Hi Ben,

Yes, you are correct that rebuilding a specific segment is only available for rebuild_in_place=true.

For replace_side_chains, you have a choice of specifying use_any_side=True (default) or False. If you say "True", then if it cannot figure out what is the right sequence it makes one up that matches the density.

You don't have to specify which map coefficients to use...it will guess.

Best to supply a PDB just of the part you are interested in.  Doesn't matter what side chains you supply (they are ignored.)

Replace side chains operates on whatever you supply, so no you can't say which chain and residues to replace. (Nice idea though!).

All the best,
Tom T



On Thu, Jan 18, 2018 at 1:44 PM, ben flath <bef206@mail.usask.ca> wrote:

Hi

I have run in to a few issues while trying to run phenix.replace_side_chains : the sequence of the output pdb does not seem to correspond to that of the input sequence file. Inspecting the .eff file it appears that the program is in fact reading my sequence file. Also, do I need to explicitly specify which map coefficients to use? Is it best to supply a pdb of my entire model or just that of the “extra” part I am interested in and as a poly-ala or with my best guess at side chains? Can I specify which chain and or residues I would like to replace_side_chains for?

 

I Also have some question about using autobuild. Trying to use gives an error that my model sequence and input sequence are not the same which is a requirement for rebuild in place. It seems for the GUI at least the rebuild_list options (rebuild_chain_list=A rebuild_res_start_list=27 rebuild_res_end_list=35) are only available for rebuild_in_place=True.

 

Thanks in advance

 

Ben Flath

 

From: Tom Terwilliger [mailto:tterwilliger@newmexicoconsortium.org]
Sent: January 10, 2018 6:20 PM
To: bef206@mail.usask.ca
Cc: phenixbb@phenix-online.org; Thomas Charles Terwilliger
Subject: [phenixbb] Autobuild in to "extra density"

 

Hi Ben,

 

I think you've done the hard part already by building a poly-ala model into your extra density!

 

If you want to try and sequence it using the density, you can use phenix.replace_side_chains with the keyword "sequence_from_density=True" to do this.  See: https://www.phenix-online.org/documentation/reference/replace_side_chains.html

 

To use autobuild with this, add an extra line to your sequence file with the sequence of the extra piece (separate it from the previous sequence with a blank line or a ">".)

 

I would try rebuild_in_place=True to try and optimize the position of your extra chain.  You can specify what residues to work on with something like:

 

  rebuild_chain_list=A

 rebuild_res_start_list=27

 rebuild_res_end_list=35

 

This will rebuild chain A27 through A35.

 

For density modification you might want to use a mask based on your model.  For this you would say:

 

 mask_from_pdb=my_existing_pdb_file.pdb

 

Let me know if that doesn't do it!

All the best,

Tom T

 

 

Date: Tue, 9 Jan 2018 20:50:41 +0000
From: "Flath, Ben" <bef206@mail.usask.ca>
To: "phenixbb@phenix-online.org" <phenixbb@phenix-online.org>
Subject: [phenixbb] Autobuild in to "extra density"
Message-ID: <52cba352941e48869fc9d5223a97ccaf@Mail06.usask.ca>
Content-Type: text/plain; charset="us-ascii"

Hi All
I have a structure I am building/refining at 2.6 A resolution, there are three NSC molecules in P321 unit cell. Each monomer contains 150 aa and are at least 95% built. Here is where I need some help. There is extra density. Gel and Mass Spec data suggests that there is a smaller MW weight protein/band contained in the crystals which is most likely a fragment of my target protein. Basically I have 3 NCS copies plus a half a molecule that does not appear to have the same tertiary structure (LSQ & SSM do not align well) as the other 3 molecules. I have been able to build several poly-A helices in to this density and R/Rfree are now 0.25/0.31.

My Question is, how do I set up a phenix run (presumably autobuild) So that I can begin to dock my sequence in to these chains and have them extended? Does my sequence input file need to include more than one sequence string? Asking autobuild to find and build 4 NCS groups does not seem to build anything in to this extra density (Also the sigma level for the extra density chains seems to be somewhat less than the well-defined part) I am guessing this is a rebuild_in_place=false job? Can I specify only a certain chain only to be rebuilt? Also are there some special considerations with regard to density modification I should be aware of?

Any help would be much appreciated.

Regards


Benjamin Flath, M.Sc.
Research Assistant, Cygler Lab
University of Saskatchewan
Department of Biochemistry
Room 3D40 HSCI
Box 27, 107 Wiggins Road
Saskatoon, SK
S7N 5E5
Phone: 306 220-7355
email: benjamin.f@usask.ca<mailto:benjamin.f@usask.ca>

 

--

Thomas C Terwilliger

Laboratory Fellow, Los Alamos National Laboratory

Senior Scientist, New Mexico Consortium

 




--
Thomas C Terwilliger
Laboratory Fellow, Los Alamos National Laboratory
Senior Scientist, New Mexico Consortium
100 Entrada Dr, Los Alamos, NM 87544
Tel: 505-431-0033