novel approach of drug repurposing in immuno-oncology therapeutic agents using machine learning algorithm

Abstract

Drug repurposing processing plays a significant role in evaluating Drug Target identification in the field of Drug discovery. Today Machine learning algorithms based automated systems are gaining attention for creating new device applications in the field of artificial intelligence. [2] The present study used machine learning approaches to develop a computational predicted model for predicting the unknown pharmacological effects with immune-oncology compound. The authors applied similarity features on drug and disease to see indication also authors observed that if number of descriptor is increased then accuracy of model is also increased. Classification linear and nonlinear algorithms such as logistic regression,Support Vector Machine (Kernel) and Random Forest were used to build the computational models. We used 10 fold Cross-validation on dataset for train and test data to performed best (Accuracy = 0.932, AUC-ROC = 0.971), finally, the developed model was used to anticipate probable indications for currently available medications for cancer protein.