Cancer Genomics: Integrative and Scalable Solutions in R/Bioconductor

Abstract Bioconductor is an ecosystem of more than 1,500 open-source software and data packages for the statistical analysis and comprehension of high-throughput genomic data. It is widely used by the cancer genomics research community for statistical analysis and visualization. This software ecosystem is supported by core

data classes and methods, reused by both users and developers, that provide convenient representations and efficient operations for many kinds of high-throughput molecular data. Falling sequencing costs and single-cell assays enable increasingly resolved study of the molecular biology of cancer, through combined assaying of

DNA sequence, epigenetics, gene expression, protein, and other aspects, even at the single-cell level, for a single specimen. These developments present new challenges in complexity, size, and interpretability of the data. The overarching goal of this project is to create and adapt core Bioconductor software infrastructure to

meet these challenges, through the following aims. First, we develop infrastructure for the analysis of single-cell multi-omic experiments. Second, we implement FAIR principles for improved somatic variant prioritization, by defining performant data architecture that harmonizes and integrates the large amount of

experimental and annotation data available through Bioconductor. Users of our system will be able to create provenance-rich interoperable reports on structural and functional contexts of somatic variants for use in prioritization. Third, we develop scalable infrastructure for the curation, distribution, maintenance,

discoverability, and usability of cancer data resources within and externally to Bioconductor. Finally, we develop a program of user training and new outreach approaches to support adoption of advanced Bioconductor infrastructure by developers of new cancer-related packages and existing packages critical to the

cancer research community.