| View Issue Details [ Jump to Notes ] | [ Issue History ] [ Print ] |
| ID | Project | Category | View Status | Date Submitted | Last Update |
| 0000138 | Rosetta | [All Projects] Incorrect Results | public | 2012-10-19 12:07 | 2012-10-19 12:07 |
|
| Reporter | smlewis | |
| Assigned To | smlewis | |
| Priority | low | Severity | minor | Reproducibility | always |
| Status | assigned | Resolution | open | |
| Platform | | OS | | OS Version | |
| Product Version | | |
| Fixed in Version | | |
|
| Summary | 0000138: InterfaceAnalyzer mishandles cross-interface disulfides |
| Description | InterfaceAnalyzer fails to treat cross-interface disulfides as broken in the separated state, leading to grossly incorrect dg_separated values, poisoned by about 6000 REU worth of disulfide energies. It is unclear how it should treat interfaces linked by a cross-chain disulfide.
A) Refuse to score such poses?
B) Perhaps score as disulfide in the complex case, and score as not-disulfide in the separated case (with repackable cysteines)?
C) Perhaps B but in both cases multiply the disulfide terms by zero post-scoring? (turning them off in the scorefunction may result in incorrect packing?) |
| Tags | No tags attached. |
|
| Application(s) Affected | InterfaceAnaylzerMover and its app |
| Command Line Used | InterfaceAnaylzer -s 1s78.pdb (an antibody FAB with cross-chain disulfides; use the CH1/CL subcomplex) |
| Developer Options | Confirmed As Bug |
| Fixed in SVN Version | |
|
| Attached Files | |
|
Relationships 
Relationships