Machine Learning Models for Predictive Studies

In the context of drag design, biological activity is a function of the descriptor or property, so the general form of a ML model can be given as  

X refers to model parameters and y denotes model output deseribing activity/ property/toxicity (Fig. 3.2). 

The main task of ML models in drag design context is to distinguish between active and inactive molecules in a given database. There are generally two types of models that can be developed, viz. continuous and binary, depending upon the type 

The major ML-based predictive models in drag design comprise the following 

Quantitative Structure-Activity Relationship (QSAR) Models 

---