QSAR studies of bioactivities of 1-(azacyclyl)-3-arylsulfonyl-1H-pyrrolo[2, 3-b]pyridines as 5-HT6 receptor ligands using physicochemical descriptors and MLR and ANN-modeling

Mohammad Goodarzi, Matheus P. Freitas, Nahid Ghasemi

Research output: Contribution to journalArticle

20 Citations (Scopus)

Abstract

Four molecular descriptors were selected from a pool of variables using genetic algorithm, and then used to built a QSAR model for a series of 1-(azacyclyl)-3-arylsulfonyl-1H-pyrrolo[2,3-b]pyridines as 5-HT6 receptor agonists or antagonists, useful for the treatment of central nervous system disorders. Simple multiple linear regression (MLR) and a nonlinear method, artificial neural network (ANN), were used to model the bioactivities of the compounds; while MLR gave an acceptable model for predictions, the ANN-based model improved significantly the predictive ability, being more reliable for the prediction and design of novel 5-HT6 receptor ligands. Topology and molecular/group sizes are important requirements to take into account during the development of novel analogs. Few physicochemical descriptors were used to model the bioactivities of a series of 1-(azacyclyl)-3-arylsulfonyl-1H-pyrrolo[2,3-b]pyridines as 5-HT6 receptor inhibitors, giving predictive QSAR models, especially when using ANN as a nonlinear regression.

Original languageEnglish (US)
Pages (from-to)3911-3915
Number of pages5
JournalEuropean Journal of Medicinal Chemistry
Volume45
Issue number9
DOIs
StatePublished - Sep 1 2010

Fingerprint

Quantitative Structure-Activity Relationship
Bioactivity
Linear regression
Linear Models
Ligands
Neural networks
Neural Networks (Computer)
Central Nervous System Diseases
Neurology
serotonin 6 receptor
pyrrolo(2, 3-b)pyridine
Genetic algorithms
Topology

Keywords

  • 5-HT receptor ligands
  • Artificial neural network
  • Central nervous system disorders
  • Multiple linear regression
  • Quantitative structure-activity relationships

ASJC Scopus subject areas

  • Drug Discovery
  • Organic Chemistry
  • Pharmacology

Cite this

@article{99417d6b5bd44800b05bf6de21c5d8e2,
title = "QSAR studies of bioactivities of 1-(azacyclyl)-3-arylsulfonyl-1H-pyrrolo[2, 3-b]pyridines as 5-HT6 receptor ligands using physicochemical descriptors and MLR and ANN-modeling",
abstract = "Four molecular descriptors were selected from a pool of variables using genetic algorithm, and then used to built a QSAR model for a series of 1-(azacyclyl)-3-arylsulfonyl-1H-pyrrolo[2,3-b]pyridines as 5-HT6 receptor agonists or antagonists, useful for the treatment of central nervous system disorders. Simple multiple linear regression (MLR) and a nonlinear method, artificial neural network (ANN), were used to model the bioactivities of the compounds; while MLR gave an acceptable model for predictions, the ANN-based model improved significantly the predictive ability, being more reliable for the prediction and design of novel 5-HT6 receptor ligands. Topology and molecular/group sizes are important requirements to take into account during the development of novel analogs. Few physicochemical descriptors were used to model the bioactivities of a series of 1-(azacyclyl)-3-arylsulfonyl-1H-pyrrolo[2,3-b]pyridines as 5-HT6 receptor inhibitors, giving predictive QSAR models, especially when using ANN as a nonlinear regression.",
keywords = "5-HT receptor ligands, Artificial neural network, Central nervous system disorders, Multiple linear regression, Quantitative structure-activity relationships",
author = "Mohammad Goodarzi and Freitas, {Matheus P.} and Nahid Ghasemi",
year = "2010",
month = "9",
day = "1",
doi = "10.1016/j.ejmech.2010.05.045",
language = "English (US)",
volume = "45",
pages = "3911--3915",
journal = "European Journal of Medicinal Chemistry",
issn = "0223-5234",
publisher = "Elsevier Masson SAS",
number = "9",

}

TY - JOUR

T1 - QSAR studies of bioactivities of 1-(azacyclyl)-3-arylsulfonyl-1H-pyrrolo[2, 3-b]pyridines as 5-HT6 receptor ligands using physicochemical descriptors and MLR and ANN-modeling

AU - Goodarzi, Mohammad

AU - Freitas, Matheus P.

AU - Ghasemi, Nahid

PY - 2010/9/1

Y1 - 2010/9/1

N2 - Four molecular descriptors were selected from a pool of variables using genetic algorithm, and then used to built a QSAR model for a series of 1-(azacyclyl)-3-arylsulfonyl-1H-pyrrolo[2,3-b]pyridines as 5-HT6 receptor agonists or antagonists, useful for the treatment of central nervous system disorders. Simple multiple linear regression (MLR) and a nonlinear method, artificial neural network (ANN), were used to model the bioactivities of the compounds; while MLR gave an acceptable model for predictions, the ANN-based model improved significantly the predictive ability, being more reliable for the prediction and design of novel 5-HT6 receptor ligands. Topology and molecular/group sizes are important requirements to take into account during the development of novel analogs. Few physicochemical descriptors were used to model the bioactivities of a series of 1-(azacyclyl)-3-arylsulfonyl-1H-pyrrolo[2,3-b]pyridines as 5-HT6 receptor inhibitors, giving predictive QSAR models, especially when using ANN as a nonlinear regression.

AB - Four molecular descriptors were selected from a pool of variables using genetic algorithm, and then used to built a QSAR model for a series of 1-(azacyclyl)-3-arylsulfonyl-1H-pyrrolo[2,3-b]pyridines as 5-HT6 receptor agonists or antagonists, useful for the treatment of central nervous system disorders. Simple multiple linear regression (MLR) and a nonlinear method, artificial neural network (ANN), were used to model the bioactivities of the compounds; while MLR gave an acceptable model for predictions, the ANN-based model improved significantly the predictive ability, being more reliable for the prediction and design of novel 5-HT6 receptor ligands. Topology and molecular/group sizes are important requirements to take into account during the development of novel analogs. Few physicochemical descriptors were used to model the bioactivities of a series of 1-(azacyclyl)-3-arylsulfonyl-1H-pyrrolo[2,3-b]pyridines as 5-HT6 receptor inhibitors, giving predictive QSAR models, especially when using ANN as a nonlinear regression.

KW - 5-HT receptor ligands

KW - Artificial neural network

KW - Central nervous system disorders

KW - Multiple linear regression

KW - Quantitative structure-activity relationships

UR - http://www.scopus.com/inward/record.url?scp=77955560416&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77955560416&partnerID=8YFLogxK

U2 - 10.1016/j.ejmech.2010.05.045

DO - 10.1016/j.ejmech.2010.05.045

M3 - Article

C2 - 20547432

AN - SCOPUS:77955560416

VL - 45

SP - 3911

EP - 3915

JO - European Journal of Medicinal Chemistry

JF - European Journal of Medicinal Chemistry

SN - 0223-5234

IS - 9

ER -