Identifying non-coding drivers of cancer

Division Of Basic Sciences - Nci

Investigators

Linked publications, trials & patents

Abstract

We have made major progress in the developing the methods and tools and showing its efficacy in identifying mutations in the non-coding regions of the genome during human diversification from other great apes that presumably have created novel regulatory enhancers active in the developing prefrontal cortex. Using deep learning model of embryonic neocortical enhancers, and human and macaque embryonic neocortex H3K27ac data, we identified 4000 enhancers gained de novo in the human, largely attributable to single-nucleotide essential mutations. The genes near de novo gained enhancers exhibit increased expression in human embryonic neocortex relative to macaque, are involved in critical neural developmental processes, and are expressed specifically in the progenitor cells and interneurons. The gained enhancers, especially the essential mutations, are associated with central nervous system disorders/traits. Integrative computational analyses suggest that the essential mutations establish enhancer activities through affecting binding of key transcription factors of embryonic neocortex. Overall, our results suggest that non-coding mutations may have led to de novo enhancer gains in the embryonic human neocortex, that orchestrate the expression of genes involved in critical developmental processes associated with human cognition. This work is published in Science Advances. Encouraged by the effectiveness of the method in prioritizing non-coding mutations during evolution, we are now in the process of evaluating the non-coding mutations in cancer. The challenge has been identifying matched healthy and tumor whole genome sequence data in sufficient numbers. We have now identified 10 such matched datasets in Esophaegeal cancer. Also, for lack of whole genome sequencing data, we have utilized the H3K27AC ChIP-seq data in these samples to identify non-coding mutations. We have also initiated a new collaboration to prioritize race-specific polymorphisms in the enhancer regions to understand the racial disparity in prostate cancer incidence. These efforts are still in the early stages. Additional activity under this projects involves a collaboration where we analyze the spatial organization ofTranscription factor (TF) binding in the nucleus. We find that in multiple cell line contexts, TFs form two groups such that TFs within a group 'attract' each other and those across the groups 'repel'. The same trend is also seen in the linear organization of the TF binding on the genome, suggesting a co-evolution of the linear genome with the spatial organization. Attractive TF pairs exhibit significantly more physical interactions suggesting an underlying mechanism. The two TF groups differ significantly in their genomic and network properties, as well in their function-while one group regulates housekeeping function, the other potentially regulates lineage-specific functions, that are disrupted in cancer. We also show that weaker binding sites tend to occur in spatially interacting regions of the genome. Our results suggest a complex pattern of spatial cooperativity of TFs that has evolved along with the genome to support housekeeping and lineage-specific functions. This work is accepted for publication in Cell journal "Heliyon".