Merge pull request #4844 from RosettaCommons/vmullig/update_vscode_git_settings
Minor tweaks to versioning of .vscode configuration files
We don't really need anything in the .vscode directory to be under version control except for `.vscode/c_cpp_properties.json`. Spurious warnings about changed files pop up whenever new file extensions are opened with VSCode, since the `.vscode/settings.json` file gets auto-extended whenever this happens. This pull request:
- Removes redundant lines in the .gitignore file (but keeps the .vscode directory listed, removing only a redundant listing).
- Removes `.vscode/settings.json` from version control (since this file seems to auto-configure itself well).
- Adds `.vscode/settings.json.template`, just in case we do need to restore the settings file for any reason.
- Keeps `.vscode/c_cpp_properties.json` under version control (since this contains Rosetta-specific setup, specifying that we use the cxx11 standard, that the external directories should be included, and that PTR_STD should be defined).
notify author
notify list [rosetta-logs@googlegroups.com]
Add residue level control to the AlignPDBInfoToSequences mover (#4826)
This pr enables the ability to set pdb info residue numbering that is more complicated than simply starting at 1
Included are unit tests and integration tests that utilize this new functionality.
% source/src/core/sequence/Aligner.cc -> Alter pointer error to throw instead of hard exiting.
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #4798 from RosettaCommons/roccomoretti/index_selector_reverse_message
Add warning message regarding the Index selector's reverse option.
Potentially counter-intuitively, the reverse option of the Index selector doesn't invert the selection, but instead invokes a count-from-end behavior.
Attempt to catch instances where users misuse this option, and print appropriate warning messages. (Particularly in cases with PDB numbering, where the count-from-end behavior doesn't really make sense).
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #4791 from RosettaCommons/vmullig/rename_within_HierarchicalHybridJDApplication
Replace "emperor"/"master"/"slave" within HierarchicalHybridJDApplication, which some people found objectionable.
I'm still not a fan of making an major effort to avoid _analogies_ to slavery when the tech industry and the University of Washington still make use of _actual_ slavery in the form of forced prison labour (i.e. I think we're just making institutions look virtuous that don't deserve to be made to look virtuous, which engage in a practice that's banned in most developed nations), but maybe once we've made this code change we can talk about something more concrete that we can do that actually helps to close the 13th Amendment loophole and to end slavery in the U.S.
TODO:
- [x] Change "emperor"/"master"/"slave" to "director"/"manager"/"worker".
- [x] Carefully check the integration tests. (Everything will change, but make sure nothing breaks.)
- [x] Update documentation --> see RosettaCommons/documentation#42
- [x] Check scientific tests --> they failed.
- [x] Update scientific tests.
- [x] Re-check scientific tests --> they passed.
- [x] Update `threads_per_slave` option.
- [x] Beauty.
- [x] Note that pull request #4761 must be merged before this one.
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #4766 from RosettaCommons/roccomoretti/vtune_1
Some simple efficiency tweaks.
I was playing around with the Intel VTune Profiler (on relax runs), and noticed a few "high spots" which could be addressed without too much restructuring.
Mainly these were in reducing the number of memory allocations, though there were a few runtime-related issues. I don't think there's substantial performance improvements here, but I think all of the changes fall into the category of "not worse", either in runtime or comprehensibility.
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #4768 from RosettaCommons/vmullig/multithreaded_frag_picker
Remove support for Boost::thread (in favour of cxx11thread)
Since switching to C++11, there hasn't really been a reason to use Boost::thread instead std::thread, and we don't really want two threading frameworks that we have to test. Indeed, the Boost::thread build seems to have bitrotted. This PR removes support for this, and adds a cxx11thread test for the fragment picker.
Task:
- [x] Add back missing .vscode file so that VSCode works "out of the box".
- [x] Remove Boost::thread support from fragment picker.
- [x] Remove Boost::thread build.
- [x] Switch other apps that use Boost::thread to use std::thread instead.
- [x] Integration test for fragment picker with threads.
- [x] Update documentation. --> No updates needed. It seems that the Boost::thread build was undocumented.
- [x] Fix single-threaded case.
- [x] Add documentation for multithreaded fragment picking.
- [x] Beauty.
Also:
- [x] This pull request updates the tools submodule to point at the latest master.
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #4839 from RosettaCommons/guangfeng/galigdock_genparams
GALigandDock mol2genparams python scripts update.
1. Added a function to rename atom names in mol2 files in order to fix duplicated atom names issue;
2. There is an issue that sometimes NH-CR in (C=O)-NH-CR is incorrectly assigned as a conjugated bond type. In some cases, e.g. APAJIL from CSD, NH-C=O is conjugated but NH-CR is not. To fix this issue, the dihedral angle of the input structure is used as a second check, if the angle (current range is from 40 to 140 degrees) is clearly not planar don't assign it as a conjugated bond.
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #4787 from RosettaCommons/BYachnin/bonusmap_for_ResidueTypeLinkingConstraint
Allow more complex behaviour with ResidueTypeLinkingConstraints
This class previously applied a bonus/penalty to scoring if two residues were the same identity. Reviewing the code, it looks like the intention was to expand this to allow for alternative pairing options, but this was never implemented. There were constructors that appeared to be unused and untested in the code base that had behaviour that didn't seem to be working.
We've updated this class to allow apply a custom bonus/penalty applied for any of the 400 possible residue pairs. A fully customized option is obtained by passing a std::map. The "legacy" behaviour is maintained by constructing this map based on user-supplied options.
All existing tests pass, and new unit tests have been added to test the new (and previously untested) functionality. Getters and setters have also been added to improve PyRosetta interface.
Thanks to @roccomoretti for the review!
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #4823 from RosettaCommons/benchmark
updating how Python virtual environment is setup and fixing Python virtual environment issues on Mac
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #4761 from RosettaCommons/vmullig/all_to_all_pnear_analysis
Add PNear analysis for many states.
I'm using peptide structure prediction to train ML models that can recognize peptides that fold well. A single structure prediction run actually generates far more data than I'm using. It would be useful to compute and output PNear to all of the negative states in addition to PNear to the state that I'm designing for, to give me more data for training. This PR adds this functionality.
Tasks:
- [x] Add new option.
- [x] Register new option.
- [x] Have the new option do something:
- [x] Add a new step to the `HierarchicalHybridJDApplication` class for this computation:
- [x] Sends a request for each state in turn from the emperor node to all slave nodes.
- [x] Each slave node broadcasts that state to all other slave nodes.
- [x] All slave nodes compute RMSD to that state.
- [x] Emperor collects RMSDs up the hierarchy and carries out PNear calculation.
- [x] Repeat for each relevant state.
- [x] Write out PNear list at end.
- [x] Error if used with `-compute_rmsd_to_lowest option.`.
- ~~Check and address issue #4738~~ --> separate pull request.
- [x] Documentation.
- [x] Integration test.
- [x] Beauty.
notify author
notify list [rosetta-logs@googlegroups.com]
Add UDPSocketClient access to pymolMover (#4753)
This pr adds getters/setters for the address/port of the pymolmover (or internally the UDPSocketClient)
the expected usage is shown in the test that have been added -> specifically the new set_link function.
notify author
notify list [rosetta-logs@googlegroups.com]