Predicting Transcriptional and Epigenetic Networks in Cancer from Sequencing Data

点击翻译按钮获取中文摘要

Limitless replicative potential is a key hallmark of cancer and critically depends on telomere maintenance. Many cancers thus aberrantly reactivate the telomerase reverse transcriptase (TERT), a catalytic subunit of the telomerase complex that elongates telomere. It has been recently discovered that this common path to immortality in multiple cancers is through two activating point mutations in the TERT promoter (TERTp), found in more than 50 different cancer types, often at strikingly high frequencies, e.g. roughly 83% in glioblastomas (GBM) and 71% in melanomas. In the previous funding period, the PI has identified the molecular function of these highly recurrent mutations, demonstrating that the transcription factor (TF) GABP binds the mutant TERTp with exquisite specificity, but not the wild-type TERTp. The high prevalence of TERTp mutations across multiple cancer types and the selectivity of GABP recruitment to mutant TERTp thus provide an unprecedented opportunity for treating a large number of cancer patients with minimal toxicity to healthy cells. Despite the clear significance of this opportunity, however, several important questions surrounding the molecular functions and modulators of TERTp mutations remain poorly understood, hindering the development of effective and safe therapeutic strategies. Our long-term goal is to establish a rigorous computational framework for understanding the aberrant transcriptional and epigenetic networks in cancers and to apply the resulting knowledge to devise novel therapeutic strategies that account for the genetic background of individual patients and that can a priori predict and avoid potential resistance mechanisms. The objective of our current renewal proposal is to develop powerful computational methods for transforming our knowledge about the non-coding TERTp mutations into an effective and safe molecular target. At the same time, the resulting methods will help resolve several outstanding challenges in the field of transcriptional gene regulation and have broad applications in cancer genomics. We will accomplish our objective my pursuing the following Aims: (1) Develop and test a computational framework for inferring sequence features that determine the distinct and shared binding patterns of paralogous TFs; (2) Develop and validate integrative tools for discovering the molecular basis of genetic interactions between germline variations and oncogenic mutations; (3) Develop and apply computational methods for studying the role of DNA helical phase between adjacent binding motifs in recruiting ETS factors to chromatin; (4) Perform a systematic genomic characterization of the effects of knocking out GABPB1L in TERTp-mutant cancer cells and healthy cells. The results of this proposal will have a broad impact on cancer research by providing powerful tools for studying paralogous oncogenic TFs and revealing novel insights into a highly promising therapeutic strategy.