Figure 1
From: Fast structure similarity searches among protein models: efficient clustering of protein fragments
Ratio of the number of RMSD computations performed over N ( N - 1)/2 versus the logarithm of the number of structures in subsets from the semfold 1khm decoy set.