You are here

Rescoring protein docking decoys to get Irms

4 posts / 0 new
Last post
Rescoring protein docking decoys to get Irms


I'd like to rescore protein docking decoys to calculate the Irms (interface rms) against the lowest energy decoy. If I rescore using the scoring application, I can get the overall rms but not the Irms. I've tried including flags like -evaluation:Irms but can't seem to get it to work. Also, although less important, is there away to apply the same scoring as the docking application does - without specifying any scoring patches or weights or anything the overall score is completely different to the docking output score.


Post Situation: 
Thu, 2014-10-23 13:23

For rescoring protein-protein docking structures, you'll probably want the Interface Analyzer application:

Thu, 2014-10-23 13:39

Thanks for the reply. I hadn't been aware of that, looks like some really useful analysis to have a look at. It doesn't however appear to calculate the Irms against an input interface, any ideas how to do that?

Thu, 2014-10-23 15:36

I think the (undocumented) "IRmsd" filter will do it. Just add it as a "confidence=0" filter to your RosettaScripts protocol, and it will output the calculated final value in the scorefile. This filter calculates the backbone atom rmsd for all residues within 8 Angstroms of the interface. It references off of the structure passed to -in:file:native (or the input structure if that is omitted), and you specify the interface by using the tag option "jump" to specify the numeric jump. (1 for a two chain structure.)

Wed, 2014-10-29 14:30