Site Search
Computer Science


Alioune Ngom, Ph.D.Dr. Alioune Ngom
Dr. Alioune Ngom
Dr. Scott GoodwinDr. Scott Goodwin
Dr. Scott Goodwin
Dr. Luis RuedaDr. Luis Rueda
Dr. Luis Rueda
Dr. Robert KentDr. Robert Kent
Dr. Robert Kent
Jessica Chen, Ph.D.Dr. Jessica Chen
Dr. Jessica Chen
Dr. Ziad Kobti lecturingDr. Ziad Kobti
Dr. Ziad Kobti
Imran Ahmad, Ph.D.Dr. Imran Ahmad
Dr. Imran Ahmad
Xiaobu Yuan, Ph.D.Dr. Xiaobu Yuan
Dr. Xiaobu Yuan
Robin Gras, Ph.D.Dr. Robin Gras
Dr. Robin Gras
Windsor WaterfrontWindsor Waterfront Park
Windsor Waterfront Park
Christie Ezeife, Ph.D.Dr. Christie Ezeife
Dr. Christie Ezeife
Lambton TowerLambton Tower
Lambton Tower
Arunita Jaekel, Ph.D.Dr. Arunita Jaekel
Dr. Arunita Jaekel

Drug-Target Interaction Networks Prediction Using Short-linear Motifs

Add this event into your calendar using the iCAL format
  • Thu, 05/04/2017 - 1:00pm - 3:00pm

Drug-Target Interaction Networks Prediction Using Short-linear Motifs

MSc Thesis Defense by:

Wenxiao Xu

Date:  Thursday, May 4, 2017
Time:  1:00 pm – 3:00 pm
Location: 3105, Lambton Tower

Abstract: Drug-target interaction (DTI) network prediction is a fundamental step of drug discovery and genomic research. Various computational methods have been developed to find potential DTIs. Machine learning (ML) approaches have been currently being proposed for identifying new drug-target interactions from existing DTI networks. There are ML-based approaches for DTI network prediction: similarity-based methods and feature-based methods. In this thesis, we propose a feature-based approach, and firstly use short-linear motifs (SLiMs) as descriptors of protein. Additionally, chemical substructure fingerprints are used as features of drug. Moreover, another challenge in this field is lack of negative data for the training set because most data which can be found in public databases is interaction samples. Many researchers regard unknown drug-target pairs as non-interaction, which is incorrect, and may cause serious consequences. To solve this problem, we extend a strategy to select reliable negative samples according to positive data. We use the same benchmark datasets as previous research in order to compare with them. After trying two classifiers, 1-NN and Random Forest, on the training sets, we find that our results are better than existing research, while Random Forest performs the best.

Thesis Committee:
Internal Reader: Dr. Mehdi Kargar  
External Reader:  Dr. Dilian Yang   
Advisors: Dr. Alioune Ngom & Dr. Luis Rueda
Chair: Dr. Scott Goodwin

See More: