Use of the Kohonen Neural Network for Rapid Screening of Ex Vivo Anti-HIV Activity of Styrylquinolines

Using the Kohonen neural network, the electrostatic potentials on the molecular surfaces of 14 styrylquinoline derivatives were drawn as comparative two-dimensional maps and compared with their known human immunodeficiency virus (HIV)-1 replication blocking potency in cells. A feature of the potenti...

Full description

Saved in:
Bibliographic Details
Published in:Journal of medicinal chemistry 2002-10, Vol.45 (21), p.4647-4654
Main Authors: Polanski, Jaroslaw, Zouhiri, Fatima, Jeanson, Laurence, Desmaële, Didier, d'Angelo, Jean, Mouscadet, Jean-François, Gieleciak, Rafal, Gasteiger, Johann, Le Bret, Marc
Format: Article
Language:English
Subjects:
Anti-HIV Agents - chemistry
Anti-HIV Agents - pharmacology
Antibiotics. Antiinfectious agents. Antiparasitic agents
Antiviral agents
Biological and medical sciences
Cell Line
HIV-1 - drug effects
Human immunodeficiency virus 1
Humans
Medical sciences
Neural Networks (Computer)
Pharmacology. Drug treatments
Quinolines - chemistry
Quinolines - pharmacology
Static Electricity
Structure-Activity Relationship
Styrenes - chemistry
Styrenes - pharmacology
styrylquinolines
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Using the Kohonen neural network, the electrostatic potentials on the molecular surfaces of 14 styrylquinoline derivatives were drawn as comparative two-dimensional maps and compared with their known human immunodeficiency virus (HIV)-1 replication blocking potency in cells. A feature of the potential map was discovered to be related with the HIV-1 blocking activity and was used to unmask the activity of further five analogues, previously described but whose cytotoxicity precluded an estimation of their activity, and to predict the activity of 10 new compounds while the experimental data were unknown. The measurements performed later turned out to agree with the predictions.
ISSN:0022-2623
1520-4804
DOI:10.1021/jm020845g