I wanted to share an idea that might help the Rosetta project. I'm not sure is it good because I don't know exacly how the program works, but maybe someone with better knowledge could assess it.

The visualisation of Rosetta's work resembles a puzzle of making different patterns and finding the one with the lowest energy. And it could be an interesting task to try to make such a structure by a creator. People like logical puzzles and if there was a possibility of making own patterns, and thanks to that beeing able to participate by own in finding a shape of the lowest energy with a ranking of who made the best result, it would appeal to people.

As I said I'm not well aware of the project's work and can't be sure would this idea be helpful, but what is especially interesting is to use people's readiness to play logic games in researching projects that require logic operations.

Helping people try out their own ideas, and see for themselves how the atoms within a protein interact is basically the idea behind the Rosetta home game they have been working on. You can't actually enter in your own protein. But you can take an existing protein and study various possible conformations and compare your resulting score with others that are working on the same one.

Here is a thread with more information on the game. If you'd like to try it out, just follow the instructions provided there.

---

Rosetta Moderator: Mod.Sense

Helping people try out their own ideas, and see for themselves how the atoms within a protein interact is basically the idea behind the Rosetta home game they have been working on. You can't actually enter in your own protein. But you can take an existing protein and study various possible conformations and compare your resulting score with others that are working on the same one.

Here is a thread with more information on the game. If you'd like to try it out, just follow the instructions provided there.

Thank you for the link to the game. I just wanted to ask one more question. Wouldn't it be easier to make a model of amino acids chain, include information about every atom's size and forces, and simulate how this forces would change the chain until it folds into a structure of the lowest energy, of balance of forces? Sorry if because of my incomplete knowledge in that field I'm wrong somewhere.

The atomic structure of the amino acids is well-known. And yes, the modelling is operating at an atomic level. The current problem is that there are literally trillions of conformations possible for the given set of atoms.

Have a look at a protein in the protein data bank
http://remediation.wwpdb.org/pyapps/ldHandler.py?query=44C
The atom count, individual atoms, and bonds are all shown. In fact, with the JMOL viewer used there, you can define how you would like to rendure the vanderWaals forces of the atoms.

This is the raw data that is being worked with. The items in the protein data bank have structures that have already been discovered experimentally. This is shown with the JMOL viewer. The problem is there are hundreds of thousands more proteins with unknown structures. And the knowledge of how proteins interact with one another is quite limited. Rosetta is trying to accurately predict the structures on computers rather then with the expensive processes that are currently used. And to do so without having to run through all of the trillions of possibilities by finding ways to skip passed the possibilities that would not occur naturally.

---

Rosetta Moderator: Mod.Sense

Another problem is that although publically funded, the protein folding projects will not release their "energy functions" which means people developing their own folding algorithms cannot compare their results to that of another labs, thus cannot tell if they are getting anywhere.

---

Wave upon wave of demented avengers march cheerfully out of obscurity into the dream.