QSBR Study of Bitter Taste of Peptides: Application of GA-PLS in Combination with MLR, SVM, and ANN Approaches

Table 8

Developed MLR model statistics for subsets of peptides compared with a previously developed model.

Data set	Developed model					Previous model
	ND	𝑅^2	𝑅^2adj	𝑅	RMSE	𝑅 (PLS)	𝑅^2 (PLS)
Dipeptides^a	76	0.51	0.47	0.72	0.41	0.63	0.40
𝐃𝐢𝐩𝐞𝐩𝐭𝐢𝐝𝐞𝐬^b	45	0.62	0.56	0.79	0.44	0.91	0.83
𝐃𝐢𝐩𝐞𝐩𝐭𝐢𝐝𝐞𝐬^c	47	0.59	0.53	0.77	0.42	0.85	0.72
Three peptides	51	0.65	0.60	0.81	0.38	0.71	0.50
Tetrapeptides	23	0.72	0.62	0.85	0.48	0.90	0.81
Pentapeptides	12	0.89	0.80	0.94	0.37	0.88	0.77
Hexapeptides	20	0.65	0.48	0.80	0.47	0.75	0.56
Heptapeptides	16	0.79	0.68	0.89	0.38	0.95	0.90
Octa-tetradecapeptides	24	0.57	0.42	0.76	0.44	—	—
Whole data set^d	227	0.80	0.79	0.89	0.46	0.81	0.66
Test and validation sets	46	0.76	0.73	0.87	0.56	—	—
Training set	181	0.81	0.81	0.90	0.43	—	—

 ^aAverage of experimental values was used when there were different values in different references.
^bExperimental data were taken from different references.
^cExperimental data were taken from [3].
^dThe 𝑅 values for whole data set using SVM and ANN methods are 0.90 and 0.91, respectively.