Merge pull request #393 from RosettaCommons/everyday847/nrchi_minimization
Non-rotameric chi minimization, the utterly trivial option to select all the non-rotameric chis in a protein and minimize them after every repack, is now implemented. Default to off because of course it is! Won't work for NCAAs, but that's okay--once I've made enough NCAA densities style rotlibs to make it worthwhile to actually use the SEMIROTAMERIC_NCAA property enough, I'll use it here too...
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #410 from RosettaCommons/aleaverfay/dfpmin_memory_alignment
This commit addresses the memory access pattern for DFPMin. It produces many (real) integration test changes on my machine (though not on the testing server), but the change is small enough that I am confident the changes do not reflect a bug.
...
Changing dfpmin's access pattern to the HESSIN matrix to better align with memory.
This change is small, and though it looks like it ought to produce no integration test changes, it does. Why? I suspect that it's due to code that previously prevented vectorization now suddenly allowing it. Vectorized arithmetic produces subtle numeric differences.
Updating the Hessian accounted for a large amount of the running time in relax when run on some of Stampede's machines.
Thanks to Cyrus Proctor at TACC for finding this issue!
notify author
notify list [rosetta-logs@googlegroups.com]
Merging code duplication elimination efforts from everyday847/everyday847/simian
The pull request documents the specifics:
1) A refactored agglomerative hierarchical clustering system in numeric
2) rsigma_dot gets its data from the database now, instead of 3 separate, identical 866 line regions of the code
3) ChiralMover relies on data from the database (CHIRAL_EQUIVALENT in params files) rather than big std::maps
4) NCBB design apps rely on common main loop, statistics functions, and perturbation foldtree functions
5) Ditto the NCBB dock-design pilot apps/protocols, but the main loops are still separated (sampling differences)
Integration tests are limited to log output changes (a couple tracer name changes, basically)
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #398 from RosettaCommons/tlinsky/remodel_mover_option_fix
Checking for the pymol mover option should use .user() rather than .acti...
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #399 from RosettaCommons/jkleman/jkleman/mp_updates
MP updates:
* rewrote and extensively tested SpanningTopology constructor from PDB
* it is vastly improved and recognizes spans better
* I tested on 34 previously broken cases, except a handful they work now
* small problems that still occur are due to Dssp recognizing loops instead of
secondary structure elements
* a difference now is that spans that might not completely span the membrane
might be recognized as "TMspans"
* also, loop residues (except up to 3 residue loops bridging spans) are
now excluded from the spans (and there seem to be quite a couple in the
structures that I looked at)
* this impacts the spanfile_from_pdb app!!!
* added unit tests for this constructor
* added unit test for average_antiparallel_embeddings in MembraneGeometryUtil
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #407 from RosettaCommons/rfalford12/update_pHmode
Update pH Energy Function for Membranes
Adding electrostatic term (fa_elec) with a weight of 0.875 to the membrane pH enabled full atom energy function - needed for pH mode pyrosetta demo. No test changes expected.
notify author
notify list [rosetta-logs@googlegroups.com]
A few updates to hybridize protocol.
Adding a few new parsable options to hybridize, allowing control of the fraction of fragment versus template insertions. Ading some code to cart bonded energy for better handling of non-protein residues.
notify author
notify list [rosetta-logs@googlegroups.com]