How to interpret the scores given by variant pathogenicity predictors such as Polyphen2, SIFT and MutationTaster provided in the SOPHiA DDM™ Platform, and compare these values to those provided by the respective tools?
The SOPHiA DDM™ Platform normalizes the scores provided by the different tools to enable comparisons among the different sources. For all predictive scores, a value of 1 means likely pathogenic and 0 likely benign irrespective of the database. The scores are transformed following the rationale and procedure established by dbNSFP and discussed in Xiaoming Liu et al., 2011 (https://doi.org/10.1002/humu.21517). Briefly:
“1 – SIFT” score (the original SIFT score is opposite, with 0 being predicted pathogenic and 1 benign)
No transformation (there are two different PolyPhen2 scores, HVAR and HDIV, based on different training sets)
Combination of prediction and score (1 – confident, 0 not confident). The SOPHiA DDM™ Platform displays the score if the prediction is pathogenic or “1 – score” if the prediction is benign