Hello,
I have been docking two proteins with HADDOCK3 (local installation).
One of my proteins contains a small molecule ligand (which has the same chain identifier with the protein on the pdb file), and this ligand is labeled with HETATM.
After docking, I saw that the the docked pdb files do not have the ligand anymore.
Does HADDOCK 3 automatically remove ligands on proteins?
If so, is there any way to keep the ligand?
Thank you!
HADDOCK3 does not generate parameters and topologies for small ligands. You would have to provide those and define them in the config file.
Check the protein-ligand docking example.
Eventually, get the files from a HADDOCK2.4 server run (download the full archive of the run and look into the toppar directory for the ligand.top and ligand.param files.
Thank you!
I’m not sure how to generate the ligand.top and ligand.param files.
Did you mean I will have to run HADDOCK 2.4 locally to get the files?
Or did you mean to get the files using HADDOCK web server? ( I tried uploading my files to the web server and downloaded “xxxx.haddock_input”, but it does not contain the ligand.top and ligand.param files you mentioned)
Did you mean I will have to run HADDOCK 2.4 locally to get the files?
No - those file are not generated by HADDOCK but by prodrg.
Or did you mean to get the files using HADDOCK web server? ( I tried uploading my files to the web server and downloaded “xxxx.haddock_input”, but it does not contain the ligand.top and ligand.param files you mentioned)
Strange - did you try to submit a run a see if they are present if the result archive?
If they are and missing when downloading the input files then we need to fix it.
And tips regarding ligand topo/param generation:
https://www.bonvinlab.org/software/haddock2.4/faq/#small-ligand-docking-with-haddock
So I was able to find the files in the result archive but not when I downloaded the input files.
PRODGR mentioned in the tutorial seems to be down.
I am wondering what would be the easiest way to get the ligand.top and ligand.param files?
Would it make sense to submit the proteins with ligand onto HADDOCK2.4 and run a short docking there to get the ligand.top and ligand.param files from the result archive and provide them in HADDOCK3.0?
Would these files generated from the results of HADDOCK2.4 server be the correct input files for the ligand on my protein? (I noticed that in the toppar directory, there are a couple other files such as RNA_DNA_amber.top. My input has no RNA or DNA though. So I am confused if the directory is coming from the inputs I provided)
Thank you!
Also, another question I have is:
Suppose I get the ligand.top and ligand.param files, do I still need to keep the ligand information in the pdb file of the protein where the ligand is attached to?
In other words, should I keep the HETATM rows in the protein pdb file?
So I was able to find the files in the result archive but not when I downloaded the input files.
Ok - an issue for us to solve
PRODGR mentioned in the tutorial seems to be down.
Thanks for point it out - there might be a new website for it.
Would it make sense to submit the proteins with ligand onto HADDOCK2.4 and run a short docking there to get the ligand.top and ligand.param files from the result archive and provide them in HADDOCK3.0?
Yes that would work - you only need to extract the ligand.param and ligand.top files
Also, another question I have is:
Suppose I get the ligand.top and ligand.param files, do I still need to keep the ligand information in the pdb file of the protein where the ligand is attached to?
In other words, should I keep the HETATM rows in the protein pdb file?
Yes
We have updated the server to allow the download of the ligand param/topo file in the haddock-input file
Thanks for pointing this out, I tried to look for a new site but could not find anything - we’ll reach out to them and ask if where the domain has moved.
Each of these files describe different things so that CNS (the molecular engine used by Haddock) can interpret the molecule; the topology represents what are the atomic types and their properties (mass, charge), the parameters are a description of the values that should be used for the inter and intra-molecular calculations, the pdb itself described the spatial cordinates of the atoms. So indeed you need all of them.
Yes they should! I’ve updated the server so that once you click the Download input files, you will also get the PRODRG generated topologies and parameters. From there what you need to do is feed these files into your toml configuration for haddock3:
ligand_param_fname = "/path/to/the/data/ligand.par"
ligand_top_fname = "/path/to/the/data/ligand.top"
You can use this docking-protein-ligand-full.toml as a template for your setup.
Good luck!