Anticancer Drug Discovery from Natural Compounds: Identification of Molecular Fingerprints for Inhibition of Breast Cancer Resistance Protein (BCRP) by Monte Carlo Optimization and Machine Learning Approaches

Posted: 3 Feb 2020

See all articles by Kalyan Ghosh

Kalyan Ghosh

Laboratory of Drug Design and Discovery, Department of Pharmaceutical Sciences

Arghya Banik

Dr. Harisingh Gour University

Date Written: November 1, 2019

Abstract

A promising hope for the treatment of breast cancer is different types of Natural Compounds. Extensive research over the past several decades has identified that numerous fruits, vegetables and dietary products have been used as chemo-preventive agents because of their safety, low toxicity, and anti-oxidant properties. The natural compounds that we consume contains a wide variety of plant secondary metabolite, which resembles drug molecules and takes part in various ADME reactions in human body. They also have the potential to interact with Breast Cancer Resistance Protein (BCRP) and regulate the ADME profile of several drugs. To identify potent BCRP inhibitors, in our present modelling study we have used a dataset of 124 natural compounds. In order to gain insight into the relationship between the molecular structures of these compounds with their anticancer activity various modelling approaches (Monte Carlo Optimization and Machine Learning Algorithm) were considered. The statistics of the best model by Monte Carlo Optimization (Sensitivity = 0.6786, Specificity = 1.0000, Accuracy = 0.7750, and MCC = 0.6227); Machine Learning Approaches: SPCI (Balance accuracy = 0.88, Sensitivity = 0.94, Specificity = 0.82, and Kappa = 0.77) and QSAR Co. (Sensitivity = 91.6667, Specificity = 69.2308, Accuracy = 83.7838, Precision = 84.6154, MCC = 0.6361, and AUROC = 0.737179). The best model obtained, was further used for prediction of some naturally occuring plant based anti-cancer compounds against BCRP. The two natural compounds like: Apigenin (found in Parsley, Celery, Celeriac, and Chamomile tea) and Naringenin (found in Grapefruit, Sour Orange, Tart Cherries, Tomatoes, Cocoa, and in Beans) have shown higher predicted inhibitory activity, which was further validated by Docking analysis. So, such kind of food supplements are highly recommended to the Breast Cancer patients to prevent development of resistance against Chemotherapy. Therefore, the combined modelling study will guide the medicinal chemist to act faster in the discovery of new potent natural inhibitors against Breast Cancer Resistance Protein (BCRP) in near future.

Keywords: BCRP, Natural Compounds, Monte Carlo optimization, Machine Learning

Suggested Citation

Ghosh, Kalyan and Banik, Arghya, Anticancer Drug Discovery from Natural Compounds: Identification of Molecular Fingerprints for Inhibition of Breast Cancer Resistance Protein (BCRP) by Monte Carlo Optimization and Machine Learning Approaches (November 1, 2019). Proceedings of International Conference on Drug Discovery (ICDD) 2020, Available at SSRN: https://ssrn.com/abstract=3529264

Kalyan Ghosh (Contact Author)

Laboratory of Drug Design and Discovery, Department of Pharmaceutical Sciences ( email )

UTD
Sagar
Sagar, MD Madhya Pradesh 470003
India
07679394667 (Phone)
713422 (Fax)

Arghya Banik

Dr. Harisingh Gour University ( email )

UTD
Sagar
Sagar, MD Madhya Pradesh 470003
India

Do you have negative results from your research you’d like to share?

Paper statistics

Abstract Views
449
PlumX Metrics