Help improve the Selection Interface

[beta_helix]

"We had been told quite a while ago that it was not going to be improved, an experiment that would stay as it was but likely phased out."

I believe that was my bad!
The selection interface was created by the Foldit Team at CMU (because they saw some of the advantages of it that were not possible in the original interface, see below). Unfortunately it became neglected for a very long time, so that did seem to be the case that nobody was going to be working on it anymore.

But as more and more scientists started using the Academic Standalone version of Foldit (tinyurl.com/academic-foldit), we started finding out that they all prefer the selection interface!

"Can you tell us why the scientists like the selection interface?"

Almost all the scientists that use Foldit do so for Protein Design, and they feel that it is infinitely better for designing proteins.
For example, the only way to mutate a bunch of residues to a particular amino acid (all at once) is with the selection interface. There is no way to do that (in a non-recipe way) with the original interface.

Thank you all for your feedback on this!
I just want to clarify that we are not trying to force any of you to switch to a completely different interface than the one you have mastered and been using for years!
Rather, we wonder if we were able to improve the selection interface with your feedback, would it made it easier for new players… would the steep learning curve that is Foldit be a tiny bit less steep?

(Obviously this would require significant work to get the wiki, and all the wonderful tools you have created for new players, updated to the selection interface)

[auntdeen]

Thanks for the reply!

It does seem, as far as I know, that the main advantage to the selection interface is that ability to mutate a bunch of residues to a particular amino acid. In fact, that's the only time I ever use it.

Is it possible to add that ability to the original interface?

IMHO, becoming adept at either interface likely has the same learning curve.

However, having new players start in the Selection interface would require many many hours of work for volunteers to change the wiki pages, etc. To do this simply because the scientists are happier with it for designing proteins is asking us to do a major amount of work for no real benefit to the players (who can adopt either interface once they are playing the real puzzles).

Clarifying the Tutorials, fixing the bugs in them, getting rid of the balloons and replacing with a box of clear instructions would likely make the Foldit learning curve a little less steep - without the volunteer work of doing over all the wiki pages involved.

[marie_s]

Cost of moving the tutorials to selection interface:

• http://foldit.wikia.com/wiki/Tutorial_Puzzles the wiki page on tutorials, easy if the text are provided (I use the translation file) and not very long but have to be anticipated and programmed as this page is used by newbies and the folders who help them on global. The volunteers have to be available for the task.

• remake all the videos on tutorials by Madde and rav3n_pl (http://foldit.wikia.com/wiki/Rav3n_pl_intros_corner)

• the other pages on the wiki: As I don't think they are use to know to access a tool but only to how use the tools, updating the other pages is not very important, they should have the 2 versions of the command,

• annoying the ones who help newbies in global who fold in old interface and never in selection. I think the only major problem is on "how to make a band?", the other questions on commands(the move tool, Q, zoom, save structure…) have the same answers in the 2 interfaces.

• remake all the translations: it is a very painful and long task to make a translation. Translations have many bugs (see here a collection of feedbacks http://fold.it/portal/node/995894). Moving the tutorials to selection interface means losing all the translations for a very very long period of time.

[jflat06]

"- rebuild less than 3 segments, I don't understand why it is not possible to rebuild a twisted segment,"

This has to do with how the rebuild tool works - I'll explain the steps that it takes.

1. The program looks at the selected sequence, looking at which amino acids and secondary structures are present.

2. The program picks a random subsequence of 3 continuous segments and inserts a cut at one end.

3. The program then does a look up into a database for protein fragments that have similar AA and SS to this 3 continuous segment sequence.

4. The program then copies this segment onto the current protein.

5. The program applies an algorithm called CCD to get the new endpoints of the cut close, and then closes the cut.

When you rebuild anything less than 3 continuous segments, all rebuild does is insert a cutpoint, apply CCD, and then close it. This is because there aren't fragment sizes of less than 3 in the database. You could hypothetically use fragment sizes of 2, but at 2 segments, the fragments are so small that you're essentially just modifying local DOFs instead of actually capitalizing on real-world configurations of that subsequence.

[bertro]

Adding:

• Segment Information window is vertically too small to display all the information. I have to repeatedly scroll down to see it all when the information is changing.

[spmm]

For new players I don't think there is any advantage in the 'general' interface, with a bar instead of numbers in the score display and restricted view and other options. That interface could go and the 'advanced general' interface be the default after the tutorials.

That would save one set of explanations :)

I think that most people playing who are not, really new or using the selection interface would be using the advanced general interface, not the simpler one.

Getting a full list of the hot keys for the selection interface on the wiki, or even in a post (hint hint) would be a great start for new and existing players.
If documentation around selection and the interface itself were more abundant then imo more people would use it.
There is a lot more to selection than just the mutation options.

[spmm]

sorry accidental down vote - is selection reading the same library as general because the rebuilds do seem different from time to time? I sometimes try rebuilds in main for a different result.

[jflat06]

The libraries are the same. However, in both interfaces, the fragments are chosen at random, so you'll get different results. But no, the interface should not matter.

[bertro]

Adding:

• Move Tool: make it "hover segment + 3" instead of "left-click to select segment + 3". Faster, less clicks.

[RicGray]

As of a couple of releases ago, they should be listed in the help dialog. However, their documentation could be better.