Development of classification- and regression-based QSAR models and in silico screening of skin sensitisation potential of diverse organic chemicals

Abstract

With the advancement of technology and industrial revolution, skin sensitisation has emerged as a major environmental and occupational health hazard. A chemical may induce B-cell and T-cell infiltration from lymph nodes resulting in irritation of skin and carcinoma on prolonged use. To minimise the animal study and also to reduce time and expenditure, development of in silico predictive models has achieved a considerable attention over the last few decades. In this study, we have developed classification- and regression-based QSAR models for skin sensitisation potential of 67 diverse organic chemicals. The developed models strongly suggest the importance of the number of H atoms in a molecule inferring that the unsaturated compounds are more skin-sensitising agent than the saturated compounds. Other two descriptors, and nCb − , signify the importance of the lipophilic character of the molecules. Here, we have also screened the DrugBank database (http://www.drugbank.ca) containing a large number of compounds using our developed models in an attempt to identify molecules with skin sensitisation potential.

Keywords::

• QSAR
• skin sensitisation
• EC₃
• LDA
• OECD

Acknowledgements

This research was funded by a major research project (KR) of Council of Scientific and Industrial Research, New Delhi. The authors thank the Department of Science and Technology (DST), Government of India, for awarding a Research fellowship under the INSPIRE scheme to SK. Ministry of Human Resources and Development, New Delhi, is thanked for a scholarship to AN.

Notes

^1. EUCLIDEAN (a program written in Cþþ) is developed and validated on known datasets by Pravin Ambure ([email protected]) of Drug Theoretics and Cheminformatics Laboratory, Jadavpur University, Kolkata, West Bengal, India.