Merge pull request #3557 from RosettaCommons/vmullig/update_beta_aa_properties
A few updates to properties for beta-amino acid params files
This is intended to match the properties for the corresponding alpha-amino acids (and to make it easier to control amino acid composition by properties when doing mixed alpha/beta design).
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #3559 from RosettaCommons/vmullig/residue_index_description_issue
Resolve an issue with assignment of -1 to an unsigned int.
Tweaking ResidueIndexDescription, which was invalidly assigning -1 to an unsigned int. Note that this was *not* causing any breakage here, but it's a design pattern that confused the ICC compiler (which threw a warning) and might have confused future developers. It's probably best to avoid this in order to prevent *human* error.
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #3531 from RosettaCommons/JackMaguire/SerializedSAndL
Add Inputter/Outputter For Saving Serialized Poses To Disk
Sometimes we run rosetta_scripts multiple times without needing the intermediate output. For example:
```sh
rosetta_scripts @flags1
sort -nk2 score.sc | awk ... | head ... > pdblist
rosetta_scripts -l pdblist @flags2
```
I am adding an option to pass poses by printing serialized poses to disk instead of PDB files, silent files, etc. This allows for more of the information inside of the pose to survive from round to round and will result in faster saving/loading.
New flags:
-in:file:srlz
-in:file:srlz_override (does not terminate Rosetta when the serialized pose was made with a different version of Rosetta)
-out:file:srlz
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #3558 from RosettaCommons/vmullig/add_gordon_settings
Adding compilation settings for Gordon cluster at UCSF.
This does not affect code, the database, or default compilation settings. It can't cause problems, so I'll skip testing.
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #3555 from RosettaCommons/vmullig/fix_a3b
Small fixes to the a3b.crick_params file in the Rosetta database
The refactor of the parametric code added some checks for parameter values, which fail if a Crick params file has r1 values less than 0. I'm correcting the a3b.crick_params file, added by Andy, so that all r1 values are positive. (When flipping the sign of an r1 value, it's necessary to shift omega1 for the same atom by π radians).
This does not change the geometry. This is a minor change that will not affect 99.9% of users.
@everyday847
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #3526 from RosettaCommons/JackMaguire/MultistageRosettaScriptsSimple
JD3 can now create an archiving directory if needed (very small PR)
JD3 has an option to archive job results on disk instead of in memory. The existing way to do this is to create and empty directory and pass the name of that directory to JD3. I'm adding a hefty 8 lines of code to allow JD3 to create that directory if it does not exist.
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #3543 from RosettaCommons/sergey/f2
Updating nlohmann/json library to 3.3.0. This should enable JSON support in GCC v4.8+ and fix PyRosetta GCC build
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #3548 from RosettaCommons/sergey/binder
PyRosetta update. Refactoring include deduction logic so it give more more stable results across different C++ libs / lib-versions
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #3547 from RosettaCommons/mlnance/set_fail_status_BackboneMover
Set last move status in BackboneMover to fail do not retry when there are no movable positions
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #3336 from RosettaCommons/JWLabonte/sugars/database
Carbohdyrates: unit test to check stereochem./atom-placement of monosacharide residues in the database
This merge adds a complicated "dynamic" unit test that used a "rubric" file with expected answers to check carbohydrate .params files for ICOOR errors.
The errors it checks for include:
Proper superimposition of virtual atoms with real atoms
Expected stereochemistry of ring atoms (based on a rubric data file, stereochem_test.key)
Expected ring conformer (based on the same rubric file)
Hydrogen placement
The test checks both reducing-end and non-reducing end versions of the residues, by creating disaccharides.
The unit test reveals an assortment of ICOOR errors, as I expected, which I am systematically fixing. Notably, post-merge, hydrogen placement should always be correct — the problem was in the patch files for lower termini — and the RingConformerSet should correctly assign the current RingConformer consistently — the problem was in virtual atom superimposition.
I have also decided to organize the .params files in a deeper subdirectory scheme.
Finally, and perhaps most significantly, I have "idealized" heavy atom ICOOR records to use simpler, average, less precise values. I do not think this will affect scientific performance, though it may affect Cartesian minimization. If it does not, I think this is a better, more standardized way of writing sugar .params files and will simplify the implementation of my long desire to script the generation of hundreds of possible topography files for future design work.
All unit tests pass.
This merge will result in integration test changes for ALL tests involving glycans, as literally every sugar .params file has been modified.
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #3544 from RosettaCommons/JWLabonte/sugars/database3
Carbohydrates: enabling input of some N-acetyl sugars with Glycam codes
This merge will expand the list of Glycam codes that are recognized by Rosetta.
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #3533 from RosettaCommons/sergey/binder
PyRosetta update. Updating Binder to add support for banned types, refactoring is_bindable for classes so it no longer check if template parameters is bindable. Refactoring Rosetta includes.
notify author
notify list [rosetta-logs@googlegroups.com]