Hi Oliver and Xiaocong,

I understand your point, and it does seem to be a right choice to handle my systems. I am basically dealing with UDP-GlcNAc and some glycans attached to the Ribitol phosphates. Thus, parameters would be crucial in this case. I don't think there is any easy workaround, instead parameterizing them (which also needs to be validated first).

Many thanks,

On Thu, Dec 15, 2016 at 10:00 PM, Xiaocong Wang <[log in to unmask]> wrote:

I agree with Oliver.  We do this when the connection part is not important for the essential properties of the molecule that you are modeling.  You can also check out the supplemental material of the following paper:

I used parm for amino acid, but GLYCAM for sulfate.

Good luck!


Xiaocong Wang
Complex Carbohydrate Research Center
The University of Georgia
315 Riverbend Road,
Athens, GA, 30602
Tel: (706) 254-7958

From: Users of GLYCAM & GLYCAM-Web <[log in to unmask]> on behalf of Oliver Grant <[log in to unmask]>
Sent: Wednesday, December 14, 2016 10:26:29 AM
To: [log in to unmask]
Subject: Re: About using GLYCAM and GAFF for a molecule
Hi Sushil,

You can do this, but I wouldn't trust it for something that needs to be high accuracy and/or can't be validated. I did it for modeling linkers attached to glycans here https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3854501/, where it didn't matter about getting populations of rotamers correct, or if some of the shapes produced weren't populated in reality. 

See attached frcmod file for what we did. There is a note with each parameter explaining where I took it from. Sp8, Sp10 etc refer to linkers (spacers) on the consortium for functional glycomics (CFG) glycan arrays. Check Table 1 in the publication for the molecular structures of the linkers.


On Sat, Dec 10, 2016 at 9:04 AM, Sushil Mishra <[log in to unmask]> wrote:

I do not have particular specific case but rather a general question about using GLYCAM and GAFF together. I am frequently seeing some ligands, which have a glyacn attached to it (e.g. UDP-GlcNAc, dTDP-Rhanmonse, heparin sulphate etc). I wonder if there is an easy workaround to use GLYCAM for glycan part of the ligand and GAFF for rest of the molecule. 

What I was thinking as of now is generate GAFF parameters for the whole molecule and later renaming glycan part (atom names and residue name) to GLYCAM. Missing parameters can be be generated in frcmod file (using parmchk ) ? What I may be missing is parameters for the "connecting bond, angle, dihedral etc". 

Have someone tried such a case and can suggest me something ?