Questions about Chicken Anemia Virus (puzzle 731b)

Case number:845799-995388
Opened by:Susume
Opened on:Friday, June 21, 2013 - 19:20
Last modified:Saturday, June 22, 2013 - 16:38

The only pictures I can find (by a layman's search of the web) of CRM1-binding NES (as mentioned in the comment on the blog post) seem to be of helices. Are there other examples of proteins with sheets that perform this function? (Would love an answer to this even if it's after the puzzle has closed.)

The Zhang predictions we were given have the isoleucines in residues 30-46 (which I gather are the active part of the NES) pretty deeply buried. Is it possible (maybe with Rosetta?) to generate predictions that have them nearer the surface (only slightly buried)? I have been trying to position them near the surface, with just a single "flap" over them that might pull away during multimerization, but doing that changes the protein pretty radically from the zhang predictions, to the point where any predictive value in the remainder of the zhang prediction is probably lost and I may as well fold it completely de novo.

(Fri, 06/21/2013 - 19:20  |  2 comments)

marie_s's picture
User offline. Last seen 1 year 33 weeks ago. Offline
Joined: 05/18/2008
Groups: None

Maybe, a next puzzle with Rosetta templates and best solutions by players chosen by Beta Helix or by votes.

beta_helix's picture
User offline. Last seen 13 hours 49 min ago. Offline
Joined: 05/09/2008
Groups: None

Reply from the Worcester Polytechnic Institute Scientists:

Great question.
We were also bothered for a while that most CRM1-binding NES are helical. We only found one other NES that was a sheet: STAT1, specifically 392-413 (
However, when we lined up the known active residues of Apoptin (which are between 40-46) with the STAT1 NES and a well-known Helical NES, SNUPN-1, two crucial leucine residues in all three proteins lined up almost perfectly. So we believe that the NES is a beta-sheet, even though it is unconventional.

As for the second question, just about all of the structure predictions we've used have the NES buried. Please feel free to move away from the Zhang predicted structures, since the estimated accuracy of the model we generated is fairly low; there are few proteins similar to Apoptin with a known structure. Our thoughts are also that anti-parallel beta sheets act as a flap to expose the NES sequence, so keep following that idea.


Developed by: UW Center for Game Science, UW Institute for Protein Design, Northeastern University, Vanderbilt University Meiler Lab, UC Davis
Supported by: DARPA, NSF, NIH, HHMI, Amazon, Microsoft, Adobe, Boehringer Ingelheim, RosettaCommons