SIFT on Blocks
SIFT on Blocks
SIFT predicts what amino acid substitutions would affect function. The prediction is based on the amino acids appearing at the position in the block.
[SIFT home page][
Implementation details]
Side effects for submitting blocks
SIFT usually takes an alignment as input. Since blocks are used instead, there are some side-effects to the way the output is printed. Method for predicton has not changed.
An example of the output:
Predict Not Tolerated |
Position |
Seq Rep |
Predict Tolerated |
| |
| c |
w | d |
f | m |
i | y |
v | g | p | s | h | n |
a | l |
t | | 7Q | 0.95 |
e |
K | Q |
R | | | | | | | |
| |
| |
| | |
| |
|
Meaning:
At position 7 in the block, 95% of the sequences in the block have an amino
acid
appearing at this position. K, Q, R, e are predicted as tolerated for this block. K, Q, R are observed
in
the block at this position (capitalized) but E is not (lower cased). C, W, D, F, M, I, Y, V, G, P, S, H, N, A, L, T
are
predicted to be deleterious and none of these appear at that position in the block (lower case).
- The Q in 7Q is the amino acid in the first sequence of the block.
- If you are submitting more than one block, the position of prediction is relative to the block and previous blocks.
For example, you want to know what amino acids are predicted to be deleterious for position 5 in block IPB000184B.
Because IPB00184B is preceded by block IPB000184A (which is 25 amino acids in width), rather than looking at position 5 of the SIFT prediction, the prediction for position 5 in IPB00184B is at position 30 (25 + 5).
Page last modified August 2001
Questions or comments?
Contact us