Skip to main content

Table 3 Performance comparison between miTAR1 and DeepMirTar on an independent dataset

From: miTAR: a hybrid deep learning-based approach for predicting miRNA targets

Model: dataset

Identified positive miRNA:targets/Total number of positive pairs (%)

DeepMirTar: DeepMirTar Independent test set

24/48 (50%)

miTAR1: DeepMirTarIn

36/48 (75%)

  1. The positive pairs for both models are the same set of pairs