RNA Lab: "Christmas Trees," "Cub Scout Projects," and "Optical Illusions" :

  • 13
  • Article
  • Updated 6 months ago
Hi All,

As one of the EteRNA Beta-Test Players, I have the advantage of having already seen and learned a few things that really work well in Lab Designs - and also, quite a few more things that simply do not work at all.

Now I am not any kind of expert, and I have little to no science or microbiology background, however, the little bit I have learned in the past couple of months of playing EteRNA, I now feel obligated to share with the HUGE new community of brand new players.

The reason I am writing this is because I have been painstakingly going over most, if not all of the new Lab Design Submissions by brand new players, and have found that some of the mistakes that were made by the Beta-Test Players a couple of months back are now being repeated by the incoming bunch of brand new EteRNA-Players - only this time there are nearly 10,000 instead of merely a few hundred.

I almost convinced myself that I should just keep my mouth shut, and not say anything, but after thinking it over have decided that I need to speak out, because I believe that we will all benefit if someone does. When I say "all," I mean not only new players, but veteran players, and the Developers and Scientists of EteRNA as well, because, we are all after the very same goal: RNA Design Success!

Ok, Enough preamble.

What's a "Christmas Tree?" It's a design made solely, or almost solely of G-C pairs; Guanines and Cytosines; all reds and greens, hence the name: "Christmas Tree."

What's a "Cub Scout Project?" Similarly, a design made solely (or mostly) of A-U pairs; Adenines and Uracils; all Yellows and Blues - Cub-Scout Colors.

Finally, as you've probably already guessed: an "Optical Illusion," is a design which is overly heavy on G-U Pairs; Red and Blue (nobody will probably ever actually make this mistake - I just wanted to fill out the three)

I coined these three un-complimentary names for these types of designs, because, as any of the more experienced players will readily testify if asked, these designs simply do not work. They do not fold properly.

(By the way, the same is true for most extremes; super-high or super-low energy designs, overly repetitious patterns in designs, etc); Most winning designs have been moderate, balanced entries in all respects.

Anyway I was chagrined to find that not one or two, but some higher number (approaching a dozen) "Christmas Trees" were submitted in the Lab by New Players this round. Even worse, at least one of them had amassed a dozen or more votes (by other new players, I 'm sure - it had to be, no experienced player would ever spend a precious vote on a "Christmas Tree")

There were also a couple of "Cub-Scout Projects", but I did not see any "Optical Illusions," thanks God.

Now, there is no shame in any of this, not for designing one, nor for voting for one, (after all, how could you have known?), and nor is this a criticism or condemnation of any design or any player, but it IS an attempt to shorten the learning curve and prevent new players from making some common early mistakes; it is an attempt to help new players to preserve their precious design slots for VIABLE designs, and also help keep them from squandering even one of their precious few votes on designs which are certain to not pay them dividends in the form of points earned.

Remember that you get points not only for creating a winning design, but also for recognizing and voting-for a winning design.

There is still time to "un-vote," any "Christmas Tree Vote" you may have made, and, for those who designed these "Christmas Trees," (if your design is "un-voted" all the way back to zero votes) you can delete it, and re-submit a better design in its place.

I would like to Strongly Urge all new players to study all the past winning designs in earnest; try to see what characteristics they have in common, which you can then identify and abstract out, and then use in your own designs, as well as use to evaluate others' designs.

Remember ONLY 8 designs can be synthesized each round! The EteRNA Scientists must accept and synthesize whatever 8 we vote the highest. The holidays are over, please let us not send any "Christmas Trees" to the Lab to synthesize this January! (or ever again).

Let us not waste even one of our precious few design slots, even one of our precious 8 votes, and also let us especially not waste even one ultra-precious Synthesis Slot.

I should add that there were also a handful of new players whose designs looked like those of much more experienced players - some great designs; especially for first efforts from brand new players; many of these will receive many, many votes when the Veteran Players start voting, which will probably happen Friday night or Saturday Morning after the results of Round One are published. (Experienced players often wait to see the results of the previous round before submitting for the following round, first, to benefit from seeing what worked, and what didn't, and second, because if someone scored over 94% already, the current lab will end, and a new one will begin).

Study up! Get Familiar! Excel in your designs and in your design choices!

And Have Fun!

Thanks for putting up with this admittedly overly impassioned Rant.

Good Luck and Best Regards to all

-d9
Photo of dimension9

dimension9

  • 186 Posts
  • 45 Reply Likes
  • hoping this will be accepted as helpful

Posted 9 years ago

  • 13
Photo of Chris Cunningham [ccccc]

Chris Cunningham [ccccc]

  • 97 Posts
  • 13 Reply Likes
Has there been a "what did we learn from Lab 102 Synthesis Round 1" discussion thread?

It seems worthwhile for me to go through and look at the synthesized designs to see if I can pick anything up, but I'd much rather do so, post it, and ask for other people to add things that they learned. If we did this for every set of synthesized RNAs, we could make it highly-recommended reading.

The admins might consider making such a discussion thread officially linked somewhere, so that random new users will find it. Officially-sanctioned discussion threads seem like the only way to build an informed community about such things.
Photo of Chris Cunningham [ccccc]

Chris Cunningham [ccccc]

  • 97 Posts
  • 13 Reply Likes
Basically I'd like something more organized than GetSatisfaction for discussing what we should learn from old labs -- even if each thread of discussion actually happens on GetSatisfaction. Think XKCD's Comic Discussion Threads.
Photo of dimension9

dimension9

  • 186 Posts
  • 45 Reply Likes
I think this is a Fantastic Idea, Chris - I agree completely that we absolutely must try everything we can to build an informed community, and get the new players educated and up-to-speed. If you do not start that thread after this round - I will! :)
Photo of Chris Cunningham [ccccc]

Chris Cunningham [ccccc]

  • 97 Posts
  • 13 Reply Likes
Do you think it should be an Idea, titled something like "Things to Learn from Lab 102 round 2?" That makes it sound like the original poster has a good handle on it. Maybe a more democratic Question titled "What did you learn from Lab 102 round 2?" would be better -- with the understanding that the original poster should Reply to the question instead of including their interpretations as part of the question.

What do you think? If you like it or make some modifications, I'll go ahead and start it for one of the previous rounds.
Photo of dimension9

dimension9

  • 186 Posts
  • 45 Reply Likes
I like the second idea a little better; keeping it a bit on the impersonal side, so even original poster could answer as if he hadn't made the original post... I like your title, or perhaps something like..."Lab 102 - Round 2: A Retrospective" ...then inside say something like "Please add the most valuable things you learned this round to this post to record them for the benefit of the whole EteRNA community." The, we could add something to get the ball rolling. I think this is just a superb idea!
Photo of Chris Cunningham [ccccc]

Chris Cunningham [ccccc]

  • 97 Posts
  • 13 Reply Likes
Is there any way to make some kind of pun instead of the phrase "A Retrospective"...

like "LeaRNAbles from Lab 102 Round 2"

... that's not very good though... :)

This isn't at all necessary. :P
Photo of dimension9

dimension9

  • 186 Posts
  • 45 Reply Likes
hmmm... a smattering of quick, off-the-cuff title reflections: 1)"Lab 102 Round 2 Retrospective; The Last Christmas Tree." (perhaps a bit too sarcastically snide). 2) ...or some absolutely horrible pun: "RNA of you going to make a Brand New Xmas-Tree for Round 3?" nah, same problem as #1 :) ...3) "Rocky 102, Round 2 - Ding!"...4) (Somebody Stop me!) "Lab 102 Round 2: "A Tragedy in 237 Acts" -- that's probably wrong, I never did catch the final number of entries... ok, enough, I have proven both my insanity and my multiple utter incapacities for humor and appropriateness, as well as title-creation; please, you decide :)
Photo of Chris Cunningham [ccccc]

Chris Cunningham [ccccc]

  • 97 Posts
  • 13 Reply Likes
L
O
L
Photo of Matt Baumgartner [mpb21]

Matt Baumgartner [mpb21], Alum

  • 128 Posts
  • 33 Reply Likes
Puns aside, I think that this is a good idea.
Photo of cesium62

cesium62

  • 8 Posts
  • 2 Reply Likes
Maybe a few more columns should be added to the table that displays the submitted designs. We should be able to sort the table by melting point and energy. (Is a high melting point good or bad?) We could also score and sort by how balanced the number of different kinds of pairs are. Say, ratio of number of least frequent type of bond to number of most frequent type of bond.

It might be interesting to score longer sequences of bonds: there are 9 possible pairs of bonds; how well balanced was the choice of those 9 pairs? how well balanced was the choice of 27 triplets?
Photo of dimension9

dimension9

  • 186 Posts
  • 45 Reply Likes
hi cesium62, the table already has all those columns,(locate the sliders at bottom and right of table - they are non-standard design & may be easy to overlook - slider is white circle in black slider-guide). The sorts you mentioned are there.
Photo of Matt Baumgartner [mpb21]

Matt Baumgartner [mpb21], Alum

  • 128 Posts
  • 33 Reply Likes
I like the idea of having the bond ratios. I added it to the bug tracker as bug 357 Bug Tracker
Photo of Fomeister

Fomeister

  • 12 Posts
  • 4 Reply Likes
I am going to speak my mind here. First of all, I am a new player. I really want to complement on your taking the initiative in trying to help others. Also, I especially appreciate your helping people ot realize that their vote is important, and not to get into any kind "popularity contest".
Your characterizations of the 3 types of groupings, while descriptive, very much defeats the entire purpose of the project.
While you say that "experienced players know, these designs simply do not work". It is unconscionable that this comment has been published, without moderation or at least a response by EteRNA. In fact, I am posting this to them now.

Lest you misunderstand, the entire purpose of this game:

There have been NO ANALYSIS OF RESULTS to indicate that the 3 designs you have given misnomers will fail.

If, and when, such analysis is done, it is a simple matter to modify the code in the game to prevent someone from making a "Christmas Tree" or such.

The fact remains, the reason that the "game" allows for these combinations is that there are hundreds of "Christmas Tree" and other combinations you have shown disdain for, that do work.

The point of the game, is to interject intuition, experiment with the design, and after analysis refine the process.

I truly believe that your statement: "..I would like to Strongly Urge all new players to study all the past winning designs in earnest; try to see what characteristics they have in common, which you can then identify and abstract out," is incredibly simplistic, and condescending.

If you find yourself "tired of" the game mechanics which allow for the "Cub Scout Project" or "Optical Illusion" you should go here and solve the worlds RNA development problems.

http://mfold.rna.albany.edu/?q=mfold/...

But please do not detract from the stated goal of this project, based upon your beta-testing.

The goal is:
"...to truly tap the wisdom of crowds, our game would have to expose players to every aspect of the scientific process: design, yes, but also experimentation, analysis of results and incorporation of those results into future designs."

As you are, in your own words "not an expert", we should only make decisions based upon the expert analysis of the results.
Photo of dimension9

dimension9

  • 186 Posts
  • 45 Reply Likes
Hi Fomeister

Thanks for your very informed and incisive comments. Thanks for voicing your agreements with the things I said that you approved of, as well as for speaking your mind regarding the parts you are very critical of. The former are truly enjoyed and appreciated, but the latter are the truly valuable. While many might have held their tongue, or tempered their criticisms out of courtesy, or perhaps out of a recognition that fallibility is a universal affliction, much to your great credit, you got straight to the point, and told me to my face what you believed I needed to hear, irrespective of these artificial societal niceties, and for that, I thank you sincerely .

Although I prefaced this post in the second paragraph with these words, "Now I am not any kind of expert, and I have little to no science or microbiology background, " your reply has made me realize that disclaimer was insufficient to avoid coming off as "superior sounding" in my comments.

I will confess that I did feel like I knew at least a little bit more than many of the incoming new players, but I am afraid I did not, in my enthusiasm and concern of the moment - (trying to have others avoid the pitfalls that myself and many of the other beta-testers went through) - retain enough presence of mind to realize that my comments were a)"condescending," b) that they "detracted from the stated goal of the project.," and that they c) "showed disdain for," and d) "defeated the entire purpose of the project."

I am truly horrified that I came off that badly to even one person. I stand duly chastised, and I accept your eloquent criticisms and comments in the spirit of helpfulness, in which I am certain they were given, and were intended to be received.

Also, thank you very much for that link to "The RNA Folding Forum" at "The mfold Web-Server" at the RNA Institute, College of Arts and Sciences, University of Albany, State University of New York!

I am honored, but do not think my education level will qualify me to participate at that level.

(although I will give it a go, if for no other reason than to assure myself once and for all, just how little I really do know.)

In closing, let me just assure you, that while I am certainly humbled by your remarks, I do recognize that is not necessarily a bad thing; I can also assure you that in the future, I will certainly devote more time to examining my every written word before publishing, so as not to repeat such a faux-pas again in the future; and finally, let me reassure you that I bear no hurt feelings nor any grudge against you for speaking your mind so directly, that, in fact, that is a trait I both respect and admire, and I look forward to meeting you in chat for a direct mind-to-mind exchange of views. (although, it is only fair to warn you, that as a 62 year-old two-stroke survivor, my typing abilities in live chat are pretty damn bad [appalling actually], whereas, at my leisure, I can still compose an adequate response in intelligible English, with at least passable grammar, if not totally devoid of typos).

Thanks Again, and Best Regards,

-d9
Photo of Chris Cunningham [ccccc]

Chris Cunningham [ccccc]

  • 97 Posts
  • 13 Reply Likes
d9, a good way to fix up your guide to make it more palatable to complaints like these of the FoMeister might be to post directions on where to find the specific examples of things that failed in past labs. Is there a cub-scout project that has been synthesized and has failed? If not, the complaint here is probably valid.

The Christmas Trees do fail though. Still, some direction toward past results would make the whole thing sound more scientific, which should presumably avoid people taking things the wrong way.
Photo of Fomeister

Fomeister

  • 12 Posts
  • 4 Reply Likes
@d9 - I appreciate your taking the time to respond.
I am 100% hoping that we can collaborate on many new exciting labs, and along with the rest of the community help advance science.
As I discussed in chat last night, I believe that you are correct in that collaboration is one of the keys to the success of this project.

I sent you the link to the RNA forum as I feel that it will perhaps be enjoyable to you, in addition to our work here.

I look forward to working together.
Photo of salish99

salish99

  • 295 Posts
  • 58 Reply Likes
there's an entire thread on lab 102