Insight II Notes: Simulated Annealing

To set up the Simulated Annealing:

• moleculename.car (the input coordiantes)
• moleculename.mdf (the molecular data file, containing info about bonding, charges, etc.)
• moleculename.inp (the "input" instructions for the simulated annealing calculation)

1. Build your structure using the Builder or Biopolymer modules in InsightII. Or get a PDB-style molecular coordinate file.
2. Using Forcefield/Potentials in Builder or Biopolymer, FIX your potentials, FIX your partial charges, and FIX your formal charges. If you get no error messages, go to the Discover module. If you get error messages, you must manually fix your molecule. See the Building molecules in InsightII page for details and tips.
3. Using the Constraints menus, fix regions of your molecules that should not change conformation, add distance restraints between pairs of atoms, add torsion restraints, tether atoms in regions of your molecule that should change their coordinates by a small amount, etc.
4. Click on Run/Run, and select the local machine, the molecule name, COMMAND FILE, auto-assign parameters, minimize, and Reduce Output. Then click EXECUTE

1. Set up the simulated annealing in a UNIX window:
   • Type unix on the InsightII command line. InsightII will ask command: (/bin/csh). Just hit the ENTER key, and a UNIX window will appear. To return to InsightII later on, type exit in the UNIX window.
   • OR you can just exit from InsightII. However, you MUST perform the following work in the UNIX window which is initialized for InsightII: either use the window from which you have already started and stopped InsightII, or first type setup_biosym to initialize your window.
2. Type ls. You should have three files, named moleculename.car (the input coordiantes), moleculename.mdf (the molecular data file, containing info about bonding, charges, etc.), and moleculename.inp (the "input" instructions for the simulated annealing calculation).
3. Edit the moleculename.inp file so that it contains appropriate "input" instructions for simulated annealing. See the examples of simulated annealing input files.
4. Type discover
   • Enter the prefix (moleculename) of your files.
   • Enter the number of the forcefield used to make the molecule (if you didn't select a particular forcefield in InsightII, choose the default CVFF forcefield, or forcefield #1).
   • Enter a "nice" number---higher "nice" numbers decrease the priority of your calculation relative to other calculations. The MolViz Facility policy is to set the "nice" number to 40 for long calculations, so that other users who want to interactively use the workstation are not battling with your non-interactive calculation for CPU time. Please be "nice".
   • On multi-CPU machines, type in the number of CPUs you want to use for this calculation. MolViz policy is to use only ONE CPU per calculation.
   • When the program asks you if you want to start (default=YES), just hit ENTER to start your calculation!
5. If you did not add a line to your moleculename.inp file to limit the size of the moleculename.his file, then delete the moleculename.his file. You usually don't need it, and it eats up mondo disk space.
6. To determine if your calculation is proceeding,
   • Type top to see a list of CPU-intensive calculations. Your calculation should be listed. Type q to quit.
   • Type ps -ef to see a list of all processes running on the computer. Your calculation ("fdiscover") should be listed.
   • Type ls and check if a new file named moleculename.pek is present. If so, type more moleculename.pek to "peek" at the status of your calculation.
   • Type tail -100 moleculename.out | grep Writing to count the number of structures saved to disk.
   • Get back into InsightII with the origional structure. You can always retrieve the origional structure by clicking on Molecule/Get and get the Biosym-format moleculename.car input coordinate file. in Discover, click on Run/Attach and attach the calculation to the original molecule. InsightiI will show the dynamic process of simulated annealing in real time. An attached calculation runs much slower than a detached calculation, since attached calculations have to update graphics. Click on Run/Detach to detach the calculation so that it may run in the background.
7. Log out
8. After a while, log back in.
9. To determine if the simulated annealing is finished:
   • Type top to see a list of CPU-intensive calculations. If your calculation is done, your calculation should NOT be listed. Type q to quit.
   • Type ps -ef to see a list of all processes running on the computer. If your calculation is done, your calculation ("fdiscover") should NOT be listed.
   • Type ls and check if a new file named moleculename.pek is present. If not, your calculation is done.
   • Type tail -100 moleculename.out | grep Writing to count the number of structures saved to disk. If your calculation is done, the total number of structures specified in moleculename.inp should be written to disk.

1. Set up the simulated annealing in a UNIX window:
   • Type unix on the InsightII command line. InsightII will ask command: (/bin/csh). Just hit the ENTER key, and a UNIX window will appear. To return to InsightII later on, type exit in the UNIX window.
   • OR you can just exit from InsightII. However, you MUST perform the following work in the UNIX window which is initialized for InsightII: either use the window from which you have already started and stopped InsightII, or first type setup_biosym to initialize your window.
2. Type ls. You should have three files, named moleculename.car (the input coordinates), moleculename.mdf (the molecular data file, containing info about bonding, charges, etc.), and moleculename.inp (the "input" instructions for the simulated annealing calculation).
3. Edit the moleculename.inp file so that it contains appropriate "input" instructions for simulated annealing. See the examples of simulated annealing input files.
4. For Discover_3, type setenv FORCEFIELD cvff.frc (or substitute cff91, amber, etc. for the forcefield you want to use). Then type discovery moleculename.inp > moleculename.log & to start your job in the background and forward error messages to the file moleculename.log.
5. To determine if your calculation is proceeding,
   • Type top to see a list of CPU-intensive calculations. Your calculation should be listed. Type q to quit.
   • Type ps -ef to see a list of all processes running on the computer. Your calculation ("fdiscover") should be listed.
   • Type ls and check if a new file named moleculename.pek is present. If so, type more moleculename.pek to "peek" at the status of your calculation.
   • Get back into InsightII with the original structure. You can always retrieve the origional structure by clicking on Molecule/Get and get the Biosym-format moleculename.car input coordinate file. in Discover_3, click on Background_Job/Control_Bkgd_Job/Attach_to_Job and attach the calculation to the original molecule. InsightII will show the dynamic process of simulated annealing in real time. An attached calculation runs much slower than a detached calculation, since attached calculations have to update graphics. Click on Run/Detach to detach the calculation so that it may run in the background.
6. Log out
7. After a while, log back in.
8. To determine if the simulated annealing is finished:
   • Type top to see a list of CPU-intensive calculations. If your calculation is done, your calculation should NOT be listed. Type q to quit.
   • Type ps -ef to see a list of all processes running on the computer. If your calculation is done, your calculation ("discovery") should NOT be listed.
   • Type ls and check if a new file named moleculename.pek is present. If not, your calculation is done.

1. Start up InsightII
2. Retrieve the original structure by clicking on Molecule/Get and get the Biosym-format moleculename.car input coordinate file.
3. Go to the Analysis module
4. To get the simulated annealing trajectory, click on Trajectory/Get. The trajectory file is named moleculename.arc, the trajectory object is the original molecule. You can choose to load all frames of the trajectory, or just a range of frames, or every Nth frame.
5. To see all frames sequentially, click on Trajectory/Animate. To stop the animation, click on Trajectory/Unanimate
6. The best way to evaluate the simulated annealing is to make a graph:
   • Click on Trajectory/Construct_Graph. Select New_Graph. Click on Execute.
   • Select an axis (e.g., X), a function type (e.g., frame), and execute. Repeat for a second axis, amd even a third axis. Graphs with frame/geometry/energy axes are particularly useful.
   • Click on End_Graph to end the graph definition.
   • Click on Graph/color to change the color of the graph's features.
7. A cluster graph is a graph of the RMS deviation of the atomic coordinates for structure #1 vs structre #2, structure #1 vs structure #3, structure #2 vs structure #3, etc. Cluster graphs show similarities and differences between groups (or "clusters") of structures. See the InsightII manuals or contact the MolViz Facility Staff for details about how to interpret cluster graphs.
   To make a cluster graph:
   • Click on Trajectory/Construct_Graph. Select New_Graph and Cluster_Graph. Click on Execute
   • Select the atom set(s) and molecule spec(s) that should be compared for each pair of molecules.
   • After all regions for atomic comparisons have been selected, click on end_definition
   • IF you are evaluating many pairs of molecules, InsightII will tell you that you are about to compare lots of structures with each other; 30,000 comparisons takes ~30 minutes on chemvgx. If you think your comparisons are going to take too long, click on Done and then Cancel the Cluster_trajectory menu. Click on Trajectory/Construct_Graph and in this menu select fewer structures to compare.
   • Type in a Maximum RMS Value and the Number of RMS levels. Only trial-and-error can help you choose th best values for your particular calculation, but first try Maximum RMS Value = 4 and number of RMS levels = 8.
8. To view a particular structure, click on Trajectory/Conformation and select a particular frame #, plot point on a graph, etc.