MODIFIED GENETIC ALGORITHM for CRYSTAL STRUCTURE Prediction

M. FERRARO; D. GRILLO; A. ORENDT; J. FACELLI

Santiago de Compostela

Conferencia; Ninth Triennial Congress of the WORLD ASSOCIATION OF THEORETICAL AND COMPUTATIONAL CHEMISTS WATOC 2011; 2011

WATOC

<!-- @page { margin: 0.79in } P { margin-bottom: 0.08in } --> This presentation reports recent advances in developing a priory scanning techniques to control the population of crystal structures when using genetic algorithms in crystal structure predictions (CSP). The authors have developed over that last ten years a powerful method to predict crystal structures using genetic algorithms (GA) and empirical potentials.1 MGAC is capable of performing crystal structure searches for flexible molecules within any space group and with an arbitrary number of molecules in the asymmetric unit. The method has been applied to small molecules of pharmaceutical interest 2 and used to participate in the last three CSP blind tests organized by the Cambridge Crystallographic Data Center.3 The MGAC distributed computing framework includes a series of tightly integrated computer programs for generating the molecule’s force field, sampling possible crystal structures using a distributed parallel genetic algorithm, locally minimization of the structures and classifying, sorting and archiving the most relevant ones. Our results indicate that the method consistently can find the experimentally known structures when the GAFF force field reproduce the torsional energetics of the molecule, but unfortunately in some cases GAFF exhibit serious errors in describing this energetics. In recent studies of large pharmaceutical molecules the empirical potential is prone to predict very low energies for non plausible structures that are far from the equilibrium; they even dominated the population, to such extend that no physically possible structures resulted from the MGAC runs. A priori methods to preclude these bad structures to enter into the population appear a possible solution to avoid performance degradation due to either false positives or negatives. Exclusion of the former ones leads to an artificial restriction of the search space and incorporation into the population of the second ones leads to poisoning of the population with bad structures. Here we present the preliminary results of using a combination of two different exclusion criteria, one based on the empirical computation of the crystal density and second based on the calculation of the direct atomic contacts. While the second method is computationally more expensive, it is suitable for our purposes because it avoids the appearance of false positives produced by the volume control approach. These approaches have just been implemented in the MGAC code and tested in some pharmaceutical drugs for which MGAC has not been able to find good structures. <!-- @page { margin: 0.79in } P { margin-bottom: 0.08in } --> 1-Bazterra, V. E.; Thorley, M.; Ferraro, M. B.; Facelli, J. C., A Distributed Computing Method for Crystal Structure Prediction of Flexible Molecules: An Application to N-(2-Dimethyl-4-5-dinitrophenyl) Acetamide, J. Chem. Theory and Comp. 2007, 3, 201-209. 2-Kim, S.; Orendt, A. M.; Ferraro, M. B.; Facelli, J. C., Crystal Structure Prediction of Flexible Molecules Using Parallel Genetic Algorithms with a Standard Force Field, J. Comp. Chem. 2009, 30, 1973-1985. 3-Day, G. M.; Cooper, T. G.; Cruz-Cabeza, A. J.; Hejczyk, K. E.; Ammon, H. L.; Boerrigter, S. X. M.; Tan, J. S.; Della Valle, R. G.; Venuti, E.; Jose, J.; Gadre, S. R.; Desiraju, G. R.; Thakur, T. S.; van Eijck, B. P.; Facelli, J. C.; Bazterra, V. E.; Feraro, M. B.; Hofmann, D. W. M.; Neumann, M. A.; Leusen, F. J. J.; Kendrick, J.; Price, S. L.; Misquitta, A. J.; Karamertzanis, P. G.; Welch, G. W. A.; Scheraga, H. A.; Arnautova, Y. A.; Schmidt, M. U.; van de Streek, J.; Wolf, A. K.; Schweizer, B., Significant progress on predicting the crystal structures of small organic molecules – a report on the fourth blind test. Acta Crystallogr. 2009, B65, 107-125.