Preserving peptide secondary structure during cyclization/docking

HADDOCK
, ,
1

Hello everyone,

I am wondering if there is a way to preserve the secondary structures of the input peptide during the cyclization and subsequent docking in Haddock3. Of course it needs to undergo 3D changes for the docking but I need to freeze/restraint the backbone changes in a region.

I am aware of ssdihed parameter in the flexref module but it’s the cyclization itself where the problem starts.

Any suggestion/help is highly appreciated.

Regards,

Jameel

2

If the termini where the cyclisation happens are close in space you might not need to go through the entire cyclisation workflow.

In flexref and emref there is the ssdihe parameter that allows to restrain the backbone dihedrals.

ssdihed = ‘none’ # $minchars 0 / $maxchars 100 / $choices [‘none’, ‘all’, ‘alpha’, ‘alphabeta’] / $title Define automatically backbone dihedral angle restraints / $short If turned on, this option will automatically defined backbone dihedral angle restraints for the selected regions. / $group dihedral restraints

3

Thanks a lot @amjjbonvin!

I still have 2 questions regarding ssdihe.

  1. Was the ssdihe = "alphabeta" parameter implemented to restraint the alpha and beta secondary structures in both molecules (the peptide and the receptor protein)?
    I looked into its section in the documentation, but couldn’t find an answer to this question.

  2. Is it expected to see the secondary structures preserved in all or the majority of the output models (the docking result models)?

Thank you in advance and kind regards!

Jameel

4
  1. Was the ssdihe = "alphabeta" parameter implemented to restraint the alpha and beta secondary structures in both molecules (the peptide and the receptor protein)?
    I looked into its section in the documentation, but couldn’t find an answer to this question.

Yes it applies to all molecules.
The restraints are defined based on the backbone dihedral angles measured in the input models for that given stage of the workflow.

For alpha helices the phi/psi angles must be within the range -37 to -77 and -26 to -67
For beta sheets phi/psi angles must be within the range -105 to -45 and 105 to 145

The measured angle is then used, with additional upper/lower limit errors defined by the error_dih parameter (default is 10 degrees).
So we don’t impose ideal values, but use the values measured in the models.

Note that you could also provide your own dihedral restraints file to HADDOCK (defined by the dihe_fname parameter).

You can find all related parameters with the following command:

haddock3-cfg -m flexref |grep dih

  1. Is it expected to see the secondary structures preserved in all or the majority of the output models (the docking result models)?

Some deviations are indeed expected as these are restraints and not constraints and we allow for error bounds.
Depending on the limits used to define a secondary structure element things will vary.