Merge pull request #2293 from RosettaCommons/jadolfbr/glycan_data_updates
Jadolfbr/glycan data updates
This PR updates the default glycan conformer sampling table with new data derived from the PDB using Adaptive Kernal Density estimates (same methods as the 2010 rotamer library) and a Von Misses Kernal to derive dihedral bins for each glycan linkage type seen in the data.
Raw Data was provided by Thomas Lutteke of Glycosciences.de and the densities were generated by Maxim Shapovolov and Roland Dunbrack.
This PR also updates a few default parameters to Glycan Relax and adds a few options for benchmarking. Integration test changes expected to tests that use Glycan Relax.
Density estimates were used to generate cubic spline(s) in matlab in order to identify torsion bins and assign conformers to each linkage data point using the inflection points of the spline to define single-dihedral conformer bins.
For example, the Chi1/Omega2 of an ASN a-D-GlcpNAc has three bins, as identified by the maxima of the density and boundaries at the minima on either side of each maxima.
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #2299 from RosettaCommons/vmullig/add_trimesic_acid
Re-activiate commented-out TMA unit test
The added unit test produces weird NaN behaviour ONLY in the Mac/clang build. I think that this is illustrating an underlying bug in the way that the Mac/clang build is doing math in `core::kinematics`, and this might be related to the existing unit test failure for that build (where different numbers are appearing than what we're getting for the other builds). It's easiest to debug the problem if the master branch is highlighting it. We might get to this at the noncanonicals meeting this week.
Andrew Leaver-Fay 7 years In as much as this unit test fails in the same way using gcc and clang on both mac and linux, I think it ought to be disabled, lest people grow comfortable with the idea that unit tests failing in master are ok.
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #2303 from RosettaCommons/vmullig/fix_score12_test
Fixing currently-failing unit tests on test server
The beta_nov15-as-default merge broke the ScoreTests:test_Score12 unit test. The changes seem to be purely cosmetic changes to output formatting. I'm updating the UTracer for the test to fix the test. Note that no score values change.
Vikram K. Mulligan 7 years That pesky protocols:RotamerBoltzmannWeight2Tests:test_rbw2_vs_rbw unit test is still failing only in the Mac/clang build...[list]
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #2216 from RosettaCommons/rhiju/swm_checkpoint_fixes
Merging this fix to stepwise checkpointing -- I think it's substantially more important for this to be working at the two basic levels (i.e., good checkpoints for last and low) rather than ensuring that it gets trajectory stats right too.
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #2075 from RosettaCommons/vmullig/add_trimesic_acid
Add trimesic acid (crosslinker)
This adds trimesic acid, a three-way crosslinker that can be used in peptide synthesis.
Tasks:
- [x] Add params file (note that crosslinkers are off by default in residue_types.txt).
- [x] Set up virtual atoms.
- [x] Set up connections.
- [x] Ensure overall neutral charge.
- [x] Ensure symmetry.
- [x] MM atom types.
- [x] Chi definitions and rotamers (using PROTON_CHI lines in params file). Note that trimesic acid prefers +/- 30 or +/-150 degrees for chi1, chi2, and chi3.
- [x] Add symmetric version of params file.
- [x] Add DAP, if it doesn't already exist --> It exists; params file updated. (Called DPP in PDB; renamed to DPP in database, too.)
- [x] Add DAB, if it doesn't already exist. --> It exists; params file updated.
- [x] Update ornithine params file.
- [x] Add sidechain conjugation patch for LYS, ORN, DAB and DAP, if it doesn't already exist.
- [x] Unit test for this.
- ~~Deprecate LYX? (Will have to update integration tests that use it.)~~ --> maybe in the future.
- [x] Add support for TMA to CrosslinkerMover:
- [x] Cherry-pick in change of name from ThreefoldLinkerMover to CrosslinkerMover (from pull request #1839).
- [x] Add setup logic for TMA in the asymmetric case.
- [x] Add setup logic for TMA in the symmetric case.
- [x] Allow LYS, ORN, DAB, or DAP side-chains to present TMA.
- [x] Unit test for functional form of constraints.
- [x] Integration test -- asymmetric case.
- [x] Integration test -- symmetric case.
- [x] Documentation.
- [x] Add support to simple_cycpep_predict for TMA.
- [x] Integration test.
- [x] Documentation.
- [x] Add weight option to GaussianFunc (needed for the construction of a suitable torsional potential for TMA).
- [x] Document changes to GaussianFunc.
- [x] Calculate correct values for torsion potential to ensure +/- 30 degree minima (calculus -- yay).
- [x] Add potential offset to give minimum value of 0 (more calculus).
- [x] Update PatchOperations to report their own names for errors / debugging.
- [x] Beauty.
- [x] Document change of name of ThreefoldLinkerMover --> CrosslinkerMover.
@gbhardwaj
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #2295 from RosettaCommons/everyday847/native_screen_design
Enable design in RMSD-screening stepwise runs
Vikram K. Mulligan 7 years @Sergey: The MPI-mode integration tests seem to be intermittently failing to run. It looks like a test server thing. Would you be able to take a look? (See also 59506, 59502, and 59496. Other revisions in between are fine.)[list]
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #2145 from RosettaCommons/vmullig/correct_xsd_info
Correct some XSD information
I've seen a few instances of incorrect descriptions in XSD stuff. This pull request corrects some of that.
notify author
notify list [rosetta-logs@googlegroups.com]