Ensembler: Enabling High-Throughput Molecular Simulations at the Superfamily Scale
Fig 5
Distribution of RMSDs to all TK catalytic domain models relative to the model derived from the highest sequence identity template.
Distributions are built from data from all 93 TK domain targets. To better illustrate how conformational similarity depends on sequence identity, the lower plot illustrates the distributions as stratified into three sequence identity classes: high identity (55–100%), moderate identity (35–55%), and remote identity (0–35%). The plotted distributions have been smoothed using kernel density estimation.