VarSome's ACMG explained
Charles Chapple, PhD, VarSome's Head of Bioinformatics, demonstrates the power of VarSome's automated implementation of ACMG guidelines for genome interpretation.
Learn in more about the reasoning behind the implementation of VarSome's ACMG classifier, one of the key-stone features provided by the VarSome Suite:
Our implementation is derived from the ”Standards and guidelines for the interpretation of sequence variants” was published in 2015 by Sue Richards et al. in their seminal paper (ACMG Guidelines). Following the advice from our clinical advisors, feedback from the VarSome user community, and using statistically justified thresholds we always strive to provide the most accurate and up-to-date implementation!
Clinical Evidence is the foundation stone of our ACMG evaluation and is derived from a multitude of sources. The data are then integrated into the implementation of all the rules requiring clinical evidence and the calculation of gene statistics. Splice site prediction and conservation tests are performed as well, to be utilized in the respective rules. Allele frequencies are taken into an account for the implementation of certain rules. Finally, selecting a specific transcript can modify the ACMG verdict and the user can select a transcript through the VarSome UI. This complex process requires the involvement of numerous databases for the formation of the ACMG classifier.
Our ACMG classifier has currently 21 implemented rules. Each rule comes with a default strength recommended by ACMG. However, it is allowed to reduce or enhance the strength of a rule depending on the available annotatioon data. VarSome's ACMG implementation makes use of this flexibility and allows users who disagree with a given implementation make their own modifications using VarSome UI.
Certain rules are not implemented or not currently available to VarSome users - in most cases, this is because the necessary data required to evaluate the rules is not in the public-domain, or the rules require patient-specific information, sometimes on a per-variant basis. Should they have more evidence, users can manually toggle rules on or off, or adjust the strength used, and the resulting classification will be re-evaluated immediately.