Drug-like Index:  A New Approach To Measure Drug-like Compounds and Their Diversity

Abstract

Combinatorial organic synthesis (combinatorial chemistry or CC) and ultrahigh-throughput screening (UHTS) are speeding up drug discovery by increasing capacity for making and screening large numbers of compounds. However, a key problem is to select the smaller set of "representative" compounds from a virtual library to make or screen. Our approach is to select drug-like as well as structurally diverse compounds. The compounds, which are not very drug-like, are less taken into account or excluded even if they contribute to the diversity of the collection. Hence, the first step in the compound selection is to rank compounds in drug-like "degree". To quantify the drug-like "degree", drug-like index (DLI) is introduced in this paper. A compound's DLI is calculated based upon the knowledge derived from known drugs selected from Comprehensive Medicinal Chemistry (CMC) database. The paper describes the way of this knowledge base is formed and the procedure for selecting drug-like compounds.

Multi‐target QSAR Study in the Analysis and Design of HIV‐1 Inhibitors

“…In our study, the widely applicable set of descriptors i.e., the GD 13 is selected. In addition, the DI descriptors 14,15 is also adopted for a complementary comparison.…”

“…On the other side, DI acts as an approach to measure drug-like compounds as initially presented by Xu et al 14 Then it was used and modified as a set of descriptors by MOE. 15 DI descriptors characterized the hierarchy of drug structures in terms of rings, links, and molecular frameworks.…”

“…By using the multi-task learning method, the synergy among different set of inhibitors was exploited and an efficient multi-target QSAR modeling for HIV-1 inhibitors was obtained. The general descriptor (GD) features 13 and drug-like (DI) features 14 for compound description were ranked according to their jointly importance in multi-target QSAR modeling respectively, which will provide useful clues for the design of novel multi-target HIV-1 inhibitors with increasing likelihood of successful therapies of HIV.…”

Gradual in Silico Filtering for Druglike Substances

“…29 A much more extensive set of rules was proposed by Xu and Stevenson. 4 Considering also nonorally administered drugs, Ghose et al yielded margins that comprised the preferred 50% and a less stringent range of 80% of druglike substances taking similar variables into account. 30 The latter methods correspond to a gradual filtering of the considered substances based on a series of molecular descriptors.…”

Development of a Method for Evaluating Drug-Likeness and Ease of Synthesis Using a Data Set in Which Compounds Are Assigned Scores Based on Chemists' Intuition

“…9 The concept of "drug-likeness," which should comprise all ADME/Tox properties, has also been discussed and modeled. [10][11][12][13][14][15][16][17] Most reports that attempt to discriminate between drug-like and non drug-like molecules use databases such as the CMC (MDL Comprehensive Medical Chemistry), 18 the MDDR (MDL Drug Data Report), 19 and the WDI (Derwent World Drug Index) 20 to specify molecules considered drug-like, while the ACD (MDL Available Chemicals Directory) 21 is used to specify non drug-like molecules. However, the WDI also clearly contains what we believe to be non drug-like molecules, while the ACD contains some drug-like molecules.…”