MobleyLab / alchemical-setup

GNU Lesser General Public License v2.1
15 stars 9 forks source link

Dummy atomtypes must be added to forcefield library definitions? #9

Open rmuraglia opened 8 years ago

rmuraglia commented 8 years ago

Using the provided example output files from this package, I attempted to run a steepest descent minimization using typical parameters to test my installation. I get the following fatal warning at the grompp step:

" WARNING 1 [file forcefield.itp, line 1]: Too few gb parameters for type


Couldn't find topology match for atomtype **** Abort trap: 6 "

I believe this is because dummy atoms are denoted as "XX_dummy," but this atom type doesn't exist in a standard amber99sb forcefield. In other words, it is not present in my atomtypes.atp, ffbonded,itp nor ffnonbonded.itp files. I'm guessing there are some modifications I must make to these files to make it compatible with the dummy atoms, but it isn't clear to me how I should define them.

An upload of your custom forcefield library seems necessary to complete this distribution.

davidlmobley commented 8 years ago

Normally we just modify the topology file to add an [ atomtypes ] section at the top of your topology which defines these atom types (same as the normal atom types, but 0 LJ parameters). However, the tool needs to be modified so it does this itself. Modifying the forcefield .itp files is unnecessary since this info can go into your topology.

For your purposes, you might try adding that section and see if it eliminates this problem. However, ultimately we need to update this tool to do so.

On Sun, Mar 27, 2016 at 3:01 PM, rmuraglia notifications@github.com wrote:

Using the provided example output files from this package, I attempted to run a steepest descent minimization using typical parameters to test my installation. I get the following fatal warning at the grompp step:

" WARNING 1 [file forcefield.itp, line 1]:

Too few gb parameters for type

Couldn't find topology match for atomtype


Abort trap: 6 "

I believe this is because dummy atoms are denoted as "XX_dummy," but this atom type doesn't exist in a standard amber99sb forcefield. In other words, it is not present in my atomtypes.atp, ffbonded,itp nor ffnonbonded.itp files. I'm guessing there are some modifications I must make to these files to make it compatible with the dummy atoms, but it isn't clear to me how I should define them.

An upload of your custom forcefield library seems necessary to complete this distribution.

— You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub https://github.com/MobleyLab/alchemical-setup/issues/9

David Mobley Associate Professor Department of Pharmaceutical Sciences Department of Chemistry 3134B Natural Sciences I University of California, Irvine Irvine, CA 92697 dmobley@uci.edu work (949) 824-6383 cell (949) 385-2436

rmuraglia commented 8 years ago

Even when using the provided files in the output_files/ directory of this repository, which has that [atomtypes] directive in the topology, I run into the aforementioned issue. I'll attach my .top, .gro and .mdp files for reproducibility's sake.

The grompp call I make is the following (obviously rename the files to strip the .txt I added for github compatibility):

gmx grompp -f min_05.mdp -c out.gro -p out.top -o min_05.tpr

min_05.mdp.txt out.gro.txt out.top.txt

Let me know if additional files are needed from my end to make a minimum working example