Hi. i'm new in rosetta, but was in seti since 1999.
I think your's (mine) :) work is more important for humanity.
...so i try to understand as many as i can about this project. When i watch screensaver i observe (that's my observation)searching is almost random ??? is it correct? when my computer searching the lowest energy and (i think)
the lowest RMSD (am i wrong we search the closer result to the native?) ...in the "speed" search, it is mindless.
no matter if he found the closer solution next result may be far worse than earlier... is it ok?
can you explain me this mechanism. ... i almost sure it isn't mindless search :)) but i want to understand whole mechanism
to know more.. to be more effective help to this research.

dont be angry if its sily question. i simply try to understand how it works.
have a nice day (or night)
yours
Marek

---

Welcome to Rosetta Marek!

You are certainly right about working on Rosetta rather then SETI. I would much rather answer "yes" to the "can we find a cure?" question, then "yes" to the "is something out there?" question.

Each model starts with a random shape the protein might take, and then works to refine it based on the science behind Rosetta. Some of those random starting shapes will be a good foundation to build on, and others will be poor. By trying many many initial shapes we try to assure that we find some of the good ones.

RMSD is a comparison to the structure that has been found by other processes. The idea is for Rosetta to compute it's way to the same answer that has been found by other methods. For some of the proteins we study, no other processes have found a structure to compare to. A lower RMSD means you are closer to the "correct" structure. Zero would be a perfect match. When new proteins are being studied with Rosetta, RMSD will just show a "?". But the energy level will be your guide. Lower energy (i.e. a large negative number) means a tighter fit of the atoms that make up the protein, and the true form tends to have a very tight fit. So, finding a low energy level tends to be the best prediction.

What you are seeing in the graphic, as it jumps wildly is a high level study of the possible bends in the chain. You have to keep in mind that you are not seeing ALL of the ideas your computer tested. So it seems to wildly jump around, but it actually tested several shapes in between the two frames you see as well. You will find that later on, it settles in and makes smaller twitches to the prediction. This is called the full atom relax stage. The trick, is trying to figure out ahead of time, which high level shapes will be worth the time to do a more detailed study on.

As time goes on, the science of Rosetta changes and different ideas are tested to make better and better decisions at the high level. But, you will notice that the energy level drops significantly during the full atom relax stage, so it is very difficult to predict, ahead of time, how your energy level will look when you complete the full atom relax. This is what makes the problem difficult to solve, and why your help is needed for Rosetta to advance.

---

Add this signature to your EMail:
Running Microsoft's "System Idle Process" will never help cure cancer, AIDS nor Alzheimer's. But running Rosetta@home just might!
https://boinc.bakerlab.org/rosetta/

Hi.
I very thankful for Your absolute answer, Feet1st.

I truly agree that in higher popability answer ""yes" to the "can we find a cure?" question, then "yes" to the "is something out there?" question." :))

I'm physiotherapist, nad chiropractor.. with strong keen on helping suffering people even if they cant afford to pay me my usualy charge. I have no doubt if its only a small chance to help cure anybody (especialy without unnecessary suffer) we should try to.
I have no doubt, processing this rosetta program, we can help humanity to be more happy, healthy... etc, too.
Maybe someday when we will be heathy, happy..etc...and we will meet ET :)) :)) we would say "hallo.. can we help you?" rather than "freeze we've got a gun."

Ones more thank you for your answer. im a bit closer to understand what we are doing and how can i help (ive got some polish friend testing proteins with classic method. im convinced if they will find someting which can help with Rosetta computing they let us know)
...and once more im sure "he" was right when said "Quod scimus, gutta est, ignoramus mare." - What we know, is a drop, we know nothig about the sea.
Scio me nihil scire. - I know, that I know nothig. (Socrates)

---

i've noticed one more thing. this structrures we procesing is in almost every case similar to native but complitly up side down :)) position at the end of computing. so :)) may the better way will be manualy changing the shape and then procesing again. am i only who noticed this :)))) ???

---