Table 1.
Program | MAPP (%) | SIFT (%) | VarioWatch (%) | PolyPhen-2 (%) |
---|---|---|---|---|
MAPP | 100 | |||
SIFT | 64.6 | 100 | ||
VarioWatch | 59.4 | 62.9 | 100 | |
PolyPhen-2 | 71.4 | 64.2 | 62.9 | 100 |
For the off-diagonal values, this table looks at pairwise agreements between the four prediction programs in predicting whether a variant is tolerated or deleterious. For example, 64.6% of the time, MAPP and SIFT agreed on their prediction, whereas VarioWatch and SIFT only agreed 62.9% of the time. The highest agreement is between MAPP and PolyPhen-2, and the lowest agreement is between MAPP and VarioWatch.