Predicting inhibitor for multidrug resistance associated protein-2 transporter

The efflux transporter multi-drug resistance associated protein-2 belongs to the ATP-binding cassette super-family. This family of proteins plays a very important role in multiple drug resistance and drug-drug interactions. These efflux transporters are considered to be the most important targets for the incremental efficacy of drugs. Researchers use them for studying efflux transporters for purpose of anticipating substrates, non-substrates, inhibitors and non-inhibitors. The work that was previously done on predictive models for the inhibitors of multidrug resistance which are associated with the Protein-2 efflux transporter showed that, good results were produced after machine learning.

The aim of this present work written by Sahil Kharangarh et al. is to deal with the development of a machine learning model that is also predictive enough to categorize the inhibitors and non-inhibitors of multidrug resistance that is associated with protein-2 transporter using data that is refined. In their review, many prediction algorithms related to machine learning were utilized to develop the models such as, support vector machine, random forest and k-nearest neighbor. Other methods like variance threshold, SelectKBest, random forest, and recursive feature elimination were utilized to select the features generated by PyDPI. A total number of 239 molecules consisting of 124 inhibitors and 115 non-inhibitors were used for the development of this model. The best inhibitor model that modelled multidrug resistance associated protein-2 displayed the prediction occurrence of 0.76, 0.72 and 0.79 for training, 5-fold cross-validation and external sets, respectively.

It was also observed that the support vector machine model which was built on those features selected using recursive feature elimination method, displays the best performance. This model can be used for identifying the inhibitors of multidrug resistance associated protein-2 efflux transporter.

###

To obtain the article please visit http://www.eurekaselect.com/166564