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

Biology Direct

Table 2 Accuracies of predicting MOA’s in the originally given test set, based on classifiers developed on common gene expression sets profiled from microarray and RNASeq platforms

Platform	Classifier	Overall Acc. %	Sensitivity, Specificity
			PPARA	CAR/PXR	Control	OTHER
Microarray	Ensemble	50	44,52	44,52	100,42	39,58
	SVM	55	0,70	0,70	83,50	100,21
	RF	57	0,73	0,73	100,50	100,25
	PLS+LDA	40	67,33	56,36	67,36	6,67
	PLS+RF	55	11,67	11,67	100,48	83,34
	PCA+LDA	12	0,15	0,15	83,0	0,21
	PCA+RF	40	0,51	0,51	94,31	0,70
	RPART	45	100,30	0,57	83,39	28,58
RNASeq	Ensemble	50	33,55	33,55	100,42	50,50
	SVM	60	0,76	11,73	100,53	100,30
	RF	55	0,70	0,70	100,48	94,26
	PLS+LDA	38	56,33	56,33	100,28	0,66
	PLS+RF	55	11,67	22,64	100,48	78,38
	PCA+LDA	14	0,18	0,18	100,0	0,24
	PCA+RF	43	0,55	0,55	83,36	72,21
	RPART	43	33,46	33,46	33,45	56,33

ISSN: 1745-6150