Cancer cells often evolve through a reiterative process of clonal expansion, genetic diversification, clonal selection and adaptation within tumor microenvironments. Clinically, the presence of a small percentage of drug-resistant or tumor-initiation cells representing 1% to 20% of bulk tumor may ultimately determine patients' outcome and survival. The significant intratumoral heterogeneity represents a formidable challenge to the discovery of effective cancer treatments; and therefore dissecting of the tumor heterogeneity holds the key to the development of more effective drugs to control cancer growth and metastasis. Several recent studies have revealed the coevolution of genetic and epigenetic aberrations and highlighted the influential role of epigenetic hierarchy in tumor cell evolution. One of innovative tools that can examine intratumoral epigenetic heterogeneity is NOMe-seq (Nucleosome Occupancy and Methylome Sequencing) or MAPit-BGS, which allow simultaneously profiling of chromatin accessibility and DNA methylation on single molecules. Our preliminary study on analyzing NOMe-seq for one wild type and 3 knock down cells of HCT116 cell lines has revealed that single molecule analysis for chromatin accessibility and DNA methylation together is feasible. However, there is no appropriate bioinformatics program and software that are currently suitable for the large number of genes and genomes. The overarching goal of this project is develop bioinformatics algorithms tools for dissecting the epigenetic heterogeneity of human cancer. Specifically, we will develop bioinformatics algorithms for single molecule- and population-level analysis of chromatin accessibility and DNA methylation (Aim 1); and develop statistical approaches to identify epigenetic heterogeneity in a mixed cell population among different conditions or samples and implement modules and pipelines for chromatin accessibility and DNA methylation (Aim 2).
Clinically, the presence of a small percentage of drug-resistant or tumor-initiation cells may ultimately determine patients' outcome and survival. The objective of this project is to develop bioinformatics algorithms and statistical approaches for single molecule- and population-level analysis of chromatin accessibility and DNA methylation. The results will improve our understanding of the tumor progression with drug treatment, and will aid the refinement of existing cancer diagnosis and identify novel targets.
