Defining Known Drug Space Using DFT

Matuszek, Anna M.; Reynisson, Jóhannes

Abstract

A density functional theory (DFT) study was performed on a collection of clinically approved drugs, or Known Drug Space (KDS), to determine the statistical distribution of four properties: dipole moment (DM), polarisability (POL), ionisation potential (IP) and electron affinity (EA). The DM and POL are linked to cell permeability of drugs whereas IP and EA reflect their redox stability thus ease of metabolism. A benchmarking exercise showed a good correlation between experimental values and their predicted counterparts. It was found that KDS occupies the volume of chemical space defined by: DM≤10 D, POL≤68 Å3, IP 6.0–9.0 V and EA−1.5–2.0 eV. Only 16 % of the drugs are outside one or more of these parameters. Three categories based on known oral absorption and bioavailability (low/medium/high) were established and compared. Predominately, drugs designated as ‘low’ were found outside the established parameters. The properties were compared with mainstream molecular descriptors and a strong correlation was seen for POL to MW (r2=0.899), which can explain the success of the latter since POL reflects the ability of molecules to interact with polar and non-polar environments such as water and interior of a membrane.

Citation

Matuszek, A. M., & Reynisson, J. (2016). Defining Known Drug Space Using DFT. Molecular Informatics, 35(2), 46-53. https://doi.org/10.1002/minf.201500105

• Journal Article Type: Article
• Online Publication Date: Nov 3, 2015
• Publication Date: 2016-02
• Deposit Date: Jun 8, 2023
• Journal: Molecular Informatics
• Print ISSN: 1868-1743
• Publisher: Wiley-VCH Verlag
• Peer Reviewed: Peer Reviewed
• Volume: 35
• Issue: 2
• Pages: 46-53
• DOI: https://doi.org/10.1002/minf.201500105
• Keywords: Organic Chemistry; Computer Science Applications; Drug Discovery; Molecular Medicine; Structural Biology; B3LYP; chemical space; molecular descriptors; chemoinformatics and drug discovery