The scripts and input files that accompany this demo can be found in the
demos/public directory of the Rosetta weekly releases.
KEYWORDS: ANTIBODIES DOCKING
The entire workflow for this demo should be described in this file. It should describe an entire work flow, with command lines, tested if possible.
The first thing a user should notice is that there’s an antibody modeler in Rosetta 3, but still under development. As of August 2011, Rosetta3 should be used for camelid antibody modeling, but Rosetta2 should be used for other antibody modeling and for antibody docking via SnugDock. The current stable version of Rosetta2 (Rosetta++) is the released Rosetta-2.3.1
Obviously, you need structures of both antibody and antigen in order to do antibody-antigen docking. If you don't have antibody structures, but have antibody sequences, you can use Gray lab antibody homology modeling server (http://antibody.graylab.jhu.edu/) and input the sequence of the light and heavy chain. You will get best 10 structures.
If you want to manually run the scripts yourself, you can download the scripts source code, example, and instructions from the link below: https://svn.rosettacommons.org/source/trunk/antibody/. Again, please realize that the H3 loop modeling is still from Rosetta++.
Download the released version of Rosetta++: https://svn.rosettacommons.org/source/branches/releases/rosetta-2.3.1/
For people outside of the rosetta community: https://www.rosettacommons.org/software/academic/2.3.1/RosettaSnugDock-2.3.1.tgz
tar –zxvf RosettaSnugDock-2.3.1.tgz (if you download the second link) cd rosetta++ scons mode=release –j12 (assuming you can use 12 CPUs)
The SnugDock example in Rosetta++ can be found at: https://svn.rosettacommons.org/source/branches/releases/rosetta-2.3.1/example/
After prepack, you will see
cd ./PrePack_input # you need AB_model*.pdb, ABRM.fab, ABRM.pdb, ABRM.unbound.pdb, Antigen.pdb, pdblist1, pdblist2 cd ../Prepack # you need `paths.txt` and `prepack.bash` file available ./prepack.bash
*.ppkfiles in the PrePack_input directory. Your original pdblist1 and pdblist2 files will be modified as well, please see the README file inside that folder.
cd ../SnugDock # you need EnsembleDock_plus_SnugDock.bash and paths.txt file ./EnsembleDock_plus_SnugDock.bash # please realize that the example we used here is a little slow, due to the protein size
Some extra information you may need, besides the example tutorial linked above:
Make fab file:
The CDR loops of antibody should point to the antigen.
By specifying the antibody loops in the fab file, one can reduce the computational cost for global docking.
The scripts to make fab file can be found at:
Run makefab.pl on your pdb of choice.
./makefab.pl `input pdb` <heavy and/or light chain i.e. HL> ./makefab.pl AB_model1.pdb HL
FR02.pdb complex file:
Use pymol to open both the antibody and antigen in one session and save both into one pdb file.
In the example:
It's better to point the antibody CDRs to the antigen, and keep them at a certain distance.