Automated docking for novel drug discovery

Abstract

Docking studies include many conformations to predict binding free energies (scoring functions) and to search (scoring sampling) for the most representative binding conformations. Therefore, several biological properties, from side chain residues to complete protein motions, have been included in docking studies to improve theoretical predictions.

Understanding the challenges of protein flexibility in drug design

“…Sampling and scoring engender different challenges, but have a combined influence on docking accuracy and performance [7], [8]. Issues related to scoring have been examined in numerous reviews [4]- [6], [9], [10], and will not be discussed here. Instead, the present paper focuses on sampling methods.…”

“…Various conformational sampling techniques have been proposed to explore ligand flexibility. The most common ones involve (i) a systematic search (e.g., exhaustive search, ensemble-based search, fragment-based search) or (ii) a stochastic search (e.g., Monte Carlo, evolutionary algorithm, swarm optimization, tabu search) [2], [5], [10]. Large ligands and peptides represent bigger challenges for docking, which might exceed the capabilities of many current approaches [11]- [15].…”

“…Large ligands and peptides represent bigger challenges for docking, which might exceed the capabilities of many current approaches [11]- [15]. The treatment of ligand flexibility has been extensively reviewed elsewhere [2], [5], [10] and will not be specifically discussed here. Instead, this review focuses on strategies used to account for protein flexibility (in addition to ligand flexibility).…”

Avances en el diseño de fármacos asistido por computadora

“…El acoplamiento molecular automatizado tiene como finalidad buscar la conformación y posición óptima de un ligando dentro de un blanco molecular o la posición y conformación más favorable de 2 macromoléculas (Bello, Martinez-Archundia y Correa-Basurto, 2013). El cribado virtual consiste en filtrar series de compuestos, normalmente bases de datos moleculares grandes, para seleccionar un subconjunto de moléculas que se sometan a ensayos biológicos (Lavecchia y Di Giovanni, 2013).…”

Incorporating specificity into optimization: evaluation of SPA using CSAR 2014 and CASF 2013 benchmarks

“…As the number of determined three-dimensional protein structures grows and the power of high-performance computing improves, computational virtual screening of large chemical databases is becoming an alternative method to seek the lead compounds in academia and pharmaceutical industry [2][3][4][5]. The fast and accurate prediction of a ligand tightly and specifically binding to a target protein is a crucial step for computational virtual screening.…”