DEVELOPMENT OF INFORMATICS RESOURCES FOR INTERPRETATION OF CLINICALLY ACTIONABLE VARIANTS IN CANCER

点击翻译按钮获取中文摘要

Abstract Clinical cancer sequencing will increasingly be used to identify genomic alterations that are relevant to understanding cancer progression and improving clinical decision making for individual patients. Currently, the most critical bottleneck in the precision medicine workflow is at the interpretation step. However, there are few resources to help with the prioritization and interpretation of these alterations in a clinical context. Multiple groups are building their own databases documenting clinical interpretation of tumor mutations as they are observed in those groups. This represents a largely redundant effort with no mechanisms for capturing evolving evidence from the biomedical literature. Public knowledgebases are needed with sophisticated application programming interfaces (APIs) that allow rapid intersection of genomic alterations with interpretations of their clinical actionability. The goal of this proposal is to develop such an expert-curated knowledgebase, web interface and API for Clinical Interpretation of Variants in Cancer (CIViC - www.civicdb.org). The knowledge created from this effort will be freely available and the product of open discussion across a diverse community. This will require an interdisciplinary approach to combine the expertise of genome scientists and cancer researchers, whose efforts are otherwise often isolated. Content will be created with transparency, kept current, be comprehensive, track provenance, and acknowledge the efforts of creators. It will cover all types of alterations from single nucleotide variants to structural variants, RNA fusions, expression changes, epigenetic alterations and others. The interface will capture both structured statements of evidence for actionability to allow computational data mining and also human-readable interpretations. Content and software will be unencumbered and easy to access to encourage both academic and commercial engagement. In order to achieve these goals CIViC will require well-designed data standards, use of structured vocabularies, and a user-friendly curation interface that balances data mining needs with human accessibility. Focused `hackathons' and curation meetings will be organized to establish community standards and coordinate with synergistic efforts by the Global Alliance for Genomics and Health (GA4GH) and ClinGen working groups. The CIViC resource will serve as the foundation for development of applications critical to widespread implementation of precision medicine for cancer. Specifically, it will facilitate rapid generation of targeted sequencing assays, automated clinical report generation, and other applications built upon the open CIViC standards and APIs.