory regions of the genome. Genome wide epigenomic data including ChIP-Seq data are being used to identify candidate risk enhancers followed by enhancer validation using in vitro luciferase enhancer assays and site directed mutagenesis. Candidate gene targets are being identified through genome wide transcriptomic analyses of normal colon tissues employing RNA-Seq and CRISPR-Cas9 genome editing of risk enhancers. Finally, risk enhancer-gene target validations are being studied using chromosome conformation capture approaches (e.g. 3C, Capture C). Much of our work relies on the application of integrative computational pipelines.
2. From association to biology in colorectal cancer:
We have developed human 3D normal colon organoid models in which to model inherited genetic factors in the early stage development of colon cancer. For example, lentiviral knock down of genes such as the APC gene leads to increased proliferation and morphologic changes in 3D organoids that recapitulate what is seen in organoids developed from patients with Familial Adenomatous Polyposis cancer. We propose to use these models not only to interrogate the role of genes involved in colorectal cancer risk on 3D proliferation, apoptosis and morphology, but also to investigate the effect of environmental agents that are known to influence risk of colorectal cancer on gene expression using RNA-Seq. We are also interested in understanding how risk variants and risk genes affect global genome-wide epigenetic profiles and will use these models to map differential allele-specific chromatin states using ATAC-Seq and gene expression. These approaches should help us begin to clarify the functional interaction between common regulatory variants, the epigenetic status of cells, transcription factor binding and gene expression in different cellular environments.
3. Whole genome sequencing of individual colon crypts:
We hypothesize that one mechanism whereby risk variants can influence disease risk is through premature aging of colon crypt stem cells. All of the cells of a human crypt are renewed every 4-5 days, meaning that each crypt is essentially clonal and reflects the status of the stem cell(s). We are testing whether risk variant burden correlates with an increased accumulation of DNA mutation and epigenetic changes in individual crypts using whole genome sequencing and epigenetic profiling. Our studies will provide information on the contribution of risk variants to colon crypt cellular aging, through assessment of the frequency and types of DNA mutations and altered epigenetic profiles in colon crypt stem cells.