Gradient Boosting Based Classification of Ion Channels

Target-based drugs are one of the growing research areas to identify potential drugs for a particular target body part. The number of ion channels open at a particular time could be used to determine the effect of the drug. In some cases it may also help to determine the effects of the drugs at different concentrations and dosages, helping in developing target-based drugs. Thus by classifying the number of open ion channels at a particular time, it is possible to quantify the drug effect and explain varying effects of a drug based on the ions which are passing through those open channels. This paper works on the classification of ion channels by using gradient boosting algorithms including Light Gradient Boosting, eXtreme Gradient Boosting, and Categorical Boosting. The analysis of gradient boosting algorithms and comparison with the KNN classifier by using various metrics shows the significance of the work.