Harnessing the power of cancer whole genome sequencing for clinical utility

Research Question How can we maximise the use of mutational information in whole cancer genomes for clinical benefit?

Background Substantial expertise is required to perform analyses and make clinically-useful interpretations of whole genome sequenced (WGS) cancers.

My team have pioneering bioinformatic skills in this area, with strong foundations in exploratory data analysis, supported by experimental approaches. We have begun to use our knowledge to design computational algorithms that could have predictive/prognostic value.

We have a track record of substantiating these analytical tools through retrospective and prospective clinical studies, to demonstrate patient benefit.

The UK 100,000 Genomes Project (UK100kGP) is the largest WGS endeavour in the world focused on rare diseases, common cancers and infectious diseases.

My team have recently forged links with the UK100kGP, become familiar with their secure data infrastructure (the Research Embassy (RE)), and begun curation of >15,000 of their WGS cancers.

Clearly capable of generating data, the UK100kGP genomics infrastructure could next benefit from the best WGS insights and algorithmic tools, to become thought-leaders in offering WGS as an unbiased, comprehensive, clinically-actionable assay to inform NHS cancer care.

To achieve this potential, we must: Be able to provide holistic WGS cancer genome interpretation containing clinically-useful driver, mutational signature, algorithmic scores and germline genome information Be able to distinguish clinically-relevant mutagenesis for development into new algorithms/biomarkers for prognostic/therapeutic intervention.

Aims & Objectives Aim I: Accelerate clinical translation of computational tools Objective I.1: Develop and implement Steiner-Open-CGA Holistic Cancer Genome Interpreter for wider NHS clinical research use Objective I.2: Maintain and enhance the Mutational Signatures Reference Database Signal Objective I.3: Take specific algorithms through the final steps of implementation of clinical algorithms into clinical trials Aim II: Utilise computational approaches to identify clinically-relevant mutagenesis Objective II.1: Work with already-published cohorts of cancer genomes to make new discoveries Objective II.2: Work on clinical cohorts with treatment and outcome data in order to make clinically-useful interpretations Objective II.3: Understand the contribution of the germline to somatic mutagenesis Aim III: Use experimental approaches on patient-derived pluripotent stem cells and gene-edited human cellular models to gain mechanistic understanding of mutagenesis Objective III.1: Explore mutational mechanisms underpinning on-going mutational signatures because these may be biomarkers/developed into clinical algorithms Objective III.2: Explore tissue specificities and how the genome interacts with other modalities (transcriptome, proteome, metabolome).

Methods For computational work we will utilise: Simple and advanced statistical methods e.g. machine-learning, clustering, latent variable analyses Deep data analysis e.g. nucleotide level exploration to make insights on mutational mechanisms For experimental work, we will utilise: cellular models from DNA repair patients, perform gene editing and/or treat cells with mutagens functional DNA damage response assays, western blotting, proteomics/metabolomics/transcriptomics/sequencing to seek signatures that inform our computational efforts.

Impact Nationally, our tools/insights will be available to all UK100kGP RE users For patients, advances made through WGS are imported rapidly into cancer care For my team and I, NIHR Research Professorship support will be a strong impetus to focus on clinical computational translation and national implementation, substantially increasing our authority and reach