Warnings during a WESTPA 2.0 run

[Razvan Marinescu]

Hi everyone,

I managed to run our system of two-dimers on WESTPA 2.0 by adapting tutorial 1 (NaCl). Currently it's using the classic rectangular binning. Everything works (I ran 20 iterations and got trajectories that made progress over 28 bins), but I get some warnings nonetheless. How can I debug and fix these warnings? See them below. 

I don't find anything wrong in the log files, and I cannot find where is the missing trajectory file. I've also attached the segment logs from one full iteration as well, in case they're helpful. 

Thanks,

Razvan

=== Warnings for ./init.sh ===

(westpa2) root@v100-2x-02:/mnt/twodimers# ./init.sh

--bstate-file /mnt/twodimers/bstates/bstates.txt
Updating system with the options from the configuration file
Creating HDF5 file '/mnt/twodimers/west.h5'
1 target state(s) present
Calculating progress coordinate values for basis states.
-- WARNING  [westpa.core.propagators.executable] -- could not read any data for trajectory: trajectory file not found
1 basis state(s) present
Calculating progress coordinate values for start states.
0 start state(s) present
Preparing initial states
-- WARNING  [westpa.core.propagators.executable] -- could not read any data for trajectory: trajectory file not found
-- WARNING  [westpa.core.propagators.executable] -- could not read any data for trajectory: trajectory file not found
-- WARNING  [westpa.core.propagators.executable] -- could not read any data for trajectory: trajectory file not found
-- WARNING  [westpa.core.propagators.executable] -- could not read any data for trajectory: trajectory file not found
-- WARNING  [westpa.core.propagators.executable] -- could not read any data for trajectory: trajectory file not found

        Total bins:            72
        Initial replicas:      8 in 1 bins, total weight = 1
        Total target replicas: 576

1-prob: 0.0000e+00
Simulation prepared.
1 of 72 (1.388889%) active bins are populated
per-bin minimum non-zero probability:       1
per-bin maximum probability:                1
per-bin probability dynamic range (kT):     0
per-segment minimum non-zero probability:   0.1
per-segment maximum non-zero probability:   0.2
per-segment probability dynamic range (kT): 0.693147
norm = 1, error in norm = 0 (0*epsilon)

=== Warnings in ./run.sh ===

(westpa2) root@v100-2x-02:/mnt/twodimers# ./run.sh

-- WARNING  [westpa.core.propagators.executable] -- could not read any data for trajectory: The topology is loaded by filename extension, and the detected ".dcd" format is not supported. Supported topology formats include ".pdb", ".pdb.gz", ".h5", ".lh5", ".prmtop", ".parm7", ".prm7", ".psf", ".mol2", ".hoomdxml", ".gro", ".arc", ".hdf5" and ".gsd".
Iteration completed successfully 

...

(same error repeated many times)
...

Iteration wallclock: 2:25:19.610810, cputime: 1 day, 18:39:46.031833

[Razvan Marinescu]

To reproduce the warnings, I put the files in a github repository: https://github.com/razvanmarinescu/twodimers2 ... The README has instructions on how to setup everything. 

Thanks,

Razvan 

[Anthony Bogetti]

Hi Razvan,

Thanks for uploading your files. The warnings you received are related to the new HDF5 trajectory storage framework in WESTPA 2.0. Make sure that before you copy in the trajectory storage and restarting files (in your get_pcoord.sh and runseg.sh) you first copy in the topologies. In your get_pcoord.sh, make sure to mirror these lines from the NaCl tutorial and copy in your pdb topology before the xml files for both WEST_TRAJECTORY_RETURN and WEST_RESTART_RETURN:

cp $WEST_SIM_ROOT/common_files/bstate.pdb $WEST_TRAJECTORY_RETURN

cp $WEST_STRUCT_DATA_REF/basis.xml $WEST_TRAJECTORY_RETURN

cp $WEST_SIM_ROOT/common_files/bstate.pdb $WEST_RESTART_RETURN

cp $WEST_STRUCT_DATA_REF/basis.xml $WEST_RESTART_RETURN/parent.xml

And similarly in your runseg.sh file, make sure you add some form of the following for both WEST_TRAJECTORY_RETURN and WEST_RESTART_RETURN:

cp bstate.pdb $WEST_TRAJECTORY_RETURN

cp seg.dcd $WEST_TRAJECTORY_RETURN

cp bstate.pdb $WEST_RESTART_RETURN

cp seg.xml $WEST_RESTART_RETURN/parent.xml

cp seg.log $WEST_LOG_RETURN

Without the topologies loaded in, the simulation will still work perfectly fine, but will not make use of the HDF5 restarting functionality.

Let me know if you have any further questions!

Best,

Anthony

-- 
You received this message because you are subscribed to the Google Groups "westpa-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to westpa-users...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/westpa-users/26d3bbb6-3730-4af1-969a-99dffe5778d9n%40googlegroups.com.

[Razvan Marinescu]

Hi Antony,

Many thanks, that makes sense. I initially used to have those lines in get_pcoord.sh, but I had to make a few changes:

1) In the openMM script (common_files/run_md.py) which runs the segments, I'm reading the bstate.pdb straight from common_files/bstate.pdb, so I don't need to copy it in every folder. This is how I read them in the MD simulator:

pdb = PDBFile('../../../common_files/bstate.pdb')
psf = CharmmPsfFile('../../../common_files/bstate.psf')

2) For basis.xml, how and where is that generated? Because when I switched from NaCl to my protein system, it used to give me an error that the number of atoms didn't match. So I now just generate it myself and copy it to the bstates/01 (I renamed it seg.xml). See the following excepts from get_pcoord.sh

cp $WEST_SIM_ROOT/common_files/bstate.psf bstate.psf
cp $WEST_SIM_ROOT/common_files/bstate.xml seg.xml
cp $WEST_SIM_ROOT/common_files/bstate.dcd seg.dcd

3) I had to change the topology from .xml to .psf, as I'm using a CHARMM force field. I'm using .xml just to save/resume the state of the simulation in openMM.

What are my solutions in this case?

Razvan

To view this discussion on the web visit https://groups.google.com/d/msgid/westpa-users/D270A77A-3576-4560-B31E-5D7D602619E4%40gmail.com.

[Razvan Marinescu]

Hi Anthony,

Actually, I managed to fix the above warnings (the way you suggested), by just copying those files (bstate.pdb and basis.xml) -- even though I don't actually use them anymore. If you know a way I can remove them completely from the HDF5 framework, let me know.  

The issue with the MAB framework still persists. We'll chat about that tomorrow on zoom.  

Thanks,

Razvan