Inter-platform concordance of gene expression data for the prediction of chemical mode of action

Biology Direct

Table 3 Accuracies of predicting MOA’s in the adjusted test set, based on classifiers developed on complete gene expression sets profiled from microarray and RNASeq platforms

Platform	Classifier	Overall Acc. %	Sensitivity, Specificity
			PPARA	CAR/PXR	Control
Microarray	Ensemble	62	56,66	44,73	100,49
	SVM	50	33,60	44,54	83,39
	RF	67	89,54	22,94	100,56
	PLS+LDA	67	67,67	44,81	100,56
	PLS+RF	54	44,60	33,67	100,39
	PCA+LDA	12	33,0	0,20	0,17
	PCA+RF	8	22,0	0,13	0,11
	RPART	62	100,39	11,93	83,55
RNASeq	Ensemble	62	56,66	44,73	100,49
	SVM	54	44,60	33,67	100,39
	RF	62	78,52	22,86	100,49
	PLS+LDA	58	44,66	44,66	100,44
	PLS+RF	50	44,54	22,67	100,33
	PCA+LDA	33	33,33	0,53	83,16
	PCA+RF	25	22,27	0,40	67,11
	RPART	42	44,41	22,54	67,34

ISSN: 1745-6150