Please login first
TOMOCOMD-CARDD Method in Early Drug Discoverybased Rational Drug Selection of Antifungal Agents
1, 2 , * 1, 3, 4 , 1 , 1 , 2 , 3 , 4
1  Unit of Computer-Aided Molecular “Biosilico” Discovery and Bioinformatic Research (CAMD-BIR Unit), Faculty of Chemistry-Pharmacy. Central University of Las Villas (UCLV), Santa Clara, 54830, Villa Clara, Cuba
2  Department of Microbiology, Chemical Bioactive Center. Central University of Las Villas (UCLV), 54830, Villa Clara, Cuba
3  Institut Universitari de Ciència Molecular, Universitat de València, Edifici d'Instituts de Paterna, Poligon la Coma s/n (detrás de Canal Nou) P. O. Box 22085, E-46071 Valencia, Spain
4  Unidad de Investigación de Diseño de Fármacos y Conectividad Molecular, Departamento de Química Física, Facultad de Farmacia, Universitat de València, Spain

Abstract: The novel TOMOCOMD–CARDD approach has been introduced here for the classification and design of antifungal agents using computer-aided molecular design. For this purpose, no stochastic and stochastic atom-based quadratic fingerprinting were used to codify the antifungal-related chemical structure information from a comprehensive data set of 2478 organic compounds having a great structural variability, 1087 of them being antifungal agents covering the broadest antifungal mechanisms of action known so far. The two ligand-based antifungal-activity classification models obtained by using Linear Discriminant Analysis, including no stochastic and stochastic indices, classified correctly 90.73% and 92.47%, respectively, of 1772 chemicals in the training set. These models showed moderate-to-high Matthews correlation coefficients (MCC of 0.81 and 0.85) as well as a very good accuracy, sensitivity, specificity and false alarm rate. These models were able of classifying correctly 92.16% and 87.56% of 706 compounds in an external test set. In general, the TOMOCOMD–CARDD models were best in predicting antifungal activity when compared with six of the most recent models reported so far; indicating that this approach could be very useful to identify (design and/or select) new antifungal agents against life-threatening fungal infections.
Keywords: TOMOCOMD-CARDD Software; non-stochastic and stochastic atom-based quadratic indices; LDA-based QSAR model; Learning Machine Tools, Computational Screening, Antifungal Agent