Ranking of QSAR models to predict minimal inhibitory concentrations toward Mycobacterium tuberculosis for a set of fluoroquinolones

Marjan Vračko, Nikola Minovski, K. Heberger

Research output: Contribution to journalArticle

19 Citations (Scopus)

Abstract

CP-ANN technique was used to build 54 different QSAR models. The models were built for three sets (assays) of fluoroquinolones considering their antituberculosis activity and using different technical parameters (dimension of network and number of learning epochs). The models served as a reliable basis for ranking by a new powerful method based on sum of ranking differences (SRD). With the applied SRD procedure we can find the optimal ones. The best model can be selected easily for the first assay. Two models can be recommended for the second assay, and no recommended model was found for the assay3.

Original languageEnglish
Pages (from-to)586-590
Number of pages5
JournalActa Chimica Slovenica
Volume57
Issue number3
Publication statusPublished - Sep 2010

Fingerprint

Fluoroquinolones
Assays

Keywords

  • Method comparison
  • Modeling
  • Neural networks
  • Prediction of antituberculosis activity
  • QSAR
  • Ranking

ASJC Scopus subject areas

  • Chemistry(all)

Cite this

Ranking of QSAR models to predict minimal inhibitory concentrations toward Mycobacterium tuberculosis for a set of fluoroquinolones. / Vračko, Marjan; Minovski, Nikola; Heberger, K.

In: Acta Chimica Slovenica, Vol. 57, No. 3, 09.2010, p. 586-590.

Research output: Contribution to journalArticle

@article{736df8ffe8d041e38752342e3f4fe4ad,
title = "Ranking of QSAR models to predict minimal inhibitory concentrations toward Mycobacterium tuberculosis for a set of fluoroquinolones",
abstract = "CP-ANN technique was used to build 54 different QSAR models. The models were built for three sets (assays) of fluoroquinolones considering their antituberculosis activity and using different technical parameters (dimension of network and number of learning epochs). The models served as a reliable basis for ranking by a new powerful method based on sum of ranking differences (SRD). With the applied SRD procedure we can find the optimal ones. The best model can be selected easily for the first assay. Two models can be recommended for the second assay, and no recommended model was found for the assay3.",
keywords = "Method comparison, Modeling, Neural networks, Prediction of antituberculosis activity, QSAR, Ranking",
author = "Marjan Vračko and Nikola Minovski and K. Heberger",
year = "2010",
month = "9",
language = "English",
volume = "57",
pages = "586--590",
journal = "Acta Chimica Slovenica",
issn = "1318-0207",
publisher = "Slovensko Kemijsko Drustvo",
number = "3",

}

TY - JOUR

T1 - Ranking of QSAR models to predict minimal inhibitory concentrations toward Mycobacterium tuberculosis for a set of fluoroquinolones

AU - Vračko, Marjan

AU - Minovski, Nikola

AU - Heberger, K.

PY - 2010/9

Y1 - 2010/9

N2 - CP-ANN technique was used to build 54 different QSAR models. The models were built for three sets (assays) of fluoroquinolones considering their antituberculosis activity and using different technical parameters (dimension of network and number of learning epochs). The models served as a reliable basis for ranking by a new powerful method based on sum of ranking differences (SRD). With the applied SRD procedure we can find the optimal ones. The best model can be selected easily for the first assay. Two models can be recommended for the second assay, and no recommended model was found for the assay3.

AB - CP-ANN technique was used to build 54 different QSAR models. The models were built for three sets (assays) of fluoroquinolones considering their antituberculosis activity and using different technical parameters (dimension of network and number of learning epochs). The models served as a reliable basis for ranking by a new powerful method based on sum of ranking differences (SRD). With the applied SRD procedure we can find the optimal ones. The best model can be selected easily for the first assay. Two models can be recommended for the second assay, and no recommended model was found for the assay3.

KW - Method comparison

KW - Modeling

KW - Neural networks

KW - Prediction of antituberculosis activity

KW - QSAR

KW - Ranking

UR - http://www.scopus.com/inward/record.url?scp=77957001536&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77957001536&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:77957001536

VL - 57

SP - 586

EP - 590

JO - Acta Chimica Slovenica

JF - Acta Chimica Slovenica

SN - 1318-0207

IS - 3

ER -