Ki DoQ: using docking based energy scores to develop ligand based model for predicting antibacterials

BMC Bioinformatics

Table 6 Results obtained during external cross-validation procedure using three non-correlated descriptors

Size of training set	q ²	MAE	Size of test set	r ²	MAE
21	0.83 ± 0.02	1.90 ± 0.07	2	0.95 ± 0.03	0.98 ± 0.09
19	0.81 ± 0.01	2.01 ± 0.08	4	0.94 ± 0.02	0.98 ± 0.37
17	0.78 ± 0.03	2.28 ± 0.24	6	0.93 ± 0.03	1.09 ± 0.24

ISSN: 1471-2105