Table 4. Confusion matrix of support vector machine for combination of total length with 3′UTR length and G+C content.

	Fold change>1	Fold change≤1
Predicted fold change>1	2385	1006
Predicted fold change≤1	618	1620

Results of the support vector machine (corresponding to first row in Table 3) applied as a two-class predictor. The used eIF4E overexpression dataset consists of 4000 mRNAs for training and 5629 mRNAs for testing the classifier (see Materials and Methods).