Computational approaches to the mechanistic elucidation of the serrated pathway of human colon carcinogenesis

PROJECT SUMMARY This K25 Mentored Quantitative Research Development Award project proposal is to allow the PI, Dr. Marmar Moussa - an experienced Computer Scientist and Bioinformatician - to obtain additional training in cancer genomics and cellular and molecular cancer biology needed to lead an interdisciplinary research lab and prepare the PI to become a fully independent investigator. To that end, Dr. Moussa assembled an outstanding mentoring committee of world-class scientists and physicians from University of Connecticut and Harvard University and provided a comprehensive training plan to meet her career goals. Additionally, Dr. Moussa’s research goal is to develop advanced computational approaches to investigate and model the mechanisms of the alternative serrated pathway of human colon carcinogenesis. This research complements Dr. Moussa’s career development plan and promotes her path to achieve independence. Despite increased colorectal cancer (CRC) prevention efforts in recent years, CRC remains the second leading cause of death from cancer in the United States. CRC develops from a series of genetic and epigenetic changes described by two pathways; the well-studied ’conventional adenoma-carcinoma sequence’ and the less understood ’alternative serrated pathway’. The serrated pathway, characterized by the precursor Serrated Lesions, is increasingly described over the past decade to be contributing to 15 - 35% of CRC tumorigenesis and even more in ’interval’ CRC (I-CRC). It is therefore significant for surveillance and early detection to identify which at-risk lesions progress along this pathway, and how. This is the focus of this proposal. Characterization of the serrated lesion pathway of colon carcinogenesis has been difficult due to the het- erogeneity and the absence of comprehensive longitudinal data in humans. Single cell sequencing is the best tool for studying heterogeneity, and far exceeds the power of histology in this regard. In addition, a wealth of thousands of pre-cancerous and CRC samples is available to this study from the PI’s institution. Dr. Moussa will develop computational approaches to the genomic, and state-of-the-art single cell epigenetic (ATAC-Seq) and transcriptomic analyses of human samples to accurately characterize distinct and subtype-specific phe- notypes and elucidate the mechanisms by which carcinogenesis occurs in the human colon via the serrated pathway (Aim1). Additionally, the PI proposes interrogating the new and groundbreaking spatially-resolved transcriptomics of archival Formalin Fixed Paraffin Embedded (FFPE) samples using novel computational ap- proaches to identify distinct molecular patterns within otherwise histologically similar serrated lesions/polyps (Aim2). Elucidating the mechanisms of precursor lesion progression along the serrated pathway will help im- prove clinical predictability and identify factors that extend prevention well beyond polyp detection and removal.