DrugECs: ensemble system with feature subspaces to

accurate drug-target interaction prediction


The flowchart of the ensemble system for the drug-target prediction.

  Background: Drug-target interaction is key in drug discovery, especially in the design of new leader compound. However, the work to find a new leader compound for a specific target is complicated and hard, and it always lead to many mistakes. Therefore computational techniques are commonly adopted in drug design, which can save time and costs in a significant extent.
  Results: To address the issue, a new prediction system is proposed in this work to identify drug-target interaction. First, drug-target pairs are encoded with a fragment technique and the adopted software "PaDEL-Descriptor". The fragment technique is for encoding target proteins, which divides each protein sequence into several fragments in order and encodes each fragment with several physiochemical properties of amino acids. The software "PaDEL-Descriptor" creates encoding vectors for drug molecules. Second, the dataset of drug-target pairs is resampled and several overlapped subsets are obtained, which input into kNN (k-Nearest Neighbor) classifier to build an ensemble system.
  Conclusion: Experimental results on the drug-target dataset showed that our method performs better and runs faster than the state-of-the-art predictors.


Software available:

 A simple Matlab implement of our predictor is available here: DrugECs.



drug-target datasets: enzymes, icon channels, GPCRs, and nuclear receptors.


Jinjian Jiang, Nian Wang, Peng Chen*, Jun Zhang, Bing Wang, DrugECs: ensemble system with feature subspaces to accurate drug-target interaction prediction. Submitted.

Copyright @ 2004-2016 by Peng Chen

All Rights Reserved