Figure 2
The structure of the deep learning method contains two parts: the embedding module and task specific module. In the embedding module, RNA sequences of fixed lengths were converted into a numeric matrix vector and used as an input for the task specific module. The task specific module contains a CNN layer for extracting features, a max-pooling layer for speeding up the computation, BLSTM and attention layers, and fully connect layer for output result.

The structure of the deep learning method contains two parts: the embedding module and task specific module. In the embedding module, RNA sequences of fixed lengths were converted into a numeric matrix vector and used as an input for the task specific module. The task specific module contains a CNN layer for extracting features, a max-pooling layer for speeding up the computation, BLSTM and attention layers, and fully connect layer for output result.

Close
This Feature Is Available To Subscribers Only

Sign In or Create an Account

Close

This PDF is available to Subscribers Only

View Article Abstract & Purchase Options

For full access to this pdf, sign in to an existing account, or purchase an annual subscription.

Close