Figure 3. Applying SHO for loop modeling discrimination.
For each of 45 target loops, 500 models were generated and then the lowest−energy model was selected using the Rosetta energy function with EEF1 or using the Rosetta energy function with SHO. The RMSD of the loop region was calculated for the model selected by each method. Each point on the plot represents a different target loop; points below the diagonal represent targets for which SHO led to selection of a more “native-like” model than EEF1. Excluding “ties” (cases in which the RMSDs for both methods were within 10% of one another), SHO outperforms EEF1 for 14 targets whereas EEF1 outperforms SHO for 7 targets.
