Hi there!
I've been using MMB to fold ssDNA for subsequent MD simulation in Tinker with some success (it's a great tool by the way, thanks!). I'm now trying to do the same but with OpenMM, and I'm having difficulty with the MMB --> OpenMM translation.
Taking the output structure directly from MMB: firstly, pdbfixer's addHydrogen's method puts lots of protons in unphysical places, and secondly, whether or not we add hydrogens, the OpenMM system builder method always crashes with a message like this:
"
No template found for residue 2 (DC). The set of atoms matches DC, but the bonds are different.
"
Usually it is happy with the first base, but not the subsequent ones. It seems there may be an issue with the connections between bases (?)
I know this is the forum for MMB, not OpenMM, but I was wondering if there was any prior experience here making this transition. It seems in general the OpenMM pdb interpretation does not quite agree with the MMB outputs, and I'm not yet knowledgeable enough to guess how or why that is the case.
Running MMB outputs in OpenMM
-
Samuel Flores
- Posts: 188
- Joined: Mon Apr 30, 2007 1:06 pm
Re: Running MMB outputs in OpenMM
Hi Michael,
Thanks for raising this issue. Since birth molmodel has been naming the ribose atoms correctly using an apostrophe (e.g. C5') for the FIRST residue. However for reasons unknown it goes with stars for all subsequent residues (e.g. C5*). The latter is not consistent with PDB rules, so it is not surprising that you would have trouble. I would search-replace stars with apostrophes. That should fix the issue for now.
I have now fixed the issue in molmodel and checked it in. MMB also needed some adjustments. The latest docker image has these and seems to be working fine. If you don't mind using an unofficial release (bleeding-edge code) you can use that. I use the following syntax:
docker pull samuelflores/mmb-ubuntu
docker run -v $(pwd):/work --rm -it samuelflores/mmb-ubuntu MMB
It's probably not backward compatible, so if you read in older MMB nucleic acid structures, the new MMB will puke upon encountering asterisks.
Sam
Thanks for raising this issue. Since birth molmodel has been naming the ribose atoms correctly using an apostrophe (e.g. C5') for the FIRST residue. However for reasons unknown it goes with stars for all subsequent residues (e.g. C5*). The latter is not consistent with PDB rules, so it is not surprising that you would have trouble. I would search-replace stars with apostrophes. That should fix the issue for now.
I have now fixed the issue in molmodel and checked it in. MMB also needed some adjustments. The latest docker image has these and seems to be working fine. If you don't mind using an unofficial release (bleeding-edge code) you can use that. I use the following syntax:
docker pull samuelflores/mmb-ubuntu
docker run -v $(pwd):/work --rm -it samuelflores/mmb-ubuntu MMB
It's probably not backward compatible, so if you read in older MMB nucleic acid structures, the new MMB will puke upon encountering asterisks.
Sam
-
Samuel Flores
- Posts: 188
- Joined: Mon Apr 30, 2007 1:06 pm
Re: Running MMB outputs in OpenMM
super quick update.. the trunk version does actually appear to still read in the old asterisk-named atoms. Really it shouldn't, but there it is.
one more thing -- why are you adding hydrogens? MMB structures have hydrogens already.
Sam
one more thing -- why are you adding hydrogens? MMB structures have hydrogens already.
Sam
-
Michael Kilgour
- Posts: 7
- Joined: Mon Mar 08, 2021 5:14 pm
Re: Running MMB outputs in OpenMM
Hi Sam - I can't thank you enough for getting back to me with such quick and concise advice! Maybe it's something I would have caught with more experience in the pdb format but alas.
I did a search-replace " * " for " ' " and it looks like OpenMM is running the structure now just fine.
As to some updated version - I'm currently running the 2.14 Windows and Linux binaries from 2013 (?) - as long as I can keep this working I think I'll abstain from the bleeding edge code. I tried out the 3.4 Windows version that was just uploaded - seems to work fine but I'm planning to stick with what I know for now.
On adding hydrogens - I was just playing around with pdbfixer and thinking of maybe modelling different protonation states of other molecules mixed in with the DNA. Indeed for just the free aptamer in water I don't think I should need it.
I did a search-replace " * " for " ' " and it looks like OpenMM is running the structure now just fine.
As to some updated version - I'm currently running the 2.14 Windows and Linux binaries from 2013 (?) - as long as I can keep this working I think I'll abstain from the bleeding edge code. I tried out the 3.4 Windows version that was just uploaded - seems to work fine but I'm planning to stick with what I know for now.
On adding hydrogens - I was just playing around with pdbfixer and thinking of maybe modelling different protonation states of other molecules mixed in with the DNA. Indeed for just the free aptamer in water I don't think I should need it.