Placeholder image of a protein
Icon representing a puzzle

672: Starting Asymmetry

Closed since about 13 years ago

Intermediate Overall

Summary

Created
January 26, 2013
Expires
Max points
100
Description

This puzzle starts out asymmetric but needs to be made symmetric. You can move one half of it to make it match the other half, and fit them together!

Top groups

  1. Avatar for Void Crushers
    1. Void Crushers
    100 pts. 6,700
  2. Avatar for Anthropic Dreams 2. Anthropic Dreams 85 pts. 6,691
  3. Avatar for Gargleblasters 3. Gargleblasters 72 pts. 6,681
  4. Avatar for Go Science 4. Go Science 61 pts. 6,674
  5. Avatar for L'Alliance Francophone 5. L'Alliance Francophone 51 pts. 6,619
  6. Avatar for HMT heritage 6. HMT heritage 42 pts. 6,617
  7. Avatar for Russian team 7. Russian team 35 pts. 6,613
  8. Avatar for Beta Folders 8. Beta Folders 28 pts. 6,587
  9. Avatar for Contenders 9. Contenders 23 pts. 6,586
  10. Avatar for foldeRNA 10. foldeRNA 19 pts. 6,582
  1. Avatar for Timo van der Laan
    1. Timo van der Laan Lv 1
    100 pts. 6,700
  2. Avatar for Deleted player 2. Deleted player pts. 6,682
  3. Avatar for gramps 3. gramps Lv 1 97 pts. 6,668
  4. Avatar for frood66 4. frood66 Lv 1 95 pts. 6,645
  5. Avatar for MurloW 5. MurloW Lv 1 94 pts. 6,634
  6. Avatar for karstenw 6. karstenw Lv 1 92 pts. 6,626
  7. Avatar for O Seki To 7. O Seki To Lv 1 90 pts. 6,617
  8. Avatar for pauldunn 8. pauldunn Lv 1 89 pts. 6,615
  9. Avatar for johnmitch 9. johnmitch Lv 1 87 pts. 6,613
  10. Avatar for utaca 10. utaca Lv 1 85 pts. 6,612

Comments

BootsMcGraw Lv 1

We're supposed to make the two pieces symmetric, but the frozen piece is all stretched out. That's not right. Y'all want us to make both halves stretched out??

Bletchley Park Lv 1

How does the scoring function guide us here ? how well the unfrozen piece matches the frozen one AND how they interact ? Or is this just about how well copied the whole thing is and how nice the symmetry ends up ?

Timo van der Laan Lv 1

It took me 45 minutes to write a simple script to copy the structure.
Simple, copy the distance from segment i+50,j+50 to seg i,j. We can make a band between i+50 and j+50 and use the bandlength to do just that.

Bletchley Park Lv 1

And if I may auggest; If this is anout what the best symmetric organization is, it would help to receive both parts frozen and idential so we do not have to waste our time trying to make an identical copy.

Bletchley Park Lv 1

I would like to reiterate my question. What is the science goal to achieve ? Are we to keep the shape frozen yes or no ?

If we first need to COPY the existing shape, there is no point in KEEPING that copy if the structure is not to be frozen afterwards in order not to loose that shape. If we can unfreeze it, there is no point in COPYING the shape in the first place.

This information will greatly impact the approach. Without this information the exercise is pointless as it will only be a race towards points, guided by an undisclosed scoring function.

jflat06 Staff Lv 1

The objective is to fold the non-locked copy to match the locked one, and move the unlocked copy into what you believe is the best symmetric orientation. This protein is symmetric in nature, and this is a sort of pseudo-QTTN and docking problem.

phallicies Lv 1

There seems to be a lot of funky stuff going on in this puzzle. I can make it symmetric perfectly, dock it and get at most 3700-4100. (depending on disulfide bonding.) it seems that all the side chain scores on the locked protein as well as on the unlocked are way to low to just be by itself. i think this is severely hindering the folding and making it impossible to make it symmetric and score good. our top score for our group is so dis-formed and far from symmetric that its not even funny. we either need the entire protein if there is one, or some sort of bonus/penalty for not following symmetry.