DISCUSSION of Rosetta@home Journal (2)

Message boards : Rosetta@home Science : DISCUSSION of Rosetta@home Journal (2)

To post messages, you must log in.

Previous · 1 · 2 · 3 · 4

AuthorMessage
hugothehermit

Send message
Joined: 26 Sep 05
Posts: 238
Credit: 314,893
RAC: 0
Message 30489 - Posted: 2 Nov 2006, 5:42:30 UTC
Last modified: 2 Nov 2006, 6:00:39 UTC

good question! all of the work units we would send out would keep the part that would be important for curing the disease constant but the other parts would be variable. this is what we are doing now, your computers are exploring different possibilities for the variable regions. if we can go interactive, you would help your computer explore the different possiblities. we would rank all the solutions that each of you return by their energies, and then experimentally test the lowest energy solutions.
now, suppose that you designed a protein and it turned into a blockbuster drug--maybe you should get some fraction of the royalties? thus far all income from licensing rosetta to companies has gone straight back into research and development and I want this to continue, but the double incentive of contributing both to global health and possibly getting some money back could make this vision of an interactive multiplayer rosetta "game" quite popular globally which would be great for human health and research (and who knows, there could be a 13 year old in Nepal who turns out to be a whiz at designing proteins!). but I should emphasize that this is still more fantasy than reality right now.


I would strongly suggest, that bragging rights is the be-all and end-all of this non-profit @home project.

ID: 30489 · Rating: 0 · rate: Rate + / Rate - Report as offensive
David Baker
Volunteer moderator
Project administrator
Project developer
Project scientist

Send message
Joined: 17 Sep 05
Posts: 705
Credit: 559,847
RAC: 0
Message 30491 - Posted: 2 Nov 2006, 6:23:30 UTC - in response to Message 30461.  

Interactive Rosetta - good Idea! In this wu:

"FRA_t364_HOMOENV_hom001_6_t364_4_1s5uA_IGNORE_THE_REST_545_1324_8_0"

On the graphic display the red colored end is pointing from the main structure into the middle of nothere, but it may fit perfect into the blue colored section by wrapping the red end around the blue section. The algorithm does not find this solution. May be there is a high potential barrier between the starting point and this possible solution.



this is just the kind of thing we want to enable you to do!
ID: 30491 · Rating: 0 · rate: Rate + / Rate - Report as offensive
David Baker
Volunteer moderator
Project administrator
Project developer
Project scientist

Send message
Joined: 17 Sep 05
Posts: 705
Credit: 559,847
RAC: 0
Message 30492 - Posted: 2 Nov 2006, 6:26:07 UTC - in response to Message 30484.  

Regarding the upcoming interactive mode: How does the protein "sense" its own energy level, if it indeed does so? Could the interactive mode provide "feedback" about what the energy level is at each point of the manipulation so that the neophyte protein-artist could have an immediate sense of the usefulness of a particular move, in seeking the lowest level?



yes, feedback on the energy change produced by your move will be critical so you can learn which directions are good to search in. ideally, there would be some kind of force feedback so you would feel the energy changes, but before this there would be numbers or colors on the screen.
ID: 30492 · Rating: 0 · rate: Rate + / Rate - Report as offensive
Branislav

Send message
Joined: 23 Mar 06
Posts: 6
Credit: 450,417
RAC: 0
Message 30762 - Posted: 7 Nov 2006, 11:49:57 UTC

It would be good if the interactive features could be accompanied with a scripting feature.
For example, you start with initial configuration, then you perform some manual moves, and if you are satisfied with the result you can save it as a script. You could run the script later (or some other users could run your sript and see how did you achieve your state).
For a real scripting you would need some condition checking also (if energy>initial_energy) as well as at least some elementary variable assignment (with access to energy and rmsd). Then you could enable us to search for best policies instead of best moves (which might be more protein specific).
ID: 30762 · Rating: 0 · rate: Rate + / Rate - Report as offensive
Mod.Zilla
Volunteer moderator

Send message
Joined: 5 Sep 06
Posts: 423
Credit: 6
RAC: 0
Message 31430 - Posted: 19 Nov 2006, 18:59:58 UTC - in response to Message 31429.  
Last modified: 19 Nov 2006, 19:02:01 UTC

THREAD CLOSED!

This seems like a good time to close this growing thread and cut a new one for further discussion. Please place new comments and questions in Discussion 3.

Rosetta Informational Moderator: Mod.Zilla
ID: 31430 · Rating: 0 · rate: Rate + / Rate - Report as offensive
Previous · 1 · 2 · 3 · 4

Message boards : Rosetta@home Science : DISCUSSION of Rosetta@home Journal (2)



©2023 University of Washington
https://www.bakerlab.org