Merge pull request #488 from RosettaCommons/jkleman/mp-goodies
Jkleman/mp goodies
= added function split_topology_by_jump_noshift that doesn't shift the output topology, added unit test for this function
= bugfix in InterchainPotential that comes from virtual residues
= fixed std::out to tracer output in AddMembrane and membrane framework
= FlipMover:
* added option for random flip angle
* added unit test
= TiltMover:
* added Mover and unit test
* Tilts the downstream partner to/from the upstream partner in the membrane
= OptimizeMembranePositionMover:
* added Mover and unit test
* uses highres smooth score function to optimize membrane center and normal
* uses deterministic search, moving the protein through the membrane (+/- 10 A from initial center) and tilts the membrane planes +/- 45 degrees from initial normal in each direction
= SpinAroundPartnerMover:
* added Mover and unit test
* spins the downstream partner around the upstream partner in the membrane
= MPDockingSetup:
* fixed chain issue from before and added integration test
= MPDockingMover: added option for running lowres and highres, specified in constructor
= DockingInitialPerturbation:
* added variable stepsize for membrane proteins, which depends on the distance between the partners => much faster than 1.0 A stepsize
= MPFindInterfaceMover:
* added mover, integration test, pilot app and documentation (internal)
* pilot app is jkleman/mp_find_interface.cc
* this is for global docking in the membrane and runs lightening fast :)
= MPQuickRelaxMover:
* added mover, integration test, pilot app and documentation (internal)
* pilot app is jkleman/mp_quick_relax
* simplest relax which is lightening fast compared to FastRelax
* single model for 1000 residue protein is created in 10-15 mins, for 80 residues in ~60 seconds
= MPMutateRelaxMover:
* added mover, integration test, pilot app and documentation (internal)
* pilot app is jkleman/mp_mutate_relax
* makes a mutation and run MPQuickRelax
=====
unit tests pass
integration test changes: expected (mpdock and mpdock_prepack are fine, the rest are tracers)
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #553 from RosettaCommons/roccomoretti/cppcheckfixes
A number of fixes to some cppcheck issues.
These have been cross referenced against the earlier fixes Andy made, and should be distinct.
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #554 from RosettaCommons/vmullig/fix_library_levels
Trying to fix broken library levels.
Includes pointing to devel were left in protocols/relax/LocalRelax.cc/hh. Deleting these seems to be all that's required -- this doesn't break compilation, so whatever devel-level code that this protocol once depended on has already been removed or moved to protocols.
Library levels are fixed and basic compilation works in debug and release modes. I'm going to merge this without waiting for all tests to complete.
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #486 from RosettaCommons/roccomoretti/cppcheck
Add cppcheck test to benchmarking server framework
This test (build.cppcheck) will allow us to run the cppcheck code quality analyzer over Rosetta code and assemble the results. On the test server it takes about 0.75 hrs to run (parallelized), but should be faster if source files haven't been modified from the last run.
There are currently about 1700 things that cppcheck is flagging in Rosetta code. I tried to pick settings which were comprehensive but didn't seem nit-picky. If we disagree about particular things cppcheck is complaining about, there are various ways to make it less critical.
The intent is to turn this into an standard test, but we want to clean out most of the issues it's reporting first.
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #546 from RosettaCommons/vmullig/triazolemer_tweaks
Tweaking database patches for triazolemer class residues. Should result in no test changes.
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #483 from RosettaCommons/roccomoretti/valgrind_fixes
Fix residual valgrind errors
In order to fix the broker_membrane errors, I needed to alter slightly how the LegacyRootClaim and SequenceNumberResolver worked. I think I did it correctly, but if you're using these, you'll likely want to check that your protocols are not affected by the change.
notify author
notify list [rosetta-logs@googlegroups.com]
Merge pull request #453 from RosettaCommons/smoe/gcc-5_compatibility
gcc-5 compatiblity - increased sensitivity to type differences and possibly uninitialised values
notify author
notify list [rosetta-logs@googlegroups.com]