Rosetta 3.5
 All Classes Namespaces Files Functions Variables Typedefs Enumerations Enumerator Friends Macros Pages
Documentation for the Rosetta Flag File Builder GUI
Jared Adolf-Bryfogle



This document was last updated December 2012, by Jared Adolf-Bryfogle The PI for this application is Roland Dunbrack

The application was developed by:

Code and Demo

The code is available in rosetta_source/GUIs/RosettaFlagBuilder To run the GUI use ./ in the code directory.

The DoxygenParser class is available in rosetta_source/GUIs/pyrosetta_toolkit/modules


Refs in Publication

New_Library contains code for this purpose

The purpose of this GUI is to help running, exploring, and building flag files for Rosetta applications. It is useful for exploring both documentation and available options and default settings of the many Rosetta applications.


This GUI parses Rosetta Doxygen documentation and the -help flag for each application.


This application attempts to help in formatting a Rosetta flag file and exploring documentation. However, numerous applications have special formatting for Doxygen documention, including embeded HTML. Efforts are under way to standardize and better parse the formatting, but please refer to rosettacommons for full production runs.


Main Window

Path Builder

Qsub Cluster Setup GUI

Input Files

If there are any special input file types, describe them here.


To get all available options given to a particular app, choose Repopulate->x This overwhelms the Tk GUI, but can be useful if you want to go off the beaten path. For an individual app that is selected, you can click x. QsubClusterSetup app works on all apps. However, running Rosetta in MPI mode is recommended over using batch processing using Qsub. If you can get MPI to work on your cluster, use it instead. QsubClusterSetup is made to be applicable to all apps and all clusters. If you find it does not work for yours, please report it on the bugtracker.

Describe how to get the best use out of the protocol. Does it not require constraints, but work better with them? Do you need to remember -ex1 -ex2 to get much out of it?

Expected Outputs

What does your protocol produce? Usually it will be just a PDB and a scorefile, but note if there should be more. Note what the normal exit status of the protocol is (for example, "You'll see the jd2 x jobs completed in y seconds message if successfully completed".)

Post Processing

What post processing steps are typical? Are score vs RMSD plots useful? Are structures clustered (if so, give a command line)? Is it obvious when either the application has succeeded or if it has failed (e.g. if the protocol makes predictions like "This is the docked conformation of proteins A and B"). In the case of designs, how should designs be selected?

New things since last release

If you've made improvements, note them here.