DASH: a novel analysis method for molecular dynamics simulation data. Analysis of ligands of PPAR-gamma.

A novel molecular dynamics (MD) analysis algorithm, DASH, is introduced in this paper. DASH has been developed to utilize the sequential nature of MD simulation data. By adjusting a set of parameters, the sensitivity of DASH can be controlled, allowing molecular motions of varying magnitudes to be detected or ignored as desired, with no knowledge of the number of conformations required being prerequisite. MD simulations of three synthetic ligands of the orphan nuclear receptor PPARgamma were generated in vacuo using Tripos's SYBYL and used as the training set for DASH. Two X-ray crystal structures of PPARgamma complexed with Rosiglitazone were compared to gain knowledge of the pharmacophoric conformation; this showed that the conformation of the ligand is significantly different between the two structures, indicating that there is no distinct conformation in which rosiglitazone binds to PPARgamma but multiple binding modes. An investigation into simulation length was carried out. A simulation of 5 ns was found to give highly variable results, whereas a simulation of 25 ns gave a representative window of motion for molecules of this size. DASH was compared with Ward's hierarchical cluster analysis method. The results show that DASH analysis is as good as Ward analysis in some areas (e.g. conformation identification) and is superior in others (e.g. speed and input size).