Knowledge about the interaction between drugs and proteins is essential in the drug discovery process. Understanding the relationship between compounds and proteins through wetlab experiments alone is time-consuming and costly. To support the experimental work by prioritizing the most potent compounds for a target, numerous methods for the in-silico prediction of drug-target interaction have been proposed and high performance on binary datasets have been reported. A drawback of binary datasets is that missing values and non-interacting drug-target pairs are not differentiated. In this thesis, a model is developed that predicts the drug-target binding strengths as continuous values and thus incorporates the whole interaction spectrum from true negative to true positive interaction. The developed model combines two previously used approaches for the problem, which are Matrix Factorization and Conditional Random Fields. The model is evaluated on three datasets and a slight performance improvement is observed when compared to the state of the art method.
Copyright is held by the author.
This thesis may be printed or downloaded for non-commercial research and scholarly purposes.
Supervisor or Senior Supervisor
Thesis advisor: Ester, Martin
Member of collection