(11g) Combining Big Data Analytics and Computational Chemistry in Target-Based Drug Discovery

A new target-based drug screening method is proposed exploiting the synergy effect of big data analytics and computational chemistry. The method was illustrated in the development of a target-based drug using 45 sulphonamide derivatives (33 training compounds and 12 testing compounds) with carbonic anhydrase IX (CA IX) as a drug target. For each sulphonamide compound, about 5,000 molecular descriptors were calculated using Gaussian (Gaussian Inc.), and lipophilicity (logk_w) and inhibitory activity (logK_i) were measured using RP-LC as drug properties. Genetic algorithm-partial least squares (GA-PLS) provided a quantitative structure-property-relationship (QSPR) model with high prediction capability employing only 7 descriptors. The ideal drug structure was obtained by inverting the QSPR model numerically in respect to reference properties of a reference drug (one of testing compounds). 12 testing compounds were ranked using the ideal drug structure and the rank was further validated through molecular docking and molecular dynamic (MD) simulation of drug candidate-CA IX complexes. Parameterization of the simulation was verified experimentally using structural analyses of the complexes including MALDI-TOF/TOF-MS. The results proved that the drug screening method proposed has great flexibility and can provide drug candidates with lower risk.