normal Backmapping Martini Lipid using Amber forcefield

  • ibrahim
  • ibrahim's Avatar Topic Author
  • Offline
  • Fresh Boarder
More
1 month 2 weeks ago - 1 month 2 weeks ago #8181 by ibrahim
Hello all,
I have a Martini Lipid-rna system, I want to backmap it using Amber forcefield.
However the Martini Mapping files for DOPC dopc.amber.map and dopc.charmm36.map has exactly the same atom names which corresponds to that of charmm naming. So i modified (by hand) the .map file into charmm atom names using the CSV mapping found with "charmmlipid2amber.py".
Now AMBER has same naming for several atoms like, both C32(charmm) and C22(charmm) maps to same atom name C12(amber) and there are several others. When I use the backward tool, it crashes and the intermediate file projected.gro has only 50 atoms (there are 138 in DOPC). I suspect this is due to same atom names for several atoms. My questions is,
(1) Where can I get dopc.amber.map with correct amber atom names?
(2) Given that amber has same atom names for several atoms for DOPC, does that causes backward.py to fail?
Any suggestions would be highly helpful
Thanks
Ibrahim
P.S I obtained itp files for DOPC using tleap and acpype and used Lipid14.ff . And I could backmap DOPC using charmm forcefield (and rna using amber) without any problem.
Last edit: 1 month 2 weeks ago by ibrahim.

Please Log in or Create an account to join the conversation.

More
1 month 1 week ago #8182 by riccardo
(0) You cannot have two atoms with the same atom name in dopc.amber.map. I think that's indeed what's going wrong.
(1) The "correct" atom names for AMBER itp may be the ones you have in the itp. So, the "incorrect" ones may be the ones used in dopc.amber.map -- the reason being point (0).
(2) See (0).

Given the above, if the order of the atoms in your DOPC_amber.itp file is the same as in the original dopc.amber.map (= taken from github.com/Tsjerk/Backward/tree/master/Mapping ), you can just replace all the atom names in your itp with the ones used in dopc.amber.map.

Please Log in or Create an account to join the conversation.

Time to create page: 0.132 seconds