「view this page in B3 βῆτα server」

Revisions №58307

branch: master 「№58307」
Commited by: Frank DiMaio
GitHub commit link: 「247a53599494c20a」 「№929」
Difference from previous tested commit:  code diff
Commit date: 2015-11-23 21:09:43

Merge pull request #929 from RosettaCommons/dimaio/rama_prepro This checkin adds the latest version of the beta energy function. It is enabled with the flag -beta_nov15. Also adding the flag -beta which will always point to the most recent energy function version. Details of the energy function are available on the doc wiki at https://www.rosettacommons.org/docs/wiki/Overview-of-Seattle-Group-energy-function-optimization-project (overview) and https://www.rosettacommons.org/docs/wiki/Updates-beta-nov15 (this specific update). 29 integration tests fail. Most are related to: a) cart_bonded scores changing due to typo being fixed in the default set (e.g. hybridization) b) cart_bonded tracer change (e.g. antibody_designer) c) the addition of new scoreterms cause the database_* and features* tests to fail d) debug output removed from code (fuzzy) e) bugfix to symmetric disulfidize (symm_disulfidize) The following integration tests change for other reasons: ddG_of_mutation enzdes flexpepdock inv_kin_lig_loop_design membrane_relax mp_relax mp_relax_w_ligand next_generation_KIC pocket_relax splice_* On closer inspection, all of them look to have very small score changes in their logs, which then propagate (perhaps the new score terms introduced changes the order in which scores are summed?) Additionally, there are changes to sweep_respair_energies that are difficult to trace (since its just a binary diff) but I suspect they are related to the above.

...