Variant Annotation 2016-12-03T06:23:57+00:00

Extensive, Accurate and Up-to-date Variant Annotation

Building on accurate variant calls, we integrate numerous public and proprietary databases to provide detailed and current annotations for variants across the genome in a highly structured, consistent and thorough manner.


Proprietary Databases

The challenges inherent in providing accurate annotation and interpretation of the genome have been described by the Personalis team in prior publications. These challenges include issues with both the accuracy of the annotations themselves as well as how those annotations are synthesized and interpreted. To address quality and coverage issues, Personalis has developed, or exclusively licensed, several large manually-curated databases that provide high quality structured data to power our downstream analytics:

Personalis Disease Variant Database
The Personalis Disease Variant Database is the most comprehensive, detailed, high-quality manually-curated variant to common disease, Mendelian disease, and phenotype database of its kind containing over 600,000 variant-to-disease relationships. This database enables highly reliable biomedical interpretation of not only exonic variants, but also those located in intronic and intergenic regions. Furthermore, the database is ethnicity specific when that information is available.
Personalis Pharmacogenomics Database
The Personalis Pharmacogenomic Database is built on PharmGKB®, the world’s leading pharmacogenomics database which has been exclusively licensed by Personalis. PharmGKB content includes information linking genetic variants to drug toxicity, dosing, and efficacy. PharmGKB content has been manually-curated and extensively peer-reviewed over a 12-year period. Personalis has used this information to refine its sequencing and interpretation technologies to measure these genetic variations accurately.
Personalis Regulatory Regions Database
The Personalis Regulatory Regions Database contains several hundred thousand putative transcription factor binding sites/regulatory regions. These regulatory annotations enable us to interpret intergenic and other non-coding variants that may have a significant impact on gene expression and function.

Public Databases

Personalis’ annotation engine draws on data from a broad range of public and commercial databases to complement our proprietary databases. The Personalis annotation engine updates, integrates, and version controls a broad variety of public and proprietary databases to provide rich variant and gene annotation. Annotations include:

  • Ethnically specific population allele frequencies
  • Non-coding RNA (ncRNA)
  • Regulatory elements
  • Problematic regions of the genome
  • Conservation
  • Known structural variation
  • Variant to disease relationships
  • Structural variants associated with cancer
  • Networks
  • Interactions
  • Pathways
  • Biological function
  • Drug targets
  • Genetic tests
Together these databases serve as powerful inputs for our analytics to make sense of genomic data