Structural databases: A way of improving knowledge-based docking

LANZAROTTI, ESTEBAN; DEFELIPE, LUCAS ALFREDO; MARTI, MARCELO ADRIÁN; TURJANSKI, ADRIAN GUSTAVO

Santiago de Chile

Congreso; 2nd ISCB Latin America 2012 Conference on Bioinformatics; 2012

Drug discovery through docking simulations has proven successful strategy but has its limitations1. p38a (MAPK 14) is one of the chosen models for docking calculations because of the abundant structural data available. Using our structural database2 and binding data from BindingDB3 we performed a search of structural similarity between the drugs with crystal structures in p38a. Aromatic stacking and hydrogen bond acceptors and donors were found in the same relative positions. All the gathered data was used to place pharmacophores for aromatic and hydrogen bond acceptors/donors in a docking simulation of more than a thousand of MAPK14 drugs with binding data deposited in BindingDB. Our results show that using this strategy improves the quality of the obtained results and brings structural information on hundreds of drugs for which no crystal data is available. This methodology will be applied for all the drugs deposited in bindingDB in order to increase the available structural information for drug discovery.