Placeholder image of a protein
Icon representing a puzzle

1311: Hfq TB puzzle - fixed

Closed since over 9 years ago

Intermediate

Summary

Created
November 30, 2016
Expires
Max points
100
Description

Note: Due to a large number of crash reports, this puzzle has been made worth zero points. The puzzle will remain active, and players are encouraged continue playing the puzzle if they can. We will still analyze players solutions for scientific merit!



This TB puzzle is summarized in http://fold.it/portal/node/2003049. It is a hexameric protein responsible for virulance factors in TB. We do have crystals and phase data, but none of our models work! Help us get a model to solve this structure.
VEVKIGITDSPRELVFSSAQTPSEVEELVSNALRDDSGLLTLTDERGRRFLIHTARIAYVEIGVADARRVGFGVGVDAAAGSAGKVATSG

Top groups

  1. Avatar for Go Science
    1. Go Science
    100 pts. 12,376
  2. Avatar for Beta Folders 2. Beta Folders 77 pts. 12,278
  3. Avatar for Gargleblasters 3. Gargleblasters 58 pts. 12,249
  4. Avatar for Anthropic Dreams 4. Anthropic Dreams 43 pts. 12,060
  5. Avatar for Contenders 5. Contenders 31 pts. 11,964
  6. Avatar for L'Alliance Francophone 6. L'Alliance Francophone 22 pts. 11,413
  7. Avatar for Void Crushers 7. Void Crushers 15 pts. 10,547
  8. Avatar for HMT heritage 8. HMT heritage 11 pts. 10,287
  9. Avatar for xkcd 9. xkcd 7 pts. 9,764
  10. Avatar for Eὕρηκα! Heureka! 10. Eὕρηκα! Heureka! 5 pts. 9,308
  1. Avatar for Bruno Kestemont
    1. Bruno Kestemont Lv 1
    100 pts. 12,376
  2. Avatar for johnmitch 2. johnmitch Lv 1 98 pts. 12,285
  3. Avatar for bertro 3. bertro Lv 1 95 pts. 12,278
  4. Avatar for caglar 4. caglar Lv 1 93 pts. 12,264
  5. Avatar for ManVsYard 5. ManVsYard Lv 1 90 pts. 12,249
  6. Avatar for frood66 6. frood66 Lv 1 87 pts. 12,107
  7. Avatar for Blipperman 7. Blipperman Lv 1 85 pts. 12,060
  8. Avatar for JayD7217 8. JayD7217 Lv 1 83 pts. 12,001
  9. Avatar for LociOiling 9. LociOiling Lv 1 80 pts. 11,998
  10. Avatar for ViJay7019 10. ViJay7019 Lv 1 78 pts. 11,995

Comments

jeff101 Lv 1

The Wikipedia article Susume cites above looks very useful.

I would like to add that a BLASTP search using the 90-segment monomer sequence
run on the pdb database found no significant similarities:
https://blast.ncbi.nlm.nih.gov/Blast.cgi?CMD=Get&RID=3ZKGN6R8015

Meanwhile, a BLASTP search run on the Non-redundant protein sequences (nr)
database found many hits:
https://blast.ncbi.nlm.nih.gov/Blast.cgi?CMD=Get&RID=3ZK5XJY3014

The latter search lists the DUF3107 superfamily with accession pfam11305.

Susume Lv 1

The few Hfq proteins that are solved and in the PDB have a different SS sequence than ours - they start with a helix and then have all their sheets. The little piece of RNA they bind to goes in the middle of the hexagon - so there are presumably some sidechains in the "hub" of the hexagon that contact the RNA. I would expect those sidechains to be heavily conserved across homologs to the proteins in the PDB. Even though our solution will be different from the solved ones (because of different SS), I would expect the same or very similar sidechains to fall in the hub of our hexagon. I would expect those sidechains to be heavily conserved across homologs to our protein (a different set of homologs than the proteins in the PDB have, because of different SS). If we knew what positions those conserved sidechains have in homologs to our protein, we would know which residues to put in the hub.

markm457 Lv 1

Last night, I really tried to work on 1311. I had 18 crashes in one hour, I counted. Shake crashes, a wiggle for more than a fraction of a second crashes, standard scripts crash as soon as they try to wiggle or shake. This is beyond frustrating.

Bruno Kestemont Lv 1

I have random crashes ONLY when I'm busy on screen.

No crash in the following situation:

(Windows 7)

View not viewing the symmetrics
Graph properties 0 memory and 0 length (susume)

Limiting movements of the mouse on the window (I surround the window) and very slow actions (not doing many thinks quickly one after the other).
Don't move the mouse when an action is ongoing (e.g. shake: only start and stop, nothing else).

Recipes: never crashed on window minimized.

It looks like the problem is linked to video display.

LociOiling Lv 1

I agree with Bruno that interacting with the window while a foldit action is running seems to cause crashes. But I've also had crashes while running a recipe with the window minimized. I haven't yet tried reducing undo steps and memory usage to zero.

I also noticed that the "show isosurface" view option only shows the isosurface for three of the six chains. See feedback 2003167.

Bruno Kestemont Lv 1

Several recipes can be used without crash if they don't contain a ShakeSidechainsAll action. Following recipes would be safe for using on 1311 (minimizing the window is much safe):

Quickfix 3.3.2, for fixing after hand work

Quaking Remix 1.4.4, for science valid rebuild

Tvdl enhanced DRW 2.8.1a, here you must absolutely check the option "Do shakes only local".

Rav3n_pl GAB BiS Filter v2.2.3, good for fine tune linking the monomer elements

Acid Tweeker V2.6.1 Filter, for fine tuning at the end

Bruno Kestemont Lv 1

As structure.shakeSidechainsAll() makes most of the bugs (possibly structure.shakeSidechainsSelected as well when all or non residues are selected), a short term solution might be to re-post the puzzle without Shake.

When shake will be called, nothing would happen (and no crash).