613: CASP10 Refinement Puzzle: TR655

Closed since over 13 years ago

Summary

• Created: August 03, 2012
• Expires: August 12, 2012 at 20:00 UTC
• Max points: 100

Description

The twenty-second CASP10 Refinement Target was the best server prediction for target T0655, generated by the Tasser_VMT server. We've been told that ~69% of this model matches the native, so you'll need to significantly change the starting model. Luckily, they have told us which regions to focus on 1-15 & 46-59 (as well as which to ignore: residues 86-88 & 137-158). These regions are shown in Note Mode. More info in the puzzle comments.

Top groups

• 21. hhv9 1 pt. 9,716
• 22. Crunching Family 1 pt. 9,687
• 23. Team Canada 1 pt. 9,685
• 24. SETIKAH@KOREA 1 pt. 9,650
• 25. Czech National Team 1 pt. 9,548
• 26. Eὕρηκα! Heureka! 1 pt. 9,540
• 27. Mojo Risin' 1 pt. 9,537
• 28. Team China 1 pt. 9,256
• 29. DSN @ Home 1 pt. 9,251

Top soloists

• 1. auntdeen 100 pts. 12,514
• 2. scottyler89 99 pts. 12,512
• 3. BitSpawn 97 pts. 12,486
• 4. doug.ehlert 95 pts. 12,450
• 5. bertro 94 pts. 12,418
• 6. SimplySue 92 pts. 12,383
• 7. tokens 90 pts. 12,357
• 8. Madde 89 pts. 12,352
• 9. Galaxie 87 pts. 12,345
• 10. smilingone 85 pts. 12,338

Comments

[beta_helix]

Please look at the Notes for this puzzle (by pressing 3 in the Original Interface, or pressing 1 in the Selection Interface).

---

This is the page for this CASP10 refinement target:
http://predictioncenter.org/casp10/target.cgi?id=212

The CASP10 organizers gave us this hint:

"The experimental NMR structure shows quite a bit of disorder in regions 91-93 and 142-163. Regions 6-20 and 51-64 need special attention. Starting model GDT_TS=69 (including the internal disordered regions)."

GDT_TS=69 essentially means that 69% of the model superimposes correctly onto the native structure:

en.wikipedia.org

Global distance test

The global distance test (GDT), also written as GDT_TS to represent "total score", is a measure of similarity between two protein structures with known amino acid correspondences (e.g. identical amino acid sequences) but different tertiary structures. It is most commonly used to compare the results of protein structure prediction to the experimentally determined structure as measured by X-ray crystallography, protein NMR, or, increasingly, cryoelectron microscopy. The metric was developed by Ada...

Here is the sequence logo predicted by the SAM server:

H = helix
E = sheet
C = loop (or coil)

The taller the letter at each position, the higher the probability of that specific secondary structure for that amino acid.

For example, the Valine at residue 92 is highly predicted to be part of a sheet. However, the Valine at residue 92 is predicted to be in a helix.

[Bletchley Park]

At the moment of writing, this puzzle is not visible in either main or devprev.

[brow42]

It's there now 8:00 PDT 15:00 GMT.

[Susume]

"Disorder" is not clear to me - does that mean those areas will not be scored for CASP so it doesn't matter if we move them?

[beta_helix]

Normally, disordered regions will not be scored.
See TR754 as an example:
http://predictioncenter.org/casp10/target.cgi?id=211&view=refinement
"Residues 1-8 and 77-80 are disordered in the experimental structure and deleted from the model."

It probably means that they will simply ignore that region (but you never know… they have never mentioned this sort of thing in previous CASPs), I would just not worry about those regions and focus on Regions 6-20 and 51-64.

Sequence Logo can be found here (since it keeps vanishing):