Paloma Ruiz, a junior quantitative biology major at the University of North Carolina at Chapel Hill, analyzed the exclusive tendencies of cancer gene pairs between two datasets.
When analyzing tumor development, there are certain tendency patterns that occur among driver genes, which are genes that actively contribute to and benefit from cancer tumor growth when mutated. These cancer gene pairs can exhibit patterns of co-occurrence, where mutations occur together within a specific tumor, or mutually exclusivity, where mutations do not occur simultaneously. BSRP 2020 has been a wonderful and transformational experience in defining my identity as a scientist. As a quantitative biology major with mostly wet-lab experience, I am very grateful for the opportunity to develop computational skills that will be useful back at my home university and in my scientific career. However, the welcoming and collaborative environment at the Broad has been the most impactful aspect of my summer experience. I am constantly inspired by the work of my cohort, and look forward to see their accomplishments in the future. The relationship among gene pairs is important to understand when developing cancer therapeutics, as certain tumor types cause immunotherapy to be less effective based on the tendency pattern of gene pairs. The standard tool currently used for cancer research is the Cancer Cell Line Encyclopedia (CCLE), a database comprised of mutation analyses of human cancer cell lines. For our project, we wanted to examine the tendency patterns of gene pairs within the CCLE, and analyze how the results found compare to actual patient data, obtained from The Cancer Genome Atlas (TCGA). We looked at the most frequently mutated genes and categorized the tendencies of gene pairs within each dataset. From our analysis, we found that, while both the CCLE and TCGA demonstrated gene pair tendencies of co-occurrence, the results compiled from the CCLE were statistically insignificant due to an insufficient amount of data. Thus, in order to arrive to a conclusion on the relationship of cancer cell line data within the CCLE to patient data in TCGA, we must obtain more cell line samples. Furthermore, we hope to examine expression data in order to understand the biological function of each gene within a gene pair, which can provide more insight on the correlation between cell line and patient data.
Project: Mutual Exclusivity between Cancer Genes in Cancer Cell Lines and Patient Data
Mentors: Liang Chang, William Sellers, Sellers Lab