Chapter 1 - Quantitative structure-activity relationships (QSARs) in medicinal chemistry

Abstract

The quantitative structure-activity relationship (QSAR) is an important branch of in silico modeling approaches that has significant applications in medicinal chemistry, materials sciences modeling, predictive toxicology, agricultural science, food science, nanotechnology, etc. Different regulatory agencies promote the use of predictive modeling, especially QSARs, instead of experimental approaches to reduce the animal use and the wastage of chemicals. Thus, the importance of QSAR and the demand for subject experts are increasing. However, QSAR has gained both praise and criticism throughout its journey due to its reliability, successes, limitations, and failures. The modern methodology of QSAR model development is mostly oriented toward machine learning; however, conventional approaches are still in use. In this present chapter, we have reviewed the history of QSAR with its evolution. We have also elaborately discussed descriptors, modeling algorithms, various statistical approaches to model development, validation, and several other aspects of modeling. We have mentioned several limitations of QSAR along with suitable corrective measures. In the limited space of this chapter, we have tried to provide a few examples where the QSAR and SAR have been applied in successful drug design. As the chapter reaches its penultimate stage, we have further enriched it with the future trends of the domain along with the concluding remarks. We have a positive hope that this chapter will be handy to beginners as well as all the readers for understanding the subject, acquiring practical knowledge, and having guidance to perform the QSAR analysis.