I would like to know how to automatically position new peptides into the binding site of a enzime, in order to run peptide-protein docking approaches?
In my case, I know the binding site and I have the crystallographic structure of a reference peptide located at the binding site, but I need to generate several (thousands) of other new linear peptide structures at this same binding site, in order to run the docking calculations using the rosetta3 protocols.
Any tips on how to make automatic the generation of the linear peptide (from sequence of residues), and the positioning at the region of the binding site (maybe alignment with the crystallographic structure of reference peptide?) ?
Thanks in advance for any help.
It seems like you would like to study the binding affinities of different peptides at the active site. For that purpose the crystallographic coordinates of the known binder might be a good starting point for further minimization or refinement. To get starting structures, you can thread the peptide sequences on the existing peptide backbone using Rosetta Fixbb design protocol. You need to create resfiles for that. After threading is done, you can use FlexPepDock. Use constraints to retain important interactions present at the binding site. If FlexPepDock gives low score for a bound peptide with constraints satisfied, it is probably a good binder.
Thank you very much for your fast answer.
You are right. I'm trying to study binding affinities of several (thousands) different peptides at the active site of protein targets. But instead of work with "random" generated peptides, I already know thousands of sequences (primary structure) of peptides that I'm interested to calculate.
I am a beginner with Rosetta and I would appreciate if you could kindly give me any more help with this issue. I have already ran the demo tests with Fixbb.
The tip of Rosetta Fixbb seems very good, but It is possible, using the "resfiles", to setup every one of the thousands of different peptides I would like to calculate?
Did you have any additional material (tutorial, reference, etc.) about FixBB or "resfile" generation?
My e-mail contact is firstname.lastname@example.org
I did mean to thread the specific peptide sequences you want on the exisitng backbone conformation. It should be pretty easy to do.
First step: create specific resfile for each peptide.
795 B PIKAA L EX 1 EX 2 EX 3 EX 4
796 B PIKAA I EX 1 EX 2 EX 3 EX 4
797 B PIKAA R EX 1 EX 2 EX 3 EX 4
798 B PIKAA S EX 1 EX 2 EX 3 EX 4
You instruct to keep native rotamer by writing NATRO in the first line and instructions for specific positions are overwritten by instructions provided after the "start" section; first col = residue postion; 2nd col = chain id; 3rd col = instruction to pick amino acid mentioned in fourth col and rest of the cols give instructions to use extra rotamer for different chi angles.
After you have the resfile ready you can run fixbb design as:
$ROSETTA_BIN/fixbb.linuxgccrelease -database $ROSETTA_DB -s template.pdb -resfile resfileX -ndruns 100 -nstruct 1 >design.log
The template pdb should the solve crystal structure.
The deisged pdb can be used for running FlexPepDock; either minimization using -flexPepDockingMinimizeOnly or refinement using -pep_refine flag.
You can create an short script to automate the creation of resfiles for different peptides and running fixbb design and FlexPepDock there after.
Give it a try! I will be happy to help if you need further help.
Thanks for your detailed explanation.
I will try as you suggested.
If I have any dificulties, I will write again to you.
Thank you again.