Back to Search
Start Over
Unify QSAR approach to antimicrobials. Part 1: Predicting antifungal activity against different species
- Source :
-
Bioorganic & Medicinal Chemistry . Sep2006, Vol. 14 Issue 17, p5973-5980. 8p. - Publication Year :
- 2006
-
Abstract
- Abstract: Most of up-to-date reported molecular descriptors encode only information about the molecular structure. In previous papers, we have extended stochastic descriptors to encode additional information such as target site, partition system, or biological species [Bioorg. Med. Chem. Lett. 2005, 15, 551; Bioorg. Med. Chem. 2005, 13, 1119]. This work develops an unify Markov model to describe with a single linear equation the biological activity of 74 drugs tested in the literature against some of the fungi species selected from a list of 87 species (491 cases in total). The data were processed by linear discriminant analysis (LDA) classifying drugs as active or non-active against the different tested fungi species. The model correctly classifies 338 out of 368 active compounds (91.85%) and 89 out of 123 non-active compounds (72.36%). Overall training predictability was 86.97% (427 out of 491 compounds). Validation of the model was carried out by means of leave-species-out (LSO) procedure. After elimination step-by-step of all drugs tested against one specific species, we record the percentage of good classification of leave-out compounds (LSO-predictability). In addition, robustness of the model to the elimination of the compounds (LSO-robustness) was considered. This aspect was considered as the variation of the percentage of good classification of the modified model (Δ) in LSO with respect to the original one. Average LSO-predictability was 86.41±0.95% (average±SD) and Δ =−0.55%, being 6 the average number of drugs tested against each fungi species. Results for some of the 87 studied species were Candida albicans: 43 tested compounds, 100% of LSO-predictability, Δ =−3.49%; Candida parapsilosis 23, 100%, Δ =−0.86%; Aspergillus fumigatus 21, 95.20%, Δ =0.05%; Microsporum canis 12, 91.60%, Δ =−2.84%; Trichophyton mentagrophytes 11, 100%, Δ =−0.51%; Cryptococcus neoformans 10, 90%, Δ =−0.90%. The present one is the first reported unify model that allows one predicting antifungal activity of any organic compound against a very large diversity of fungi pathogens. [Copyright &y& Elsevier]
- Subjects :
- *QSAR models
*ANTI-infective agents
*CRYPTOGAMS
*MARKOV processes
Subjects
Details
- Language :
- English
- ISSN :
- 09680896
- Volume :
- 14
- Issue :
- 17
- Database :
- Academic Search Index
- Journal :
- Bioorganic & Medicinal Chemistry
- Publication Type :
- Academic Journal
- Accession number :
- 21664597
- Full Text :
- https://doi.org/10.1016/j.bmc.2006.05.018