Yes, that’s the way I understand it - both through conversations with people that work with it and with some reading. Note, I’ve not actually used it myself.
A recent paper has done a comparison of 10 different tools, using a huge dataset. I think in that paper (below) they use ‘specificity’ to indicate the correct identification of a benign variant. As far as I know, none of the tools would identify a change as “damaging” if it were in fact “positive,” simply because it has no other label. I suppose if the change created a neomorph (a protein with a new function), in which the previous function were lost, the phenotype might somehow point to a “loss” of function. I doubt it.
So, I don’t think that you are correct about ‘sensitivity’ and ‘specificity.’ A “true positive” is a correct identification of a “damaging” change, and sensitivity is a measure of that. A “true negative” is a correct identification of a benign change, and specificity is a measure of that. I think this is the relevant part of the legend for Table 1 from the 2012 Human Mutation paper:
Sensitivity equals the number of true positives divided by the number of true positives plus false negatives; Specificity equals the number of true negatives divided by the number of true negatives plus false positives;
This is the more recent paper, with huge dataset to compare 10 tools. PolyPhen-2 is mediocre.
1 Like
Hi Matthew, the original significance of this thread was to discuss Gauger’s defense of Behe’s claim that some of the adaptive mutations in the APOB gene in polar bears are probably damaging. Behe’s claim is that most adaptive mutations are damaging to proteins, thus further limiting Darwinian evolution. For example, he would say that the genes for digesting chitinase probably didn’t come about through a Darwinian process. We may have lost them through a Darwinian process, but we probably didn’t gain them through a Darwinian process.
This topic was automatically closed 3 days after the last reply. New replies are no longer allowed.