Dear Modeller Mailing List,
I was hoping to ask if there is a way to automate glycan-building in Modeller. I’ve been able to get all residues/atoms built using AutoModel. However, I lose the glycan information when using AutoModel. Is there another program that interfaces well with Modeller, or is there a class in Modeller that can add glycans at known positions (e.g. if I know NAG is attached to residue 6, can I add that to the PDB output somehow?).
Thank you! Steven Truong sdt45@cam.ac.ukmailto:sdt45@cam.ac.uk Cambridge University
On 3/19/21 9:33 AM, Steven Truong wrote: > I was hoping to ask if there is a way to automate glycan-building in > Modeller. I’ve been able to get all residues/atoms built using > AutoModel. However, I lose the glycan information when using AutoModel. > Is there another program that interfaces well with Modeller, or is > there a class in Modeller that can add glycans at known positions (e.g. > if I know NAG is attached to residue 6, can I add that to the PDB output > somehow?).
To do this you would need to 1. Add residues for each sugar. 2. Add bonds between your glycoslyated residues and sugar residues, and perhaps also between sugars.
You would need a fairly good understanding of CHARMM topology files, including patch residues. See for example https://salilab.org/modeller/FAQ.html#8 and https://salilab.org/modeller/FAQ.html#9
There is no recipe out of the box to do this in Modeller, but AllosMod can do it. See https://modbase.compbio.ucsf.edu/allosmod/help.cgi?type=glyc
You are certainly welcome to use the AllosMod CHARMM topology file and the Python code it uses to generate AutoModel inputs. See https://github.com/salilab/allosmod-lib/blob/main/data/top_all_glyco.lib and https://github.com/salilab/allosmod-lib/blob/main/lib/allosmod/get_pm_glyc.p... respectively.
Ben Webb, Modeller Caretaker