How do I build a glycosylated protein from PDB files manually?

Question

I have a large protein PDB file (~3500 residues) and a couple of PDB files containing glycan structures I sourced from GLYCAM.

How can I attach the N and O-glycans onto each protein residue with software? In PyMol, I am stuck at loading up the protein PDB file and glycan file together, but I don't know how to put the two elements together. Are there any commands I can use? or tutorial resources I can take a look at? I'm open to trying other platforms.

I've tried glycan modelling webtools (GLYCAM, CHARMM-GUI) but they can't seem to process my large glycoprotein - leading to errors or extended waiting times which also end in failure.

Answer 1

You can use ChimeraX.

1. Open your glycans and your protein pdb files as different models, then move the models of your glycans until they are near the residues you want (in the Toolbar: Right Mouse -> Move Model).
2. Then use combine to combine one of your glycan model with the protein
3. Finally Use Tools -> Structure Editing -> Build Structure and its Adjust Bonds menu to add covalent bonds. You can also add or delete atoms this way.

This is a straightforward but complicated and tedious way of doing this, and it will give you unideal structures. There should be an equivalent way to proceed using PyMol. For a more automatized way I can only recommend having a look at PyRosetta. There is a Jupyter Notebook tutorial about adding glycosylation.

There may be other tools out there for doing this, like MoleculeKit, but I'm less aware of them. A nasty solution would be to cut your protein in smaller pieces, glycosylate each part with CHARMM-GUI, then reassemble them.

Answer 2

answer from @coot, converted from comments (please edit this answer to improve it if you can):

Manual model building is a common thing in structural biology. Usually, Coot is used.