Table 2.

Comparison of most relevant QSAR studies on BBB permeability.

Descriptors	N train	N test	Methods	r train 2	Predictive accuracy on test set	Reference
Δlop P, log⁡P,and log⁡P cyc	20	—	Linear Regression	0.69	—	Young et al. [77]
Excess molar refraction, dipolarity/polarisability, H-bond acidity, and basicity Solute McGowan volume	148	30	LFER	0.75	r test 2 = 0.73	Platts et al. [66]
ΔG W°	55	—	Linear Regression	0.82	—	Lombardo et al. [78]
PSA, the octanol/water partition coefficient, and the conformational flexibility	56	7	MLR	0.85	r test 2 = 0.80	Iyer et al. [79]
CODESSA/DRAGON (482)	200	110	PLS SVM	0.83 0.97	r test 2 = 0.81  r test 2 = 0.96	Golmohammadi et al. [62]
Molecular (CODESSA-PRO) descriptors (5)	113	19	MLR	0.78	r test 2 = 0.77	Katritzky et al. [15]
Molecular fragment (ISIDA) descriptors	112	19	MLR	0.90	r test 2 = 0.83	Katritzky et al. [15]
PSA, log⁡P, the number of H-bond acceptors, E-state, and VSA	144	10	Combinatorial QSAR (KNN SVM)	0.91	r test 2 = 0.8	Zhang et al. [17]
Abraham solute descriptors and indicators	328	—	LFER	0.75	—	Abraham et al. [51]
Abraham solute descriptors and indicators	164	164	LFER	0.71	s = 0.25, MAE = 0.20	Abraham et al. [51]
CODESSA/Marvin/indicator (6)	260	63	GA based SVM	0.83	r test 2 = 0.84, RMSE = 0.23	This research, GA/SVM, final model C = 13.3573, γ = 0.715761, ε = 0.151289
CODESSA/Marvin/indicator (236)	260	63	GA based SVM	0.97	r test 2 = 0.55, RMSE = 0.31	This research, Grid/SVM C = 8.0, γ = 0.015625, ε = 0.0625
CODESSA/Marvin/indicator (6)	260	63	GA based SVM	0.86	r test 2 = 0.58, RMSE = 0.29	This research, Grid/SVM C = 8.0, γ = 1.0, ε = 0.125