Interpreting Phyre/SuSPect mutational sensitivity in comparison to variant effect prediction from SIFT/PROVEAN

[Mat Beale]

Hi,

I'm a relatively new user, and I was wondering if I could get some advice on interpreting the varying results I'm getting from different prediction approaches. I have a few proteins that seem to generate nice models in Phyre2, and when I look at specific variants of interest they seem to have a low mutational sensitivity at those sites. However, prediction tools like SIFT and PROVEAN (which tend to be diversity/BLAST based) predict that these sites are likely to be deleterious. What is the best way to interpret this? Is there a way to view the mutant and wildtype Phyre predictions side by side?

Many thanks,

Mat