Prediction of specificity-determining residues for small-molecule kinase inhibitors

BMC Bioinformatics

Table 1 S-Filter predictions

Compound	Primary target	PDB	Predictions	FN	FP	TN	TP	Total
SB-203580	p38	1A9U	Leu 104 Thr 106	0	0	14	2	16
PHA-00781089	MK2	2P3G	Cys 140 Gly 143	0	1	18	1	20
Roscovitine	CDK2	2A4L	Val 64 Leu 83 Ala 144	0	2	20	1	23
PP1	SRC/HCK	1QCF	Thr 338 Gly 344	0	1	18	1	20
OSI-774	EGFR	1M17	Thr 766 Cys 773	0	1	19	1	21
GW-572016	EGFR	1XKK	Cys 775	2	0	23	1	26
Fasudil	ROCK1	2ESM	Val 137 Met 153 Ala 215	1	2	14	1	17

Predicted specificity determinants are shown for each compound and are either designated as false positives (FP) or true positives (TP). Five of the predicted residues are in italics as they are self-designated true positives after validating them through structural superposition studies. Residues that were not predicted as specificity determinants are not shown but are either designated as false negatives or true negatives. The total number of residues in the binding site are shown in the final column.

ISSN: 1471-2105