Virtual screening (VS) has emerged in drug discovery as a powerful computational\napproach to screen large libraries of small molecules for new hits with desired properties\nthat can then be tested experimentally. Similar to other computational approaches, VS\nintention is not to replace in vitro or in vivo assays, but to speed up the discovery\nprocess, to reduce the number of candidates to be tested experimentally, and to\nrationalize their choice. Moreover, VS has become very popular in pharmaceutical\ncompanies and academic organizations due to its time-, cost-, resources-, and laborsaving.\nAmong the VS approaches, quantitative structureâ??activity relationship (QSAR)\nanalysis is the most powerful method due to its high and fast throughput and\ngood hit rate. As the first preliminary step of a QSAR model development, relevant\nchemogenomics data are collected from databases and the literature. Then, chemical\ndescriptors are calculated on different levels of representation of molecular structure,\nranging from 1D to nD, and then correlated with the biological property using machine\nlearning techniques. Once developed and validated, QSAR models are applied to\npredict the biological property of novel compounds. Although the experimental testing\nof computational hits is not an inherent part of QSAR methodology, it is highly desired\nand should be performed as an ultimate validation of developed models. In this minireview,\nwe summarize and critically analyze the recent trends of QSAR-based VS\nin drug discovery and demonstrate successful applications in identifying perspective\ncompounds with desired properties. Moreover, we provide some recommendations\nabout the best practices for QSAR-based VS along with the future perspectives of this\napproach.
Loading....