Hi there,
I am kinda new to HADDOCK.
I have a specific molecule prepared in martini format (not all-atom). I want to know if I can input it “as is” to HADDOCK2.4 and get the result in martini format?
My goal it is to create an artificial carbohydrate in CG and dock a CG protein on it. I will not need any all-atom feature for my specific study.
Best,
Mohammad
The 2.4 server does allow to coarse grain proteins and nuclear acids (done on the server side) but no support for small molecules.
The final stage will always be to convert the model to all atoms, but you can find the Martini files in the tar archive of the run.
@amjjbonvin: So CG ligands are not supported on the server, but if I run the code locally, is there a hack I could use? maybe by adding the ligand to toppar? The point is that I only want the it0 part of the code, i.e., rigid-body docking and I don’t need the fine tunings.
In addition, I wouldn’t mind if the code crashes when converting the CG to AA, as I don’t need the AA part.
As an additional point, can I use ions in the CG version?
@amjjbonvin: So CG ligands are not supported on the server, but if I run the code locally, is there a hack I could use? maybe by adding the ligand to toppar? The point is that I only want the it0 part of the code, i.e., rigid-body docking and I don’t need the fine tunings.
That’s a trick thing to do… You would have to figure it out yourself, adding the parameters/topologies to the Martini ones.
But not something we support I am afraid.
If your system is not that large, why not go atomistic and then generate your Martini model?
@amjjbonvin, thanks for the reply 
The system I am aiming for is a relatively large system, ~50 proteins each with 200 to 1000 residues in the final structure. It also requires custom ligands with over 1000 atoms.
I guess I was able to make a smaller and hacky version to work (I would still need to check and see)
Another point is if it is possible to use another file format instead of PDB?
PDB is limited to “small” systems and the limit of residues and atoms in this format prevents really large systems.
I am afraid this won’t work for HADDOCK. Our limit on the number of molecules is 20, and even that really only makes senses provided you have good experimental data to guide the modelling…
Hmm, is 20 the limit of simultaneous docking or in total?
Should I be fine if I could do them one by one?
It is the limit of the number of separate molecules in one docking run
If you are speaking of docking 50
Different proteins one by one then there is no problem, even for 500 to 1000 residues
Awesome. thanks