Is there a way to refine glycan chains and maintain glycosidic bonds in 2.4? Oligos are disassociating and becoming monomers in docking schemes between a circular glycan chain and small molecule ligand despite them having linkages in the original pdbs.
Dear user,
In HADDOCK, we do not take into account the linkages provided by the CONNECT statement in the input PDB file.
Instead, we identify covalent bound based on:
- distance
- topology in the force field
Therefore, if that particular linkage topology for the glycan you are interested in is not defined in our force-field, each monomer will be treated separately.
Please, give us the name of your monomeric glycans and the desired linkage, and we can think of implementing it in HADDOCK.
Best regards,
Victor
Hello,
The glycan is canonical B-cyclodextrin with a 1 - 4 linkage well defined in space. Example line of pdb is shown. The atoms are organized by residue but it is possible that I have an inconsistency in the forcefield atom naming convention. Using default settings for glycans in docking calculations.
HETATM 1 C1 GLC A 1 1.549 25.826 2.682 1.00 4.58 C
Thanks for the help,
The list of supported glycans can be found at: HADDOCK Web Server - Settings
GLC would correspond to alpha-D-glucopyranose
Where does your B-cyclodextrin model come from?
Hello,
The pdb is based on CSD Entry: WEWTOJ
Regards
Hi there
I took a cyclodextrin from the PDB database (PDB entry 5MK9) and run it through the refinement interface of the server.
No issues:
https://wenmr.science.uu.nl/haddock2.4/result/1111111111/774906-cyclodextrin
It must be your model that has some non-standard namings or something like that.
Hello!
Thank you for so much for taking the time and your attention to look into my issue. I compared the atom naming from the file used in your run and did indeed find a nomenclature error at the linkage. Hopefully, changes to these will resolve the problem. Thank you for supporting such a great resource.
Regards,
Scott McCallum