hi there, after i reported a server error in haddock this morning and it was supposedly fixed I have submitted several haddock predictions of homodimers to lingands in the guru mode of haddock2.2, all of them were running … yet i recently checked them and i have another error in all of them:
Status: ERROR
There was an error in the processing of the data. Your have been notified by email about the possible cause.
A solution to your problem might possibly be found in ask.bioexcel.eu
but i havent received any email on behalf of the error, could someone please take a look at this situation ? i would really appreciate it to be able to use this nice tool again 
Best regards
Check your spam folder.
And do try first the 2.4 server - 2.2 will be discontinued as some point and support is limited.
thanks for the support so far,
I realized my problem was because I am using a homodimer, and i read in your previous suggestions that I should “Merge the two chains, give them a unique chainID and renumber the residues to avoid overlap.”
I usually use the rename function in pymol to renumber the atoms, but i am really struggling to find decent information on how to merge my A and B chains and giving them a unique chain ID as you so nicely suggest here … without loosing my B chain in the new saved .pdb file
I tried to use “fuse” to join the sequence break between the 2 chains and the following:
alter (chain B),resi=str(int(resi)+100)
to give chains B an offset of 100
alter (all), chain=‘A’
But i end up not getting my second chain in the pdb
I would be deeply thankful if you could elaborate more on how one could adapt our .pdb files in order to use your haddock tool
Best wishes and keep up with the nice work so far
Check our PDB-tools server or software
Check the forum, this is a frequently asked question
i tryed to use the pdb tools:
$ cat parsed_1.pdb | pdb_selchain.py -B | pdb_reres.py -390
maybe i am doing it wrong in your site …
here’s a short preview od my homodimer showing the chain transition:
(…)
ATOM 2951 CA LEU A 390 25.750 32.875 15.969 1.00 50.31 C
ATOM 2952 C LEU A 390 26.156 32.344 17.344 1.00 50.31 C
ATOM 2953 CB LEU A 390 24.234 32.812 15.812 1.00 50.31 C
ATOM 2954 O LEU A 390 26.281 33.125 18.297 1.00 50.31 O
ATOM 2955 CG LEU A 390 23.641 33.688 14.711 1.00 50.31 C
ATOM 2956 CD1 LEU A 390 22.219 33.250 14.375 1.00 50.31 C
ATOM 2957 CD2 LEU A 390 23.656 35.156 15.133 1.00 50.31 C
TER 2958 LEU A 390
ATOM 2959 N MET B 1 19.234 -25.625 52.031 1.00 20.58 N
ATOM 2960 CA MET B 1 18.266 -24.609 52.406 1.00 20.58 C
ATOM 2961 C MET B 1 18.688 -23.234 51.906 1.00 20.58 C
ATOM 2962 CB MET B 1 18.125 -24.562 53.938 1.00 20.58 C
ATOM 2963 O MET B 1 17.859 -22.328 51.812 1.00 20.58 O
ATOM 2964 CG MET B 1 16.938 -25.359 54.469 1.00 20.58 C
ATOM 2965 SD MET B 1 16.766 -25.250 56.281 1.00 20.58 S
(…)
the total atom number is :
"ATOM 5915 CD2 LEU B 390 -20.562 -39.719 -5.750 1.00 50.66 C
TER 5916 LEU B 390
ENDMDL
END "
and each aa of the homodimer has seq of 390 aa
Could you maybe help a little more here ?
I really like your tools and sorry if i am not interpreting well your explanantions, it would really be of help since i am only re-doing this docking part to update the submission of the paper…
could you perhaps make more precise suggestions ?
I would really apreciate your kind understanding here
You could try something like this (at the command line level, after installing locally pdb-tools):
pdb_selchain -A parsed_1.pdb >chainA.pdb
pdb_selchain -B parsed_1.pdb | pdb_shiftres -500 >chainB.pdb
pdb_merge chainA.pdb chainB.pdb |pdb_chain -A |pdb_tidy > merged.pdb
This will shift the numbering of chain B by 500 and the merged.pdb file will have a single chain with non-overlapping numbering
Thank you so much for all the help o far I will install the pdb.tools and try this nice suggestion.
Once again thank you so much for providing such nice tools
hi there,
i tried the following commands:
$ pdb_selchain -A HSD17B2_calidrispugnax_alpha_dimer_seq.pdb > HSD17B2_calidrispugnax_alpha_dimer_seq_chainA.pdb
$ pdb_selchain -B HSD17B2_calidrispugnax_alpha_dimer_seq.pdb | pdb_shiftres -500 > HSD17B2_calidrispugnax_alpha_dimer_seq_chainB.pdb
$ pdb_merge HSD17B2_calidrispugnax_alpha_dimer_seq_chainA.pdb HSD17B2_calidrispugnax_alpha_dimer_seq_chainB.pdb| pdb_chain -A | pdb_tidy > HSD17B2_calidrispugnax_alpha_dimer_seq_merged.pdb
and the structures i get in every step are appended here, i think it work, but visually in pymol the _merged.pdb look “strange” by showing only one chain even though 2 sequences are showed there…
Is this now correct for me to perform an HADDOCK analysis with a ligand ?
And if yes, will the output of haddock to the ligand be represented in this “unique” showing chain? or is it possible to then have the original dimer shape ?
Sorry and thank you so much I really appreciate your effort
…
HSD17B2_calidrispugnax_alpha_dimer_seq.pdb (468.2 KB)
HSD17B2_calidrispugnax_alpha_dimer_seq_chainA.pdb (234.2 KB)
HSD17B2_calidrispugnax_alpha_dimer_seq_chainB.pdb (234.2 KB)
HSD17B2_calidrispugnax_alpha_dimer_seq_merged.pdb (468.4 KB)
Seems like a feature of pdb_merge to keep the MODEL statements if present. Which is not what you want here.
Try instead then the following commands:
$ pdb_selchain -A HSD17B2_calidrispugnax_alpha_dimer_seq.pdb |pdb_keepcoord > HSD17B2_calidrispugnax_alpha_dimer_seq_chainA.pdb
$ pdb_selchain -B HSD17B2_calidrispugnax_alpha_dimer_seq.pdb |pdb_keepcoord | pdb_shiftres -500 > HSD17B2_calidrispugnax_alpha_dimer_seq_chainB.pdb
$ pdb_merge HSD17B2_calidrispugnax_alpha_dimer_seq_chainA.pdb HSD17B2_calidrispugnax_alpha_dimer_seq_chainB.pdb| pdb_chain -A | pdb_tidy > HSD17B2_calidrispugnax_alpha_dimer_seq_merged.pdb
Note the extra pdb_keepcoord command (this will extract only the coordinates lines)
it seems that it still overlaps both the monomers of the alphafold predicted homodimer with those commands, these were the results:
HSD17B2_calidrispugnax_alpha_dimer_seq.pdb (468.2 KB)
HSD17B2_calidrispugnax_alpha_dimer_seq_chainB.pdb (234.2 KB)
HSD17B2_calidrispugnax_alpha_dimer_seq_merged.pdb (468.4 KB)
HSD17B2_calidrispugnax_alpha_dimer_seq_chainA.pdb (234.2 KB)
I keep seing the “two” sequences in pymol but the cartoon seems to really overlap in shape, both of the structures.
if there is something else that you could suggest i will be really gratefull to try it
But thanks in advance for all the help so far
Indeed - the problem are the MODEL/ENDMDL statements - if you delete those from chainA and B then the pdb_merge command will work fine and you will see both chains
THANK YOU SO MUCH , IT WORKED ! now i have the dimer!
I did the following:
$ pdb_selchain -A HSD17B2_calidrispugnax_alpha_dimer_seq.pdb |pdb_keepcoord > HSD17B2_calidrispugnax_alpha_dimer_seq_chainA.pdb
$ pdb_selchain -B HSD17B2_calidrispugnax_alpha_dimer_seq.pdb |pdb_keepcoord | pdb_shiftres -500 > HSD17B2_calidrispugnax_alpha_dimer_seq_chainB.pdb
#then i opened the merged.pdb file with notepad++ and manually deleted the lines that contained “MODEL” and “ENDMDL”
#then i merged the 2 manually edited files
$ pdb_merge HSD17B2_calidrispugnax_alpha_dimer_seq_chainA.pdb HSD17B2_calidrispugnax_alpha_dimer_seq_chainB.pdb| pdb_chain -A | pdb_tidy > HSD17B2_calidrispugnax_alpha_dimer_seq_merged.pdb
now i have the nice homodimer shape in pymol
Thank you so much for the support i really appreciate it
best regards
Another solution is to use pdb_splitchains first and the process the split PDB files. Those do no longer contain the MODEL statements.
thank you so much for all the nice suggestions so far, I am really happy it worked and that i can proceed to all of my haddock predictions