Precision medicine refers to the use of prevention and treatment strategies that are tailored to the unique features of each individual and their disease. In the context of cancer this might involve the identification of specific mutations shown to predict response to a targeted therapy. The biomedical literature describing these associations is large and growing rapidly. Currently these interpretations exist largely in private or encumbered databases resulting in extensive repetition of effort. Realizing precision medicine will require this information to be centralized, debated and interpreted for application in the clinic. CIViC (www.civicdb.org) is an open access, open source, community-driven web resource for Clinical Interpretation of Variants in Cancer. CIViC’s goal is to enable precision medicine by providing an educational forum for dissemination of knowledge and active discussion of the clinical significance of cancer genome alterations.

A hackathon and curation jamboree will be held as a pre-meeting of ASHG, lead by the CIViC project team, in collaboration with the ClinGen Somatic working group, and hosted by the Scripps Research Institute. The focus of this meeting will be to continue to develop community consensus on data standards for interpretation of variants in cancer, engage software developers for improved variant interpretation software, engage clinical experts for curation of content and enhance integration and interoperability of resources. Additional related topics will be proposed by meeting participants.

Meeting Goals can include (but are not limited to):

• Establish community standards and best practices for curation and representation of clinically actionable variants in cancer.

• Further curation of clinically actionable variants in cancer.

• Further development of the CIViC web interface for expert crowdsourcing the curation of clinically actionable variants.

• Promote formation of a collaborative network for community-based development of precision medicine resources.

• Enhance integration and interoperability of resources through standardization of entities, improvement of APIs, and use of ontologies.

• Improve drug normalization in variant interpretation knowledgebases.

• Further develop text-mining strategies for identifying evidence of clinically actionable variants (e.g., CIViCmine).

• Improved/enhanced integration of MyVariant.info into CIViC.

• Integrate ClinGen Allele Registry’s variant normalization into MyVariant.info.

• Standard operating procedures for developing somatic assertions.

• Development of standard operating procedures for classifying somatic variant oncogenicity.

• Develop prototype data models for representing compound/complex variants.

• Examine ways to incentivize the curation community.

• e.g. Zenodo, bioCADDIE for citable records

• Additional meeting goals will be proposed by meeting participants.

Meeting format

The Cancer Variant Interpretation hackathon and jamboree will take the form of an unconference with minimal presentations and an emphasis on hands-on activities, breakout sessions to discuss emerging standards, learning to curate variants, use the API (http://griffithlab.org/civic-api-docs/) and CIViCpy (git.io/civicpy) to extract data from CIViC, or work on other related related tools for variant interpretation. Participants will self-form into groups to tackle domain-specific variant curation (jamboree), programmatic data querying (hackathon), the problem of motivating public resource curation, or other topics proposed by meeting participants.

Pre-requisites

Participants should bring their own laptop in order to participate.

Target audience

Hackathon: Software developers, engineers, computational biologists, and bioinformaticians interested in learning how to interact with CIViC or work on related tools to facilitate variant interpretation. Prerequisites: Familiarity with Ruby or Python would be beneficial.

Curation Jamboree: Pathologists and oncologists (certified or in training), genome scientists, and those interested in the problem of sequence variant interpretation for cancer precision medicine or promotion of public biocuration resources. Prerequisites: Basic understanding of the genetic basis of cancer.

Proposed topics for curation and hackathon

Participants can create an issue describing a proposed topic and apply the 'hackathon' or 'curation' label at: https://github.com/genome/civic-meeting

CIViC bug reports and feature requests

Please create an issue in the civic-server or civic-client github repos as appropriate:
https://github.com/genome/civic-server

https://github.com/genome/civic-client

Curation Help

For questions or assistance with the CIViC curation interface please email: civic-help@genome.wustl.edu.

Acknowledgements

The CIViC resource and this meeting is supported by the National Cancer Institute (NCI) of the National Institutes of Health (NIH) under Award Number U01CA209936 to Obi Griffith (with Malachi Griffith and Elaine Mardis as co-PIs). We also thank Scripps (Andrew Su and Chunlei Wu) for co-hosting and supporting this meeting.