Merge pull request #4805 from RosettaCommons/smlewis/fix_lemmin
Update rosetta_citations.txt
fixing Gordon's name
this will probably cause cosmetic itest changes if anything uses this code in the itests
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #4797 from RosettaCommons/roccomoretti/address_unit_release
Fix the release-mode version of the EPRSpinLabelTest unit test
Convert an exact comparison of a real to a delta comparison.
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #4784 from RosettaCommons/everyday847/better_FMI_debugging
A few extra .trace output statements for debugging FullModelInfo problems
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #3470 from RosettaCommons/JWLabonte/sugars/database2
ResidueTypes: Modifying patch system to allow for multiple aliases per atom
This merge allows one to add multiple atom aliases from a patch file, which previously was only possible from a .params file.
It further adds more aliases to a few sugars, which will help Rosetta load in a few obscure sugars from the PDB that currently fail because of atom naming.
This merge also fixes an error in H-placement on acetyl groups that are not trans.
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #4495 from RosettaCommons/yingyue0414-Add-tetroseandtriose-to-database
Complete .params files for triose and tetrose residues in database
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #4148 from RosettaCommons/JWLabonte/PDB_IO/HETATM_records
PDB IO: Correcting how HETATM records are decided
This merge corrects how Rosetta decides between writing a PDB ATOM record and writing a HETATM record.
Historically, Rosetta saved LIGAND ResidueTypes with HETATM and POLYMER ResidueTypes with ATM. This was in conflict with how the records are defined by the PDB, which is that only canonical AAs and NAs get ATOM records; all other types of residues should receive HETATM.
See issue #3509 for further details/discussion.
All unit tests pass.
98 integration tests give expected changes with a multiplicity of .pdb output lines changing from ATOM to HETATM.
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #4774 from RosettaCommons/roccomoretti/sf_weighted_sum
Don't spend time multiplying by zero with linmem_ig.
Performance analysis of a fixbb run of 1qys with the -linmem_ig flag indicates we were spending ~20% of the time doing EMapVector::dot() calls to multiply the edge scores by the weights.
Given that most of the weights are typically zero, we can improve runtime by only multiplying through by the nonzero terms. This requires us to pre-calculate and store the non-zero terms for the ScoreFunction, but that turns out to be relatively straightforward.
This PR adds a ScoreFunction::weighted_sum() method, which does the dot product between an EMap and the weights of the ScoreFunction, but only iterates over the non-zero terms. I then use that for the "hot" function in the OnTheFlyInteractionGraph.
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #4693 from RosettaCommons/revert-4692-revert-4624-vmullig/threadsafe_approximate_buried_unsat_penalty
Re-merge "Make the ApproximateBuriedUnsatPenalty threadsafe". (Reverts RosettaCommons/main#4692).
This re-merges the thread-safety fixes to the ApproximateBuriedUnsatPenalty.
Tasks:
- [x] Figure out what's causing the bug and fix it. --> A spurious pair of runtime_asserts have now been removed.
Note: this diff (https://github.com/RosettaCommons/main/pull/4693/commits/d6dd51482bc8e7d857263381317db30f544b5c6d) shows the only changes from pull request #4624 (for which this is a reversion of the reversion).
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #4770 from RosettaCommons/smlewis/remove_blacklist_in_PDBdiagnostic
Smlewis/remove blacklist in pdbdiagnostic
Per a good idea I saw on the internet, renaming "blacklist" to a more neutral "blocklist" within the PDB diagnostic test.
Will alter only that test itself.
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #4191 from RosettaCommons/JWLabonte/sugars/database
Carbohydrates: Expanding recognized GLYCAM codes
This merge introduces updated and expanded GLYCAM 3-letter codes for input from AMBER, following the documentation provided here: http://glycam.org/docs/forcefield/glycam-naming-2/
@mszegedy and @thieker, this may be of interest to you.
This PR is a response to issue #3546 and a continuation of work begun in PR #3544.
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #4457 from RosettaCommons/JWLabonte/PTMs/cofactors
Coenzymes: Adding SAM and S-adenosyl homocysteine to the database
This merge adds .params files for SAM and its demethylated version.
It also adds an integration test for the loading of cofactors in our list.
This merge also fixes a bug that caused the sugar_bb scoring term to crash when ligands were involved.
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #4773 from RosettaCommons/vmullig/skip_submodule_update_with_nover_option
Two small tweaks to scons build system
This pull request:
- Alters the behaviour of the --nover flag, so that in addition to skipping the versioning script, it skips the submodule update script. This saves time on slower filesystems during rapid rounds of development and compilation, if the developer knows that these submodules do not need to be updated.
- Updates the help that one sees with `./scons.py --help` to:
- Add examples for mpi, serialization, and cxx11thread builds.
- Add the recommended default compilation command to build vanilla Rosetta for production runs.
- Clarifies that `-j` accelerates _builds_ and not _Rosetta execution_, while `extras=mpi` or `extras=cxx11thread` can accelerate _Rosetta execution_ and not _builds_.
- Adds the `--nover` flag to the help.
notify author
notify list [rosetta-logs@googlegroups.com]