Exploration of nitroimidazoles as radiosensitizers: application of multilayered feature selection approach in QSAR modeling
- 31 Downloads
Radiosensitizers are aimed to augment tumor cell killing by radiation while having much less effect on normal tissues. Nitroimidazoles and related analogues are efficient radiation sensitivity enhancers, and they particularly work on hypoxic tumor cells. In the current study, we have developed two partial least squares (PLS) regression-based two-dimensional quantitative structure-activity relationship (2D-QSAR) models using a novel class of 84 nitroimidazole compounds to understand their radiosensitization effectiveness (pC1.6). Feature selection was done by genetic algorithm along with stepwise regression, while model validation was performed using various stringent validation criteria following the strict rules of OECD guidelines of QSAR validation. The variables included in the models were obtained from Dragon (version 7.0) and simplex representation of molecular structures (SiRMS) (version 188.8.131.520) software. The developed models were robust, externally predictive, and useful tools to predict the radiosensitization effectiveness of nitroimidazole compounds. True external prediction was carried out using a group of six nitroimidazole derivatives and the model reliability was checked using the Prediction Reliability Indicator tool (http://dtclab.webs.com/software-tools). Furthermore, the developed models will give an insight for development of new radiosensitizers with enhanced radiation sensitivity.
KeywordsRadiosensitizers Radiosensitization effectiveness QSAR SiRMS
PD thanks Indian Council of Medical Research, New Delhi, for awarding with a Senior Research Fellowship. KR thanks BRNS, Department of Atomic Energy, Govt. of India, for a Major Research Project (36(3)/14/08/2017-BRNS).
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
- 9.Roy K (2018) Quantitative structure-activity relationships (QSARs): a few validation methods and software tools developed at the DTC laboratory. J Indian Chem Soc 95:1497–1502Google Scholar
- 10.Hansch C, Leo A, Hoekman DH (1995) Exploring QSAR: fundamentals and applications in chemistry and biology. American Chemical Society Washington, DCGoogle Scholar
- 16.Roy H, Nandi S (2019) In silico modeling in drug metabolism and interaction: current strategies of lead discovery. Bentham Science Publishers, SharjahGoogle Scholar
- 27.MarvinSketch software, https://www.chemaxon.com. Accessed 26 Aug 2019
- 28.Dragon version 7, Kodesrl, Milan, Italy, 2016; software available at http://www.talete.mi.it/index.htm. Accessed 26Aug 2019
- 32.Drug Theoretics and Cheminformatics (DTC) laboratory software tools https://dtclab.webs.com/software-tools Accessed 28 Aug 2019
- 34.Devillers J (1996) Genetic algorithms in molecular modeling. Academic Press, Cornwall, Great BritainGoogle Scholar
- 37.U. Simca-P, 10.0, firstname.lastname@example.org, www.umetrics.com, Umea, Sweden, 2002. Accessed 30 Aug 2019
- 42.Gadaleta D, Mangiatordi GF, Catto M, Carotti A, Nicolotti O (2016) Applicability domain for QSAR models: where theory meets reality. IJQSPR 1:45–63Google Scholar