Journal of Chemical Information and Modeling 2017-11-28

Macrocycle Conformational Sampling by DFT-D3/COSMO-RS Methodology

Ondrej Gutten, Daniel Bím, Jan Řezáč, Lubomir Rulisek

文献索引：10.1021/acs.jcim.7b00453

全文：HTML全文

摘要

To find and calibrate a robust and reliable computational protocol for mapping conformational space of medium-sized molecules, exhaustive conformational sampling has been carried out for the series of seven macrocyclic compounds of varying ring size and one acyclic analogue. While five of them were taken from the MD/LLMOD/force-field study by Shelley and coworkers (J. Chem. Inf. Model. 2014, 54, 2680), three represent potential macrocyclic inhibitors of human cyclophilin A. The free energy values (GDFT/COSMO-RS) for all conformers of each compound were obtained by the composite protocol based on the in vacuo quantum mechanics (DFT-D3 method in a large basis set), standard gas-phase thermodynamics and the COSMO-RS solvation model (QM+COSMO-RS). The GDFT/COSMO-RS values were used as the reference for evaluating performance of conformational sampling algorithms: standard and extended MD/LLMOD search (simulated annealing molecular dynamics with the low-lying eigen-vector following algorithms; employing OPLS2005 force field + GBSA solvation) available in Macromodel and the plain molecular dynamics sampling at high temperature (1000K) using the semiempirical potential, SQM(PM6-D3H4/COSMO) followed by the energy minimization of the snapshots. It has been shown that the former protocol (MD/LLMOD) may provide more complete set of initial structures that ultimately leads to identifying a greater number of low-energy conformers (assessed by the GDFT/COSMO-RS) than the latter, i.e. plain SQM MD. The CPU time needed to fully evaluate one medium-sized compound (~100 atoms, typically resulting in several hundreds or few thousands of conformers generated and treated quantum mechanically) is approximately 10 000 – 100 000 CPU hours on today’s computers which transforms to 1‑7 days on a small-sized computer cluster with few hundred nodes. Finally, our data sets based on the rigorous quantum chemical approach allow us to formulate a composite conformational sampling protocol with multiple checkpoints and truncation of redundant initial structural data that offers superior insights at affordable computational cost.