Quantum Isostere Database: A Web-Based Tool Using Quantum Chemical Topology To Predict Bioisosteric Replacements for Drug Design

This paper introduces the 'Quantum Isostere Database' (QID), a Web-based tool designed to find bioisosteric fragment replacements for lead optimization using stored ab initio data. A wide range of original geometric, electronic, and calculated physical properties are stored for each fragment. Physical descriptors with clear meaning are chosen, such as distribution of electrostatic potential energy values across a fragment surface and geometric parameters to describe fragment conformation and shape from ab initio structures. Further fundamental physical properties are linked to broader chemical characteristics relevant to biological activity, such as H-bond donor and acceptor strengths. Additional properties with less easily interpretable links to biological activity are also stored to allow future development of QSAR/QSPR models for quantities such as pK(a) and solubility. Conformational dependence of the ab initio descriptors is explicitly dealt with by storing properties for a variety of low-energy conformers of each fragment. Capping groups are used in ab initio calculations to represent different chemical environments, based on background research into transferability of electronic descriptors [J. Comput. Chem. 2009, 30, 1300-1318]. The resulting database has a Web interface that allows medicinal chemists to enter a query fragment, select important chemical features, and retrieve a list of suggested replacements with similar chemical characteristics. Examples of known bioisosteric replacements correctly identified by the QID tool are given.