Introduction:
In this work, We add ECA-Net after the dense convolutional network, which is an efficient channel attention model. After the channel level global average pooling without reducing the dimension, ECA-Net captures local cross channel interaction information by considering each channel and its k neighbors. ECA-Net can be effectively implemented by fast 1D convolution with the size of k. The convolution core size of k represents the coverage of local cross channel interaction.
Intput sequence:
The user can submit the sequence of the protein (example)). According to the sequence entered by the user, the model can predict whether the sequence contains a S-nitrosylation site and return the result to the web interface.
Intput file:
Users can submit a multi-sequence file (in FASTA format) by clicking a button. SNO-DCA Server will send the predicted results to the mailbox submitted by the user.
Program name:
User's naming of forecast project.
Email:
The email address used by the user to receive the prediction result.