Development of Conformation Independent Computational Models for the Early Recognition of Breast Cancer Resistance Protein Substrates

MELISA GANTNER; MAURICIO DI IANNI; MARÍA ESPERANZA RUIZ; ALAN TALEVI; LUIS BRUNO-BLANCH

JOURNAL OF BIOMEDICINE AND BIOTECHNOLOGY

HINDAWI PUBLISHING CORPORATION

Lugar: New York; Año: 2013 vol. 2013 p. 1 - 12

1110-7243

ATP-Binding Cassette (ABC) efflux transporters are polyspecific members of the ABC superfamily that, acting as drug and metabolite carriers, provide a biochemical barrier against drug penetration and contribute to detoxification. Moreover, their overexpression is linked to multidrug resistance issues in a diversity of diseases (e.g. cancer). Breast Cancer Resistance Protein (BCRP) is the most expressed ABC efflux transporter throughout the intestine and the blood-brain barrier, limiting the oral absorption and brain bioavailability of its substrates. Early recognition of BCRP substrates is thus essential to optimize oral drug absorption, design of novel therapeutics for central nervous system conditions and overcome BCRP-mediated cross-resistance issues. Here, we present the development of an ensemble of ligand-based machine learning algorithms for the early recognition of BCRP substrates, from a database of 262 substrates and non-substrates compiled from literature. Such dataset was rationally partitioned into training and test sets by application of a 2-step clustering procedure. The models have been developed through application of Linear Discriminant Analysis to random subsamples of Dragon molecular descriptors. Simple data fusion and statistical comparison of partial areas under the curve of Receiving Operating Characteristic curves were applied to obtain the best 2-model combination, which presented 82% of overall accuracy in the training set and 74.5% of overall accuracy in the test set. These are remarkable results considering the broad substrate specificity of BCRP. Moreover, Receiving Operating Characteristic curves may be applied to attain an optimal, context-dependent balance between specificity and sensitivity of the model ensemble.