Coupled Moves Tutorial

The most up-to-date version of this tutorial is available on GitHub.

Introduction

In this activity, you will utilize the Coupled Moves [NO2015] protocol within Rosetta to find mutations capable of redesigning enzyme specificity for an alternate target ligand.

Coupled Moves utilizes a flexible backbone modeling approach, achieving a better performance than a comparable fixed backbone method for 16 out of 17 tested benchmark cases [NO2015].

Coupled Moves operates according as follows (Fig 3. from [NO2015]):

Run Coupled Moves

Note

If you desire to use coupled_moves on your own ligand/protein system, you will need to prepare any input ligands according to the Rosetta documentation

1. From within the coupled_moves .zip folder, open run.py in your editor of choice.

2. Find the coupled_moves_path at the top of run.py and check that it is set to the appropriate location of your compiled Rosetta coupled_moves binary.

3. Run python run.py. The full command line call to each instance of Rosetta will be displayed, and will look something like this:

   /home/user/rosetta/rosetta_src_2017.45.59812_bundle/main/source/bin/coupled_moves.static.linuxgccrelease -s /home/coupled_moves/2O7B_HC4/2O7B_with_HC4.pdb -resfile /home/coupled_moves/2O7B_HC4/2O7B.resfile -extra_res_fa /home/coupled_moves/2O7B_HC4/HC4_from_2O7B.params -mute protocols.backrub.BackrubMover -ex1 -ex2 -extrachi_cutoff 0 -nstruct 1 -coupled_moves::mc_kt 0.6 -coupled_moves::initial_repack false -coupled_moves::ligand_mode true -coupled_moves::fix_backbone false -coupled_moves::bias_sampling true -coupled_moves::boltzmann_kt 0.6 -coupled_moves::bump_check true -extra_res_fa /home/kyleb/algosb/coupled_moves/2O7B_HC4/MDO_from_2O7B.params

   Important flags explained:

   • -resfile is an input file that tells Rosetta which protein positions to design (sample side chain rotamers of any amino acid) or repack (only sample side chain rotamers of the wild type amino acid). For coupled moves, designable residues are usually chosen to be those in close proximity to the target ligand, and packable residues as any residues in close proximity to the design shell residues. See the Rosetta documentation for more information on resfiles.
   • -ex1 -ex2 -extrachi_cutoff tell Rosetta's side chain packing algorithm to sample extra subrotamers for chi1 and chi2 angles of all side chains (Packer documentation)
   • -mute suppresses extraneous output from printing at the command line
   • -nstruct 1 run one independent Monte Carlo trajectory, producing one final output structure
   • -coupled_moves::fix_backbone false can be set to true to compare coupled move's performance when the backrub sampling step is skipped.
   • -coupled_moves::boltzmann_kt 0.6 the Boltzmann acceptance temperature
   • -coupled_moves::ligand_weight 1.0 can be set to greater than 1.0 to upweight ligand-protein interactions
   • -coupled_moves::ntrials 20 is normally set to 1000, meaning 1000 coupled moves trials are attempted

4. Output will be saved in a new directory named output

Analysis

Normally, you would run coupled_moves 20+ times, with many trials, for a single set of inputs in order to generate enough evaluated sequences for informative output. In the interest of time, we have set run.py to create only a few output structures and run for only a few trials. Instead of analyzing the output generated in output, you should proceed with the rest of the activity by extracting tar -xf example_output.tgz in the current folder (if the folder example_output does not already exist).

Python analysis

Three Python packages are required in order to run the analysis, and can be installed via pip: pip install numpy cogent weblogo (they are already installed if you are using the tutorial virtual machine).

Run the analysis script as follows:

python analyze_coupled_moves.py example_output/3HG5_A2G example_output/3HG5_GLA

If you cannot get the analysis script to run successfully, example output can be found in example_output.tgz as example_output/analysis.txt.

The analysis script will compare the distributions of output sequences for 3HG5_A2G over 3HG5_GLA, which are mutations enriched in the non-native substrate (A2G/N-acetyl-galactosamine) over the native substrate (GLA/galactose) in the wild type crystal structure (3HG5). Looking for enrichment of mutations in the mutant profile compared to the wild type profile helps identify specificity-switching mutations, as can be seen upon examination of the individual output sequence profiles:

Left: Sequence profile predicted by coupled moves for 3HG5 with its native substrate galactose. Right: Sequence profile predicted for 3HG5 and non-native substrate N-acetyl-galactosamine.

Structure analysis

Using PyMOL (or your preferred protein visualization software of choice), load the wild type crystal structure with the native substrate ligand (3HG5_GLA/3HG5_with_GLA.pdb) and the wild type crystal structure with non-native substrate (3HG5_A2G/3HG5_with_A2G.pdb). Focus your examination on the protein environment around each ligand, especially the residues that are designed: 170, 203, 206, 207, 227, 229, and 231.

Discussion questions

• Why is enrichment a useful metric to find specificity switching mutations?
• After examination of the output sequence profiles and the structure bound to native and non-native substrates, which highly enriched mutations would you choose as most likely to produce the desired specificity switch?

References

[NO2015]

(1, 2, 3) Noah Ollikainen, René M. de Jong, and Tanja Kortemme. Coupling Protein Side-Chain and Backbone Flexibility Improves the Re-design of Protein-Ligand Specificity. PLOS Comput Biol, 11(9):e1004335, September 2015. ISSN 1553-7358. doi: 10.1371/journal.pcbi.1004335. URL http://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1004335

[CS2008]

Colin A. Smith and Tanja Kortemme. Backrub-Like Backbone Simulation Recapitulates Natural Protein Conformational Variability and Improves Mutant Side-Chain Prediction. Journal of Molecular Biology, 380(4): 742–756, July 2008. ISSN 0022-2836. doi: 10.1016/j.jmb.2008.05.023. URL http://www.sciencedirect.com/science/article/pii/S0022283608005779.