BioStars already has some nice posts on data portals for exploring mutation observations (e.g. GDC, ICGC), databases of tumor suppressors and oncogenes, and drug-gene interactions. However, these don't quite cover another category of very useful resources for interpreting variants in cancer.
Particularly the case where a germline or somatic variant is observed in a patient and you want to determine what knowledge supports any possible clinical action or interpretation associated with that variant. These actions may range from establishing a germline variant as predisposing the patient for cancer, or determining if a somatic variant predicts response to a therapy (predictive), helps to define a sub-type (diagnostic), or predicts outcome (prognostic).
Here are some of my favorite resources for this area:
- Clinical Interpretation of Variants in Cancer (CIViC). Developed at WASHU. Known for its openly licensed content and open crowdsourced (with editor review) curation interface. Disclosure: I am part of the CIViC team.
- OncoKB. Developed at MSKCC.
- JAX-CKB. Developed at the Jackson Laboratory.
- Cancer Genome Interpreter. Developed at IRB Barcelona.
- ClinVar (largely focused on germline predisposition). Maintained by NCBI.
- ClinGen VCI and ClinGen Evidence Repository. Maintained by the ClinGen consortium (multi-institute).
- MyCancerGenome. Developed at Vanderbilt.
- COSMIC (Drug Resistance Section). Developed at Wellcome Sanger Institute.
- Pharmacogenomics Knowledgebase (PharmGKB). Developed at Stanford University.
- Precision Medicine Knowledgebase (PMKB). Developed at Weill Cornell Medicine.
- Personalized Cancer Therapy Knowledge Base for Precision Oncology. Developed at MD Anderson.
- Database of Evidence for Precision Oncology (DEPO). Developed at WASHU.
- CANDL. Developed at Ohio State University.
This list above is meant to cover the resources that are focused on curating / organizing this knowledge. Other useful resources aggregate information from across these and attempt to harmonize them, such as:
Others display clinical interpretation results in the context of other genomic information such as CBioPortal. The post Exploring cancer mutation data portals has a nice list of these.
Several resources are making a concerted effort to harmonize their efforts, develop and adopt common standards and guidelines, and generally work towards greater interoperability and reduce duplication of effort. This work is critical to achieving several important goals including: (a) achieving community/expert consensus on the clinical interpretation of variants, (b) creating a comprehensive set of interpretations that cover all clinically meaningful variants, (c) keeping interpretations up to date as knowledge evolves, and (d) ensuring maximum dissemination and accessibility of this knowledge to clinicians and patients.
Once such effort at harmonization is the Variant Interpretation for Cancer Consortium (VICC) "Driver Project", a driver project of the Global Alliance for Genomics and Health (GA4GH). Creating the VICC meta-knowledgebase linked above is one activity of this group among several.