Performance of Web tools for predicting changes in protein stability caused by mutations

BMC Bioinformatics

Table 4 General results from the assessment on the datasets of multimeric proteins

	PoPMuSiC	DynaMut	DUET	INPS-MD	MAESTROweb
Values calculated for the full dataset of multimeric proteins
True negative	131	96	127	135	125
False positive	11	46	15	7	17
True positive	21	30	25	14	22
False negative	24	15	20	31	23
Accuracy	0.81	0.67	0.81	0.80	0.79
True negative rate (specificity)	0.92	0.68	0.89	0.95	0.88
True positive rate (sensitivity)	0.47	0.67	0.56	0.31	0.49
Positive predictive value (precision)	0.66	0.39	0.63	0.67	0.56
Negative predictive value	0.85	0.86	0.86	0.81	0.84
MCC	0.44	0.30	0.47	0.35	0.39
Values calculated for the balanced dataset of multimeric proteins
True negative	36	21	31	35	31
False positive	2	17	7	3	7
True positive	21	30	25	14	22
False negative	24	15	20	31	23
Accuracy	0.69	0.61	0.67	0.59	0.64
True negative rate (specificity)	0.95	0.55	0.82	0.92	0.82
True positive rate (sensitivity)	0.47	0.67	0.56	0.31	0.49
Positive predictive value (precision)	0.91	0.64	0.78	0.82	0.76
Negative predictive value	0.60	0.58	0.61	0.53	0.57
MCC	0.46	0.22	0.38	0.29	0.32

True negative and true positive values have been considered as those predictions that correctly found a negative and a positive sign for destabilizing and stabilizing mutations, respectively. Data have been reported only for mutations causing a ΔΔG energy variation outside the range of the experimental error, in the full and the balanced datasets of multimeric proteins

ISSN: 1471-2105