Intro
While I was working on making my entropy/switch strategy more specific, I started wondering. Do turnoff labs have higher entropy than turnon labs? And it really seems to be the case. So I’m splitting strategies after this feature.
Strategy
Turnoff switches (MS2 is turned off in state 2)
Give +2 to designs having entropy between 1 to 1.4
Give +3 to designs having entropy between 1.5 to 2.
Turn on switches (MS2 is turned on in state 2)
Give +3 for entropy between 1 to 1.4
Give +2 for entropy between 1.5 to 2
General
Give -2 to designs having entropy below 0.9.
Give + 1 and decrease to 0, to designs having entropy between 2.1 to 3
Penalize exponentially for anything having an entropy above 3
Thx to Pablo for pointing out the relation between medium high entropy and switches.
Background
This strategy will miss some of the good switches, as a few of them land in entropy area of solid static design or an entropy range that is not scored high. I’m assuming that then it is the folding algorithm that in these cases are not able to predict the right entropy. Since designs that usually fall out are the ones that the energy model deems unstable, so this may be a way to identify them, so they don’t get penalized to hell. Eg. perhaps turnoff entropy judgement in these cases.
Some of the other fallout winners that ends with an entropy in the area of static switches, are full moving switches.
So this strategy will not uncover all good switches, just cover a general good entropy range for most MS2/FMN switches. If designs land in rewarded area of entropy, it’s good. But it’s no guarantee they will also switch as wished for. Just its likely they actually do switch. 
Background articles
Switches and entropy
Part II - ROBOTS, REPEATS & ENTROPY
Entropy, RNA and free energy
Pablo’s lecture on RNA, entropy and free energy
2 Likes
Dear Eli Fisker,
Your strategy has been added to our implementation queue and we should have it completed within the next two to three weeks. You can check the status of the implementation here.
We will also be uploading the code used to write your strategy on this page if you are interested in reading it later.
After implementing your idea, we will post the direct link to the code here. Then, we will then optimize the parameters, and test it. At that point, we will tell you the correlation of how good your strategy was and what the new optimized parameters are.
Thanks for sharing your idea!
Vineet Kosaraju
The EteRNA Team
Hi Eli,
When you mention entropy for the puzzle you are referring to the positional entropy calculated by Vienna correct? If so, for this strategy do you want to look at the average entropy for the entire RNA, or the average entropy at the nucleotides where the binding occurs? Or something else?
Thanks,
Vineet
1 Like
Hi Vineet!
Correct, I am referring to entropy as calculated by Vienna.
Sorry, I didn’t specify. I was thinking about maximum peak entropy. I see if entropy get to high it is not advantageous for the RNA switches. It simply opens up to way too much potential switching - and most of it unwanted. So it is absolutely best if high entropy turns up specifically in obligatory switching area or those bases that gets paired with some of those. I actually think there may be an entropy preference between which bases will rather have the high entropy - the obligatory ones or the ones they partner with. I will look out for it.
However I do think that entropy averages in in binding nucleotides versus static nucleotides are interesting.
What I actually see for RNA switches are a quite unevenly distributed entropy - compared to static designs. There is a far bigger difference in entropy in switches than there is in static designs. So comparing average entropy in switching area versus average entropy in static areas in a RNA switch, might reveal a somewhat fixed relation ship.
But I think that is covered in this strategy. Please correct me if I’m wrong: https://getsatisfaction.com/eternagame/topics/-strategy-market-switch-entropy-in-static-area-in-part…
1 Like
Hi Eli,
Thanks for the clarification Yep the other strategy covers averages, I’ll implement this one to cover peaks/maximums.
Thanks,
Vineet
Dear Eli Fisker,
I am pleased to announce that a preliminary version of your strategy has been implemented with 34 lines of code. The code might slightly change in the future to remove bugs, but you can check the current version here. We will soon work on optimizing your parameters and testing the strategy.
Thank you again for sharing your switch strategy!
Vineet Kosaraju
The EteRNA Team
