Our group is working on an ab initial density modification method to solve the phase problem.
Last year, I applied the methods to the cases with a initial template which has already been correctly placed in the unit cell.
I tried several cases in the
Terwilliger, T. C., et al. (2012). "phenix.mr_rosetta: molecular replacement and model rebuilding with Phenix and Rosetta". Starting from the templates provided in the http://www.phenix-online.org/phenix_data/terwilliger/rosetta_2011/ I can get a initial density map, then apply our method to reduce the phase difference to around 50° in the highest experiment data resolution.
Now, I might want to try more blind cases with our methods. If we can solve more structures, that would be more convincing.