DeepTAP is a deep learning approach used for predicting high-confidence TAP-binding peptide.

Contact: [email protected]

Windows/Linux

python pytorch

Download the latest version of DeepTAP from https://github.com/xuezhang335/DeepTAP

git clone https://github.com/xuezhang355/DeepTAP.git

Go into the directory by using the following command:

cd DeepTAP

Invoke the setup script:

python setup.py install

Single peptide:

classification model prediction:
    deeptap -t cla -p <LNIMNKLNI> -o <output directory>
regression model prediction:
    deeptap -t reg -p <LNIMNKLNI> -o <output directory>

List of peptides in a file:

classification model prediction:
    deeptap -t cla -f <input file> -o <output directory>
regression model prediction:
    deeptap -t reg -f <input file> -o <output directory>

DeepTAP takes csv files as input with head of "peptide" (requisite). For example (demo/demo1.csv)

V1.0
Test the suitabilty of different RNN variants (GRU,LSTM,BGRU,BLSTM,att-BGRU and att-BLSTM) and CNN on the binding prediction and select the best one (BGRU) for model construction.