QSAR-Based Virtual Screening: Advances and Applications in Drug Discovery

Abstract

Virtual screening (VS) has emerged in drug discovery as a powerful computational tool to screen large libraries of small molecules for new hits with desired properties that can then be tested experimentally. Similar to other computational approaches, VS intention is not to replace in vitro or in vivo assays, but to speed up the discovery process, to reduce the number of candidates to be tested experimentally and to rationalize their choice. Moreover, VS has become very popular in pharmaceutical companies and academic organizations due to its time-, cost-, resources-, and labor-saving. When the biological target is not known or when its 3D structure or the binding mode of ligands are not available, quantitative structure-activity relationship (QSAR) analysis can be used in VS due to their high speed of screening and good hit rate. This approach is known as QSAR-based VS. In this method, compounds with known biological property are collected from databases and the literature. Then, after data curation, chemical descriptors are calculated on different levels of representation of molecular structure, ranging from 1D to nD, where n = 0-7, and then correlated with the biological property using machine learning methods. Once developed and validated, QSAR models are applied to predict the biological property of untested compounds. Although the experimental validation of computational hits does not represent part of the QSAR methodology, this should be performed as a final step of this approach. In this mini-review, we summarize and critically analyze the recent trends of QSAR-based VS approaches in drug discovery, and demonstrate successful applications in identifying perspective compounds with desired properties. Moreover, we provide some recommendations about best practices for the QSAR-based virtual screening along with the future perspectives of this approach.