[gmx-users] Re: Difficulty building a topology for a synthetic, branched PEG-peptide molecule [SOLVED]

[Pablo Englebienne]

> That's a long bond. Does your reference length in specbond.dat suit it?
> IIRC there should be some evidence in the output of the special bond
> being formed if it actually is. If not, your symptoms are probably related
Hi Mark, indeed, I think that was part of the problem. pdb2gmx indeed 
outputs a message when a specbond.dat rule is matched and a bond formed.

In case it helps to someone else or for reference purposes, I finally 
managed to solve the issue.

Some of the lessons I learned:
- make sure that the residues/atoms in specbond.dat were correct (I had 
defined a number of residues for termini and mid-chain PEG and 
connectors, and I got them confused at some point, so not all of them 
were being recognized properly). The "dangling bond at at least one of 
the terminal ends" given by pdb2gmx is most likely due to this and/or 
the protonation state of the residues connecting the fragments
- each fragment is assigned a different chain letter in the source PDB 
file; in my case, for a system like

[N-(peptide1)-C]-[N-(PEG)-C]-[N-(peptide2)-Lys-C]
                                             |
                                             NZ
                                             |
[C-(PEG)-N]-[C-(peptide3)-N]

each fragment in square brackets is assigned a different chain name in 
[A-E] in the PDB file:

ATOM    123  N   ARG A   1      74.024  13.299  50.237  1.00  
0.00           N1+
                      ^
- always list the atoms within a chain in an N-to-C direction. This 
means that the main branch is defined first, and then the branching 
point is defined in an N-to-C direction, even if it is counterintuitive 
by the way they are connected. specbond.dat takes good care of setting 
up the connection in all cases (as long as they are well defined...).
- the PEG residues need to be defined as type "Other" in residues.dat
- call pdb2gmx to manually assign the termini and Lys protonation states 
manually, and to merge the chains into a single molecule:

pdb2gmx-ter  -f  substrate.pdb-chainsep  interactive-lys

-> the "internal" termini (i.e., the ones that are peptide termini but 
are connected to a PEG chain) need to be given a protonation state of 
"None", while the "real" termini can be assigned as charged or neutral, 
depending on the conditions

> Oh, and well done for constructing a good question. You would likely not have gotten anywhere giving less detail :)
Thanks, it actually helped putting everything in writing, as it pointed 
out the few things that I hadn't yet looked at in detail...

Pablo Englebienne, PhD
Dept. of Biomedical Engineering
Dept. of Chemistry and Chemical Engineering
Institute for Complex Molecular Systems (ICMS)
Eindhoven University of Technology, TU/e
PO Box 513, HG -1.26
5600 MB Eindhoven, The Netherlands
Tel +31 40 247 5349