Representative structures from RMSD-clustering of MD trajectories.

PDB files contain representative structures of RMSD-based clusters obtained from the simulation of each variant using the combined last 50ns of all three replicas of the simulation perfomed for each variant.

The clustering was performed on all heavy atoms of the side chains of the following residues: 13, 14, 50, 51, 72, 73, 74, 75, 113, 115, 138, 139, 155, 156, 157, 158, 159, 160, 211, 212, 317, 318, 321, 325, 330, 331, 375, as well as heavy atoms of the substrate. In each case the structures in each trajectory were aligned using backbone heavy atoms of the entire protein prior to clustering. Clustering was performed with cpptraj (Amber Tools) using average linkage algorithm.

Summary of the clustering for each system can be found in *summary.out files, which contain the list of all clusters determined through the clustering analysis, with number of frames in each cluster, total fraction of the simulation these frames consittute, average distance between points in the cluster (AvgDist), standard deviation of points in the cluster (Stdev), centroid frame (Centroid) and average distance of a given cluster to every other cluster (AvgCDist).

PDB files containing the representative structures of each cluster are labelled with the number corresponding to the number of a given cluster as listed in *summary.out file.