Protein solvent-accessibility prediction by a stacked deep bidirectional recurrent neural network

Buzhong Zhang, Linqing Li and Qiang Lyu
Biomolecules, 2018, 8(2): 33.

Residue solvent accessibility (RSA) is closely related to the spatial arrangement and packing of residues. Predicting the solvent accessibility of a protein is an important step to understand its structure and function. A more accurate prediction tool SDBRNN is provided here.

SDBRNN model for RSA prediction download link is here. Users need to install Theano and Keras2.0.

Our experiments used test datasets CASP10 CASP11 CB502 Manesh215 are provided here.

Feature_prepared.py is provided for prepare your testing dataset. example.py is provided for demo how to run our model.

mapping vector which maps a sparse 22-dimensional vector into a denser 22-dimensional vector download link is here.

The testing data style is: sequences residues features,labels. The 21-dim features are 20 PSSM and residues. PSSM is like: A R N D C Q E G H I L K M F P S T W Y V
Model of SDBRNN input data style: “sequences residues features” and the input features are: 20-PSSM, 7-dim Physical properties, 3-dim physicochemical characteristics, 1-dim conservation score, 22dim- protein encoding.

Please note that SDBRNN is free for academic use purpose only, for commercial usage, please contact us.

Thank you!

-Buzhong Zhang

If you have any suggestions or questions, Please email to: 20154027005@stu.suda.edu.cn