Ant colony optimization as a descriptor selection in QSPR modeling: Estimation of the λmax of anthraquinones-based dyes
Quantitative structure–property relationship (QSPR) studies based on ant colony optimization (ACO) were carried out for the prediction of λmax of 9,10-anthraquinone derivatives. ACO is a meta-heuristic algorithm, which is derived from the observation of real ants and proposed to feature selection. A...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2016-09-01
|
| Series: | Journal of Saudi Chemical Society |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S1319610313000422 |
| _version_ | 1818112650588979200 |
|---|---|
| author | Morteza Atabati Kobra Zarei Azam Borhani |
| author_facet | Morteza Atabati Kobra Zarei Azam Borhani |
| author_sort | Morteza Atabati |
| collection | DOAJ |
| description | Quantitative structure–property relationship (QSPR) studies based on ant colony optimization (ACO) were carried out for the prediction of λmax of 9,10-anthraquinone derivatives. ACO is a meta-heuristic algorithm, which is derived from the observation of real ants and proposed to feature selection. After optimization of 3D geometry of structures by the semi-empirical quantum-chemical calculation at AM1 level, different descriptors were calculated by the HyperChem and Dragon softwares (1514 descriptors). A major problem of QSPR is the high dimensionality of the descriptor space; therefore, descriptor selection is the most important step. In this paper, an ACO algorithm was used to select the best descriptors. Then selected descriptors were applied for model development using multiple linear regression. The average absolute relative deviation and correlation coefficient for the calibration set were obtained as 3.3% and 0.9591, respectively, while the average absolute relative deviation and correlation coefficient for the prediction set were obtained as 5.0% and 0.9526, respectively. The results showed that the applied procedure is suitable for prediction of λmax of 9,10-anthraquinone derivatives. |
| first_indexed | 2024-12-11T03:22:19Z |
| format | Article |
| id | doaj.art-efb198d96d4b4a65bb3e535b1c436e4c |
| institution | Directory Open Access Journal |
| issn | 1319-6103 |
| language | English |
| last_indexed | 2024-12-11T03:22:19Z |
| publishDate | 2016-09-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Journal of Saudi Chemical Society |
| spelling | doaj.art-efb198d96d4b4a65bb3e535b1c436e4c2022-12-22T01:22:37ZengElsevierJournal of Saudi Chemical Society1319-61032016-09-0120S1S547S55110.1016/j.jscs.2013.03.009Ant colony optimization as a descriptor selection in QSPR modeling: Estimation of the λmax of anthraquinones-based dyesMorteza AtabatiKobra ZareiAzam BorhaniQuantitative structure–property relationship (QSPR) studies based on ant colony optimization (ACO) were carried out for the prediction of λmax of 9,10-anthraquinone derivatives. ACO is a meta-heuristic algorithm, which is derived from the observation of real ants and proposed to feature selection. After optimization of 3D geometry of structures by the semi-empirical quantum-chemical calculation at AM1 level, different descriptors were calculated by the HyperChem and Dragon softwares (1514 descriptors). A major problem of QSPR is the high dimensionality of the descriptor space; therefore, descriptor selection is the most important step. In this paper, an ACO algorithm was used to select the best descriptors. Then selected descriptors were applied for model development using multiple linear regression. The average absolute relative deviation and correlation coefficient for the calibration set were obtained as 3.3% and 0.9591, respectively, while the average absolute relative deviation and correlation coefficient for the prediction set were obtained as 5.0% and 0.9526, respectively. The results showed that the applied procedure is suitable for prediction of λmax of 9,10-anthraquinone derivatives.http://www.sciencedirect.com/science/article/pii/S1319610313000422AnthraquinoneλmaxQSPRAnt colony optimization |
| spellingShingle | Morteza Atabati Kobra Zarei Azam Borhani Ant colony optimization as a descriptor selection in QSPR modeling: Estimation of the λmax of anthraquinones-based dyes Journal of Saudi Chemical Society Anthraquinone λmax QSPR Ant colony optimization |
| title | Ant colony optimization as a descriptor selection in QSPR modeling: Estimation of the λmax of anthraquinones-based dyes |
| title_full | Ant colony optimization as a descriptor selection in QSPR modeling: Estimation of the λmax of anthraquinones-based dyes |
| title_fullStr | Ant colony optimization as a descriptor selection in QSPR modeling: Estimation of the λmax of anthraquinones-based dyes |
| title_full_unstemmed | Ant colony optimization as a descriptor selection in QSPR modeling: Estimation of the λmax of anthraquinones-based dyes |
| title_short | Ant colony optimization as a descriptor selection in QSPR modeling: Estimation of the λmax of anthraquinones-based dyes |
| title_sort | ant colony optimization as a descriptor selection in qspr modeling estimation of the λmax of anthraquinones based dyes |
| topic | Anthraquinone λmax QSPR Ant colony optimization |
| url | http://www.sciencedirect.com/science/article/pii/S1319610313000422 |
| work_keys_str_mv | AT mortezaatabati antcolonyoptimizationasadescriptorselectioninqsprmodelingestimationofthelmaxofanthraquinonesbaseddyes AT kobrazarei antcolonyoptimizationasadescriptorselectioninqsprmodelingestimationofthelmaxofanthraquinonesbaseddyes AT azamborhani antcolonyoptimizationasadescriptorselectioninqsprmodelingestimationofthelmaxofanthraquinonesbaseddyes |