3D Cancer Genome Modeling for Personalized Medicine Diagnostics

Fremstillingen allerede oppdatert 11.12.18 og lagt inn

-

Cancers arise from mutations in a subset of high-risk genes, whose localization in the genome seems to be related to the underlying structure of chromatin. We seek to develop a computational framework for accurate estimation of cancer mutation susceptibility in the human genome, and following H2020 work, turn this platform into a cancer classification tool for use in precision medicine diagnostics and therapeutics. To this end, we have formed a Norwegian consortium of 3 partners at UiO (Philippe Collas; nuclear architecture, genome modeling, high-performance computing), OUS (Vessela Kristensen; breast cancer genetics, large dataset integration) and NTNU (Pål Sætrom, machine learning in genomics, gene regulation networks). We will establish, from our own genome modeling platform (Chrom3D), high-resolution 3-dimensional (3D) models of the human genome in cancer cells to infer relationships between 3D chromatin structure and mutation susceptibility. By modeling this relationship, we will then be able to generate probabilistic models of tumorigenic DNA variation; these models will constitute the basis for a highly performant tumor-typing software. The most relevant H2020 Thematic Area will be "Personalized medicine", with several putative calls already identified (see Annex). Our goals for this two-year project are ambitious, but we have already set up a working 3D genome modeling framework (Chrom3D) on which we are going to expand. We will specifically here: (1) Enhance the performance and accuracy of our 3D genome modeling platform; (2) Acquire preliminary data on 3D cancer genomes for, as model systems, melanoma and breast cancer cell lines; (3) Identify H2020 opportunities and new partners, formulate work packages, to send at least two H2020 ERC applications and at least one Thematic Area application in the 2017-2018 period. Regular meetings among PIs and with H2020 partners will also ensure progress, networking and writing of H2020 applications.