You are here

Ab-initio peptide-protein docking

10 posts / 0 new
Last post
Ab-initio peptide-protein docking

I am trying to perform ab-initio peptide (6-15 in length) - protein docking as described at:
Confused by a number of issues and would be grateful for any guidance please if anyone has performed
the proceders:

Fragment generation - The Robetta server requires a minimum aa length of 26.
So how does one generate 3,5 and 9 length fragments for sub-26 length peptides ?
is it only possible as described in:


Post Situation: 
Sun, 2013-06-23 16:20

Hi Suhail,

To create the fragment for smaller peptides you need to use the scripts which lives in rosetta/protocol_capture/FlexPepDock_AbInitio/.
The Robetta Server requires a minimum AA length of 26 as the various tools used during fragment generation like secondary structure prediction are not optimized for very small peptides. Though it is seen that the fragment generated using the same protocol shows good coverage of peptide conformational space and thus can be used.
Hope this helps. If you need a detailed tutorial on FlexPepDock please provide me your email id.


Mon, 2013-06-24 01:06

Nawsad Hi

Thanks for comment. I went through the README file in FlexPepDock_AbInitio but still struggling to make it work.
For example (sorry for boring list):
- In link on
site is dead
- where does sam.predict-2nd come from
- picker.linuxgccrelease does not exist in 3.4 (unless its fragment_picker.linuxgccrelease)
- JUFO only seems to be for debian (using RedHat)
- cannot find nnmake_database
- where does chemshift come from ...
If I have failed to read something basic please do let me know.

So yes please - if you have a working example do let me know

Mon, 2013-06-24 11:57

Hi Suhail !
Here it nicely explained (rosetta/rosetta_tools/fragment_tools/fragment.README ). Follow the instruction to set up the environment.
I am also mailing you flexpepdock tutorial which might be useful later on.
Best of luck.

Mon, 2013-06-24 14:40

Hi nawsad! I'm having the same question that how to generate fragment library for peptide(9 length).Could you provide me a detailed tutorial on FlexPepDock ab-initio( Thanks a lot!

Tue, 2013-09-03 19:11

Hi All,
The FlexpepDock tutorial is available here .


Thu, 2013-09-05 01:35

It should be pointed out that that's a link to the internal RosettaCommons wiki, and won't be accessible to people who don't have developer access.

You don't really need access to that page for FlexpepDock, as that section basically just links to which, from what I can tell, is a publicly available Google Doc.

Thu, 2013-09-05 11:15

Thanks!I know how to generate a fragment file for peptide(6-15 length)now.

Mon, 2013-09-09 17:51


I was trying abinitio protein-peptide docking, for a short peptide of length of 6 amino acids. I was trying to use, but I want to inactivate the parallel jobs option. After making the required changes for the number of CPUs for Fragment_Picker and BLAST_NUM_CPUS, I tried to inactivate the "EXTRA OPTIONAL FEATURES", but the same is also required for secondary stucture prediction. If I inactivate the $SLAVE_LAUNCHER  and  $SLAVE_LAUNCHER_MAX_JOBS  line, it is still not working as the rest part of the script uses it.

I would only like to run the job on my headnode. Can you suggest me on this.



Tue, 2016-02-09 11:38
Payal Chatterjee

As I read things, by default it should test if the SLAVE_LAUNCHER command exists, and if it doesn't will run things serially and locally -- this should actually be the default case, as the bin/ command doesn't exist in the Rosetta release: it's a command that you would have to provide yourself, if you want to take advantage of that feature.

So leave the SLAVE_LAUNCHER line alone (if you want to change the SLAVE_LAUNCHER_MAX_JOBS line to 1 as a safety, that should be fine.) That should work for you. If it doesn't, what is your error message? It might not actually be related to the SLAVE_LAUNCHER settings.

Fri, 2016-04-29 08:57