You are here

Extract PDB structures from database

4 posts / 0 new
Last post
Extract PDB structures from database
#1

I am using a RosettaScripts which outputs structure coordinates to a database (.db3). I am trying to extract the individual PDB files. I tried using the following command:

score_jd2.static.linuxgccrelease -inout:database_filename struct.db3
 
 
ERROR: Option matching -inout:database_filename not found in command line top-level context
caught exception ERROR: Option matching -inout:database_filename not found in command line top-level context
 
Any idea of what I can do to fix this?
 
In case you need more info, the RosettaScript XML I am using is in the supplementary materials to this paper: https://www.biorxiv.org/content/early/2017/11/17/221689
Category: 
Post Situation: 
Mon, 2018-01-29 09:30
cossio

It looks like there's a typo on that page. The correct option is `-inout:dbms:database_name` -- the default is a SQLite database, which is a file, but if you do the appropriate extras compile, you can also have MySQL and PostgreSQL support.  Those two database types also use the -inout:dbms:database_name option, but it's not a filename.

Also note that the -inout:dbms:database_name option only sets the database file to be used - it can be used both for input and output (or both). To select whether you are doing structure input or output to the database, use the `-in:use_database` or `-out:use_database` (or both) as appropriate.

Tue, 2018-01-30 07:35
rmoretti

This worked fine some months ago. I am now getting strange errors. Could there have been a change in the newer releases of Rosetta affecting this? I am now using the numbered version 3.10. Here is the error output.

 

Error: [ ERROR ] 
Error: [ ERROR ] Treating failure as bad input; canceling similar jobs
protocols.jd2.FileSystemJobDistributor: job failed, reporting bad input; other jobs of same input will be canceled: 4_0001
protocols.features.ProteinResidueConformationFeatures: Attempting to set atom coordinates for residue '121'.
The pose residue has '18' coordinates but '17' were provided.
The residue type in the database is 'PRO:NtermProteinFull' and it matches the pose residue type.

Error: [ ERROR ] ERROR: Exception caught by JobDistributor while trying to get pose from job '40_0001'
Error: [ ERROR ] 

File: src/protocols/features/ProteinResidueConformationFeatures.cc:521
Atom mismatch in ProteinResidueConformation
 

Wed, 2018-10-17 15:59
cossio

There's been a change in how the PRO:NtermProteinFull is represented in Rosetta under the "REF2015" scoring environment versus the previous Talaris environment.

Specifically, there's an extra "virtual" atom that's been added to better represent the geometry. Normally this doesn't make a difference, as virtual atoms aren't output to PDBs, and are automatically re-added on PDB readin, but with database representation (and silent files), those virtual atoms are explicitly represented, and things can error out.

Generally speaking, the silent and database file formats aren't necessarily 100% compatible between different versions of Rosetta. We try to keep them compatible as much as possible between versions (or at least preserving the ability to read an older version with a newer Rosetta) but there's inevitably changes which come in which will result in the inability to read in older data.

If you don't have the version of Rosetta which created the database to extract with, one possibility is to add the `-restore_talaris_behavior` option to the commandline. The Talaris environment still uses the older version of the PRO:NtermProteinFull, so it might be able to read in the older version of the database.

Fri, 2018-11-23 11:23
rmoretti