DeepTorrent is a deep learning-based approach and tool for accurately predicting 4mC sites from DNA sequences. It combines four different feature encoding schemes to encode the raw DNA sequences and employs multi-layer convolutional neural networks (CNN) with the inception module integrated with the bidirectional long short-term memory (BLSTM) to effectively learn the higher-order feature representations. Dimension reductions and concatenates feature maps produced by filters of different sizes are applied in the inception module. In addition, the attention mechanism and transfer learning techniques are also employed to train the robustness predictor. Empirical benchmarking experiments demonstrate DeepTorrent significantly improves the performance of 4mC site prediction compared with the state-of-the-art methods.

Keywords for this software

Anything in here will be replaced on browsers that support the canvas element