A convolutional neural network (CNN) for the prediction of Tetratricopeptide reapeats (TPR) in amino acid sequences.
The architecture of DeePR consists of multiple convolutional layers with different properties (filters etc.), dropout layers, pooling layers and one final fully connected layer.
The network was trained on multiple hundred thousand sequences both TPR sequences and non-TPR sequences.
For more information on datasets and/or implementation feel free to contact me via mail.