Eye-color and Type-2 diabetes phenotype prediction from genotype data using deep learning methods

BMC Bioinformatics

Table 5 Results of random forest and extreme gradient boosting classifier

Classifiers	107	3	32	1560	9824	36,961	50,260	86,688
Random forest	0.92	0.89	0.91	0.90	0.86	0.81	0.81	0.82
(No scaling) (booster = gbtree, gblinear, dar)	0.88	0.89	0.88	0.92	0.92	0.92	0.92	–
	0.88	0.89	0.91	0.87	0.82	0.83	0.82	–
	0.88	0.89	0.88	0.92	0.92	0.92	0.92	–
(No scaling) (loss function = hinge, logistic, logitraw)	0.91	0.89	0.92	0.92	0.92	0.92	0.92	–
	0.92	0.89	0.92	0.93	0.92	0.92	0.93	–
	0.91	0.89	0.92	0.91	0.92	0.92	0.93	–
(Scaling) (booster = gbtree, gblinear, dar)	0.88	0.89	0.88	0.92	0.92	0.92	0.92	–
	0.85	0.89	0.90	0.86	0.78	0.78	0.73	–
	0.88	0.89	0.88	0.92	0.92	0.92	0.92	–
(Scaling) (loss function = hinge, logistic, logitraw)	0.91	0.89	0.9211	0.93	0.92	0.92	0.92	–
	0.92	0.89	0.92	0.93	0.92	0.92	0.93	–
	0.91	0.89	0.92	0.91	0.92	0.92	0.93	–

“–” means no results because of too long computation time. The different boosters used for XGBOOST are gbtree, gblinear, and dar. Different loss functions used for XGBOOST are hinge, logistic, and logitraw. The first row represents the number of SNPs used for Eye-color classification. The first column represents the classifier and different boosters used for the model. Scaling and No Scaling means dataset is scaled or not scaled for particular experiment or not

ISSN: 1471-2105