18 Nov
2011
18 Nov
'11
3:54 p.m.
On Fri, Nov 18, 2011 at 7:50 AM, Wei Shi
The sequence identity of the search model to my protein is 33%. I haven't tried mr_rosetta using this search model without pre-refinement. I only used this search model directly to do molecular replacement using phenix, and no viable solutions are generated.
It may also help to run the model through phenix.sculptor, which will modify it to more closely match the target protein (without doing anything to the conformation). Sometimes this is all that you need to make MR successful - although it's never guaranteed to work at 33% identity. -Nat