MAESTRO - multi agent stability prediction upon point mutations

BMC Bioinformatics

Table 2 Performance comparison using the SP2 set

Method	#Predictions ^a	Pearson’s ρ ^b	σ (kcal/mol) ^b
AUTOMUTE	315	0.46/0.45/0.45	1.43/1.46/1.99
CUPSAT	346	0.37/0.35/0.50	1.91/1.96/2.14
Dmutant	350	0.48/0.47/0.57	1.81/1.87/2.31
Eris	334	0.35/0.34/0.49	4.12/4.28/3.91
I-Mutant-2.0	346	0.29/0.27/0.27	1.65/1.69/2.39
PopMuSiC-2.0	350	0.67/0.67/0.71	1.16/1.19/1.67
SDM	350	0.52/0.53/0.63	1.80/1.81/2.11
mCSM	350	0.73/0.74/0.82	1.08/1.10/1.48
MAESTRO-Score	350	0.56/0.57/0.68	−/ −/ −
MAESTRO	350	0.70/0.69/0.76	1.13/1.17/1.67

Results except for MAESTRO are taken from Dehouck et al. [12] and Pires et al. [14] respectively. ^aThe test set contains 350 entries, however several methods failed to compute the Δ ΔG prediction for some mutants, resulting in a reduced number of predictions. In these cases Δ ΔG was set to 0.0 kcal/mol for calculating the correlation coefficient. ^bThree values are given for Pearson’s ρ as well as for the associated standard errors. They correspond (i) to the whole validation set, (ii) the subset of 309 mutants for which all methods provide a result, and (iii) the subset of 87 mutants with an experimental Δ ΔG≥2 kcal/mol or Δ ΔG≤2 kcal/mol respectively.

ISSN: 1471-2105