Relationship between the predictions derived from SIFT and Polyphen and the structural features
calculated for the selected subset of variants. Unfavourable structural properties of the mutations
are highlighted in bold and underlined, and partially unfavourable properties are highlighted in
bold. The remaining properties were neutral. The first column contains the mutation; the second
column contains the gene name; the third column contains the PDB identifier of the structure
template; the fourth column lists the SIFT scores; the fifth column lists the Polyphen scores; the
sixth column contains the prediction according to the SIFT and Polyphen scores as: ND: non-damaging
and D: deleterious; the seventh column contains the polarity change in a hydrophobicity scale (see
Materials and Methods); the eighth column contains the change in the protein charge, based on
changes in the charge of the changed residue (see Materials and Methods); the ninth column lists the
SNAP prediction, N: neutral and NN: non-neutral, with the accuracy in brackets; the tenth column
contains the SMD prediction N: neutral, S: stabilizing, SS: slightly stabilizing, D: destabilizing,
SD: slightly destabilizing and HD: highly stabilizing and causing protein malfunction; the eleventh
column contains the SMD pseudo ΔΔG; and the twelfth column lists the change in
percentage of solvent accessibility, coded as A: accessible, PA: partly accessible and B: buried