Loss of function rule - PVS1

Here you can find more details about Franklin's implementation of the ACMG PVS1 rule used for SNP variants classification

Updated over a week ago

The PVS1 rule is related to null variants which usually result with a loss of function (LOF) effect. These variants include stop gain, frameshift insertions or deletions, splice donor or acceptor, a start loss. The original 2015 ACMG/AMP sequence variant interpretation guidelines suggest applying the PVS1 rule with a recommended of PVS on the occasion of a null variant in a gene known to be sensitive to LOF, with a caveat being cautious when this happens in the last exon. While this could be used as a rule of thumb, there are a variety of exceptions and other parameters need to be considered. Due to that, in November 2018, the ClinGen SVI group published a paper refining the PVS1 rule, where they describe a detailed decision tree suggesting how to apply the PVS1 rule to each variant type (e.g start loss should be handled differently than stop gain), as well different weights should be given rather than the default PVS.

ACMG PVS1 Decision Tree

When looking at the recommendations we recognize some computational challenges which make it difficult to follow these recommendations.

Some examples are:

  • In the case of a splice donor/acceptor variant, is the potential skipped exon will result in a frameshift or not?

  • Is a specific exon critical to the protein function?

  • Does the variant fall in the last exon or the last 50 bases of the preliminary exon?

To facilitate the process, Genoox AI Engine automatically implements these recommendations, while giving a detailed explanation for the supporting evidence which it’s based on.

As these recommendations can be implemented using different thresholds, the default behavior some of the next parameters:

  • LOF is a known mechanism for disease in the gene (equivalent to biologically relevant transcript(s) in the decision tree):

  • Truncated/altered region/exon is critical to protein function

To validate our algorithm we benchmarked against the official dataset from the paper which was used to create these recommendations.

More details can be found in the poster “Using Artificial Intelligence for Implementing New Recommendations of the PVS1” which we presented at the 2019 ACMG conference.

Still have questions? Reach out to our Support Team, they'll be happy to help!

Did this answer your question?