Feature-map vectors: a new class of informative descriptors for computational drug discovery

Abstract

In order to develop robust machine-learning or statistical models for predicting biological activity, descriptors that capture the essence of the protein-ligand interaction are required. In the absence of structural information from X-ray or NMR experiments, deriving informative descriptors can be difficult. We have developed feature-map vectors (FMVs), a new class of descriptors based on chemical features, to address this challenge. FMVs, which are derived from the conformational models of a few actives, are low dimensional, problem specific, and highly interpretable. By using shape-based alignments and scoring with chemical features, FMVs can combine information about a molecule's shape and the pharmacophores it can match. In five validation studies, bag classifiers built using FMVs have shown high enrichments for identifying actives for five diverse targets: CDK2, 5-HT(3), DHFR, thrombin, and ACE. The interpretability of these descriptors has been demonstrated for CDK2 and 5-HT(3), where the method automatically discovers the standard literature pharmacophore.

Deep Generative Models for 3D Compound Design

“…The shape and colour similarity score (SC RDKit ) uses two RDKit functions, based on the methods described in Putta et al 38 and Landrum et al 39 . The colour similarity function scores two 3D conformers against each other based on the overlap of their pharmacophoric features, while the shape similarity measure is a simple volumetric comparison between the two conformers.…”

Generation of three‐dimensional pharmacophore models

“…Note that a pharmacophore may be used to perform overlays, whereas the reverse is not true—e.g., if a set of amide bonds are overlaid, does one ascribe a H‐bond donor to the NH, or an acceptor to the CO, or to both? Also, note that methods such as comparative molecular field analysis10 or feature‐map vectors11 require an alignment as input, and hence aim to solve problem (2) given the results of (1) or (3).…”

Computational Intelligence in Medicine and Biology: A Survey

“…Since nowadays we speak about molecular and personalized medicine, 179 drug design should also be examined from this point of view. Therefore, there have been developed methodologies that include molecular modelling, molecular dynamics and docking procedures [180][181][182][183][184] in order to simulate the feasibility of interactions and the related affinity between ligand and target in order to acquire strong in silico evidences for drug design [185][186][187][188][189][190][191][192][193][194][195] and finally drug production.…”