Rosetta is now FAT!
NEWS: This push introduces lipids into Rosetta.
In addition, it adds MM atom types to all carbohydrates and introduces the
MAINCHAIN_ATOMS field for topology (.params) files.
Full details described below:
* Two lipids have been added to the database -- the common fatty acid, linoleic
acid, and the amino lipid "head group", sphingosine.
This permits me to begin testing glycolipids in addition to glycopeptides,
oligosaccharide ligands, and polysaccharides.
You can read details of implementation decisions I made regarding lipids
here:
database/chemical/residue_type_sets/fa_standard/residue_types/lipids/README.txt
* To load the 2 lipid ResidueTypes, use the -include_lipids flag.
* I have added a glycolipid, galactosylceramide (GalCer) to the integration
test system (tested with the carbohydrates).
* A LIPID property has been added to ResdidueType with an appropriate private
datum and getter.
* The default main chain designation has been restored to what it used to be a
few months ago, i.e., Rosetta will, by default, search for the shortest path
from the LOWER_TERMINUS to the UPPER_TERMINUS. For any ResdidueType where
this is not correct, one can specify the main chain explicitly in the
topology file using the MAINCHAIN_ATOMS field, for example:
MAINCHAIN_ATOMS C1 C2 C3 C4 C5 C6 O6
* I improved debug output in several places, providing warnings where there
were easy ways to lead to silent seg faults from bad .params files.
* I added show() methods to both Residue and ResidueType, providing superior
information for either debugging purposes when introducing new ResdidueTypes
or using PyRosetta interactively.
* I performed some minor code refactoring to break up obnoxiously long
stretches of code in residue_io.cc.
* Added correct designations of main-chain, backbone, and side-chain atoms to
all carbohydrate and lipid .params files.
* Expanded, corrected, and clarified documentation/comments in ResidueType.cc
and residue_io.cc; in particular, I clarified the distinctions between main
chain and backbone atoms. (Also moved some comment blocks into the Doxygen-
correct location in the file.)
* Added MM atom types to all carbohydrate and lipid .params files.
* Made some other minor tweaks to carbohydrate patch files to improve accuracy
of scoring and nomenclature.
* Fixed severe bug in internal carbohydrate code regarding alignment of virtual
atoms.
----------------------------------------
Unit test status: Pass
Integration test status: 8 tests changed
rings, carbohydrates:
expected
hbs_design, oop_create, oop_design, oop_dock_design, scaffold_matcher:
from spelling error corrected in output
loop_creation:
missing fold tree output in one line
(I have no idea what this is about; please let me know if this
seems to be a problem.)