Supporting Information S1 - Replica Exchange Improves Sampling in Low-Resolution Docking Stage of RosettaDock

This Supporting Information file (PDF) contains supporting Figures S1–S11, Tables S1–S3 and Methods S1–S7. Figure S1, The distribution of rotation angles. Figure S2, Example of refined decoys overlap with the RelaxedNative ensemble. Figure S3, Automated setup work flow. Figure S4, Detailed analysis of shotgun and ReplicaDock sampling on a bound target 1sq2. Figure S5, Initial survey of temperature selection on bound target 1sq2. Figure S6, Energy distribution of the three temperatures in ReplicaDock for all the benchmark targets. Figure S7, Frequent exchange between bound and unbound state. Figure S8, Different start conformation converge to the same populations. Figure S9, Fraction of hits in low-resolution decoys with different cutoffs. Figure S10, Interface RMSD vs. Interface Energy after refinement of ZDOCK and ReplicaDock ensembles. Figure S11, Interchain_contact dominates docking centroid energy. Table S1, Targets in the benchmark set. Table S2, summary of structure prediction accuracy in unbound docking. Table S3, Average exchange rate between temperature levels across the tested benchmark. Method S1, protocol_capture/rosetta_dock/centroid. Method S2, protocol_capture/replica_dock/centroid. Method S3, protocol_capture/rosetta_dock/refine. Method S4, protocol_capture/replica_dock/refine. Method S5, protocol_capture/relax_native. Method S6, Flags in commandlines in Method S1–S5. Method S7, Automated Setup. (PDF)