search

patrickfuchs / buildH

:computer: Build hydrogen atoms from united-atom molecular dynamics of lipids and calculate the order parameters.
https://buildh.readthedocs.io/
BSD 3-Clause "New" or "Revised" License
14 stars 7 forks source link

Missing carbon atoms #164

Closed larabort closed 2 years ago

larabort commented 2 years ago

When implementing a data set with a DPPC bilayer using Poger GROMOS 53A6_L force field into the NMRlipids databank, I was wondering whether there are missing two lines in here: https://github.com/patrickfuchs/buildH/blob/master/buildh/lipids/GROMOS53A6L_DPPC.json

I suspect that there also have to be the two carbon atoms C1H and C2H and the related information: "C1H": ["CH2", "C1G", "C1I"], "C2H": ["CH2", "C2G", "C2I"],

Is that right?

patrickfuchs commented 2 years ago

Thanks @larabort for reporting this. There is indeed a pb. And it's even detected by buildH which raises this error:

$ buildH -c 1dppc.pdb -l GROMOS53A6L_DPPC -d ../buildH/def_files/GROMOS53A6L_DPPC.def
Constructing the system...
Some carbons (C1H,C2H) from the definition file are notpresent in the json file.
Atoms defined in ../buildH/def_files/GROMOS53A6L_DPPC.def are not consistent with the chosen topology.

I didn't check this lipid when it was added. Indeed your suggestion looks correct. I fix this ASAP and make you know.

patrickfuchs commented 2 years ago

I pushed the fix, it should be OK now. Could you give it a try?

HubLot commented 2 years ago

My bad here. I thought I tested the Poger FF before added the JSON file. I guess not...

larabort commented 2 years ago

Thank you for fixing that. Now it seems to be correct.

patrickfuchs commented 2 years ago

Thanks @larabort. So I close this issue.