Combined machine learning models, docking analysis, ADMET studies and molecular dynamics simulations for the design of novel FAK inhibitors against glioblastoma

Table 1 Performance of the ML models on the ten-fold cross-validation of test set using LightGBM method

Fingerprint	Input	Test set
Fingerprint	Number	R²	MAE	RMSE
CDK	1024	0.876 ± 0.006	0.356 ± 0.010	0.500 ± 0.012
CDK extended	1024	0.881 ± 0.005	0.353 ± 0.006	0.489 ± 0.011
Substructure	307	0.785 ± 0.006	0.492 ± 0.007	0.658 ± 0.009
Substructure count	307	0.816 ± 0.006	0.452 ± 0.007	0.609 ± 0.010
CDK + CDK extended	2048	0.887 ± 0.007	0.339 ± 0.010	0.477 ± 0.014
Substructure + Substructure count	614	0.816 ± 0.006	0.452 ± 0.007	0.610 ± 0.010
CDK + Substructure count	1331	0.881 ± 0.005	0.349 ± 0.006	0.490 ± 0.010
CDK extended + Substructure count	1331	0.883 ± 0.004	0.351 ± 0.006	0.487 ± 0.008
CDK + Substructure + Substructure count	1638	0.881 ± 0.005	0.349 ± 0.006	0.490 ± 0.010
CDK extended + Substructure + Substructure count	1638	0.883 ± 0.004	0.351 ± 0.006	0.487 ± 0.008
CDK + CDK extended + Substructure	2355	0.885 ± 0.006	0.344 ± 0.009	0.482 ± 0.012
CDK + CDK extended + Substructure count	2355	0.888 ± 0.005	0.342 ± 0.006	0.476 ± 0.010
CDK + CDK extended + Substructure + Substructure count	2662	0.888 ± 0.005	0.342 ± 0.006	0.476 ± 0.010

ISSN: 2661-801X