RESUMO
Proteins are essential macromolecules for the maintenance of living systems. Many of them perform their function by interacting with other molecules in regions called binding sites. The identification and characterization of these regions are of fundamental importance to determine protein function, being a fundamental step in processes such as drug design and discovery. However, identifying such binding regions is not trivial due to the drawbacks of experimental methods, which are costly and time-consuming. Here we propose GRaSP-web, a web server that uses GRaSP (Graph-based Residue neighborhood Strategy to Predict binding sites), a residue-centric method based on graphs that uses machine learning to predict putative ligand binding site residues. The method outperformed 6 state-of-the-art residue-centric methods (MCC of 0.61). Also, GRaSP-web is scalable as it takes 10-20 seconds to predict binding sites for a protein complex (the state-of-the-art residue-centric method takes 2-5h on the average). It proved to be consistent in predicting binding sites for bound/unbound structures (MCC 0.61 for both) and for a large dataset of multi-chain proteins (4500 entries, MCC 0.61). GRaSPWeb is freely available at https://grasp.ufv.br.
Assuntos
Aprendizado de Máquina , Proteínas , Proteínas/química , Sítios de Ligação , Ligantes , Domínios Proteicos , Ligação ProteicaRESUMO
Molecular dynamics simulations and binding free energy calculations were employed to examine the interaction between E-selectin and six structurally related oligosaccharides including the physiological ligand sialyl Lewis x. Molecular dynamics simulations revealed that sialyl Lewis x and its mimics share a common binding region and similar interactions with E-selectin involving the formation of hydrogen bonds with Glu80, Asn82, Asn83, Arg97, Asn105, Asp106, and Glu107 residues and electrostatic contacts with Ca2+ and the positively charged Lys111 and Lys 113 residues. Regarding binding free energy calculations, the performance of the rigorous but computationally expensive pathway methods TI, BAR, and MBAR was compared to the less rigorous but faster end-point methods MM/PBSA and MM/GBSA aimed at identifying a suitable approach to deal with the very subtle binding free energy differences within the ligands under study. All methods succeeded in predicting increased binding affinities for sialyl Lewis x analogs compared to the native ligand with absolute errors <1 kcal/mol. The best correlation with experimental data was obtained by TI (r2 = 0.84), followed by MBAR (r2 = 0.80), BAR (r2 = 0.73), MM/PBSA (r2 = 0.73) and MM/GBSA (r2 = 0.47). These results provide valuable information to increase understanding about E-selectin-oligosaccharide interactions and conduct further research aimed at designing novel ligands targeting this protein.