I am using Rosettascripts (version 3.5) to redesign the protein-protein interface, I use the following applications to redesign the interface,
1. Minmover - for energy minimization
2. PackrotamersMovers - for repacking and redesigning of interface residues
3. set of filters to filter the output
I would like to apply some restrictions on the amino acid selection during the redesing process which are,
1. retain the native ALA amino acids while redesigning and
2. do not substitute ALA aa while redesinging
I used a resfile for this with "* A NOTAA G", this removed all the native ala residues and also didt consider the ala for substituion. But I want to retain the native ala residues at the same time not substituting a amino acid with ala residue. Can someone give me a solution for this?
Thanks in advance for your suggestions.
You want to retain native alanines, and at all other positions allow design without designing to alanine. You need this:
# c NATAA
# c NATAA
# c NATAA ...
NOTAA A in the header will indicate the default state for packing is "design, but not to alanine". Then, in the main body of the resfile before the start token, you need one line per alanine position (where # c in my file should be a residue number # and a chain letter c, so you'll have 25 B, 67 D, etc), where NATAA represents "do not redesign this position" which translates to 'leave it as alanine'. You should use similar NATAA or NATRO (NATRO = do not repack) for other positions you do not want to change.
There is no way I know of to automatically select where the alanines are and not mutate them. If you can't do it by resfile (perhaps if you have inputs of varying sequence), let us know, maybe we can brew something else.
In Rosetta 3.5 with RosettaScripts you can use the OperateOnCertainResidues task operation (https://www.rosettacommons.org/manuals/archive/rosetta3.4_user_guide/Tas...) to do complex residue level selections (this mechanism has actually been altered and expanded upon in recent weekly releases).
In this case, you want to select all the positions which currently have alanines, and then set them to be repack only (or set them to not repack - for alanine the difference is minimal).
Something like this would do it (substitute angle brackets for square brackets - the forum has problems with angle brackets):
Remember that taskoperations are restrictive, so if you combine this task operation with other task operations (like your resfile operation or some other operations) the most restrictive one will win out, so if one operation prohibits design at a certain position, that position will never be designed, even if the other operations allow it.
Thanks for your reply.
I will try to use (OperateOnCertainResidues) keyword. My rosetta 3.5 version was downloaded on 1st week of January, hope this version will work with the operateoncertainresidues keyword.
In a previous calculation I followed smlewis's suggestion, it works well to retain the native Gly resiudes (I am sorry, its not alanine). But still the gly residue is used to substitute the interface residues while redesigning the interface. Here is my resfile for this task,
NOTAA G #I think this will not chose Gly for redesigning
11 A NATAA #List of gly positions to retain the native aa.
27 A NATAA
32 A NATAA
49 A NATAA
67 A NATAA
74 A NATAA
86 B NATAA
93 B NATAA
101 B NATAA
104 B NATAA
118 B NATAA
124 B NATAA
128 B NATAA
129 B NATAA
134 B NATAA
135 B NATAA
155 B NATAA
159 B NATAA
186 B NATAA
Per Rocco's comment, your 3.5 should be fine. If you want a more recent Rosetta, we intend to designate one of the recent weekly releases as 3.6.
This resfile ought to work fine for your purpose; I'm surprised that it doesn't. Could you consider posting a log file and/or your script? (Try posting them as .txt attachments so the forum formatting doesn't eat them.) Maybe you have more than one design step and the resfile is applied unevenly?
Please find the attached xml and resfile files. Let me know if they are ok. and yes you are right, I have different steps and this resfile is called only in few steps. May be its one of the reason. I used a resfile with "* A NOTAA G" keyword which redesigns the way I wanted to. It didt change the native Gly and also didt pick up the gly residue for redesign.
The resfile with "*A NOTAA G" keyword is attached with this post.
Even I have several steps, the redesign step is done with the resfile (attached with previous post), so I guess there should be no problem with this.
It is straightforward to use the keywork "" in the taskoperations to have this contraint. Now the issue is solved. Thanks for your help.
the missing taskoperation of the above post is
(DisallowIfNonnative name=name resnum=0 disallow_aas=G /)