GROMACS version: 2020.2
GROMACS modification: No
Hi,
I have a docked protein-protein complex and performed md simulations. I want to make a hydrogen bond study between these 2 proteins using gmx hbond ( gmx hbond -f md.xtc -s md.tpr -num hbnum_Pro-Pro.xvg and issuing Protein , Protein (same atom numbers for both)) however, I believe this way I’m looking at intraprotein H bonds instead of interprotein H bonds, right?
The md simulation was run using a pdb file with the two protein chains in it with. I created an index file out of a combined pdb file containing non overlapping atom or residues numbers of these 2 proteins. However, when I run gmx hbond again I get this message: “Fatal error: Partial overlap between groups ‘Protein’ and ‘Protein’”. What should I do? Is the problem in the md files? Is there another way?
Thanks!
Your index groups are incorrect. You do need to set each protein unit as its own group, but however you created them was incorrect such that you have some atom numbers present in both groups, which should not happen.
Thank you for you reponse! This is how I’m doing it
Protein 1
ATOM 10 N ILE B 21 184.889 103.053 229.858 0.00 0.00
ATOM 11 CA ILE B 21 184.185 102.017 230.641 0.00 0.00
ATOM 12 CB ILE B 21 185.182 101.045 231.307 0.00 0.00
ATOM 13 CG2 ILE B 21 186.010 101.728 232.403 0.00 0.00
ATOM 14 CG1 ILE B 21 184.510 99.774 231.865 0.00 0.00
ATOM 15 CD1 ILE B 21 183.748 98.968 230.804 0.00 0.00
~~
ATOM 7077 CA PRO B 733 138.210 155.200 163.232 0.00 0.00
ATOM 7078 C PRO B 733 138.990 156.426 162.714 0.00 0.00
ATOM 7079 O PRO B 733 138.393 157.496 162.567 0.00 0.00
ATOM 7080 CB PRO B 733 137.008 154.870 162.339 0.00 0.00
ATOM 7081 CG PRO B 733 135.848 155.612 163.004 0.00 0.00
ATOM 7082 CD PRO B 733 136.195 155.511 164.488 0.00 0.00
Protein 2
ATOM 7083 N CYS E 735 170.844 109.510 284.170 0.00 0.00 N
ATOM 7084 CA CYS E 735 172.051 109.359 283.332 0.00 0.00 C
ATOM 7085 CB CYS E 735 171.724 108.412 282.167 0.00 0.00 C
ATOM 7086 SG CYS E 735 173.175 108.063 281.147 0.00 0.00 S
ATOM 7087 C CYS E 735 172.534 110.729 282.843 0.00 0.00 C
ATOM 7088 O CYS E 735 171.696 111.587 282.570 0.00 0.00 O
~~
ATOM 8544 OXT LEU E 917 181.415 89.289 278.174 0.00 0.00 O
ATOM 8545 CB LEU E 917 181.907 92.313 279.345 0.00 0.00 C
ATOM 8546 CG LEU E 917 183.294 91.660 279.447 0.00 0.00 C
ATOM 8547 CD1 LEU E 917 184.022 91.664 278.105 0.00 0.00 C
ATOM 8548 CD2 LEU E 917 184.142 92.455 280.442 0.00 0.00 C
TER 8549 LEU E 917 END
command:
gmx make_ndx -f protein1.pdb -o protein1.ndx
q or a 1 - 5822 or r 21 - 733 (I’ve tried these 3 ways)
cat protein1.ndx protein2.ndx > index.ndx
gmx hbond -f md.xtc -s md.tpr -num hbnum_Pro-Pro.xvg -n index.ndx
Group 0 ( System) has 5822 elements
Group 1 ( Protein) has 5822 elements
Group 2 ( Protein-H) has 5822 elements
Group 3 ( C-alpha) has 713 elements
Group 4 ( Backbone) has 2139 elements
Group 5 ( MainChain) has 2852 elements
Group 6 ( MainChain+Cb) has 3529 elements
Group 7 ( MainChain+H) has 2852 elements
Group 8 ( SideChain) has 2970 elements
Group 9 ( SideChain-H) has 2970 elements
Group 10 ( a_1-5822) has 5822 elements
Group 11 ( System) has 1466 elements
Group 12 ( Protein) has 1466 elements
Group 13 ( Protein-H) has 1466 elements
Group 14 ( C-alpha) has 183 elements
Group 15 ( Backbone) has 549 elements
Group 16 ( MainChain) has 733 elements
Group 17 ( MainChain+Cb) has 902 elements
Group 18 ( MainChain+H) has 733 elements
Group 19 ( SideChain) has 733 elements
Group 20 ( SideChain-H) has 733 elements
Group 21 ( a_1-1466) has 1466 elements
10
21
Fatal error:
Partial overlap between groups ‘a_1-5822’ and ‘a_1-1466’
Is there something wrong?
If there wasn’t, you wouldn’t be getting an error :)
Your groups in this instance were created by selecting atom numbers, and of course (1-5822) is partially overlapping with (1-1466), and these atom numbers do not even make sense. Your first chain is from 10-7082 and the second from 7083-8548. Selecting by residue number should also work fine as there are no redundant residue numbers. You can also easily select by chains B and E.
Are there some clear examples online showing how to do this with chain selection?