SummaryDTA drug-target binding affinity prediction

Introduction

The target sequences obtain prior representation on the pre-trained protein sequence model, and the sequence feature vectors are obtained through the variable-length RNN network and averaging operation. The targets also use the concat maps as the structure information, which describe the interaction information between the residues. The residual blocks obtain the structure feature vector, and finally represent the target information through the concat operation. The drug SMILESs are constructed graphs through the node feature and adjacency matrix, which are obtained by the RDKit tool. Hence, the three GCN layers and the max-pooling operation are designed to obtain the graph representation vector of the drugs. Finally, the Kronecker Product is used to perform feature fusion on the candidate drug-target pair, and the final prediction binding affinity values are output through the linear layers.

Publications

No Publication Information

Credits

  1. Fei Guo fguo@tju.edu.cn
    Investigator

    School of Computer Science and Technology, College of Intelligence and Computing, Tianjin University, China

Community Ratings

UsabilityEfficiencyReliabilityRated By
0 user
Sign in to rate
Summary
AccessionBT007258
Tool TypeApplication
CategoryOther tools
PlatformsLinux/Unix
TechnologiesPython3
User InterfaceTerminal Command Line
Latest Release0.1 (September 14, 2021)
Download Count254
Country/RegionChina
Submitted ByFei Guo
Fundings

2018YFC0910400