Why don't you use Molecular Dynamics in the full atom relax?

I would of thought, that with the Monte Carlo Minimization, you would have a random (averaged) possible structure (if I understand what I have read), that would lend itself to MD, in that, if the MD was taking too long you could chuck it and go to another ab-inito structure.

Let me say this another way for clarity:

MD should be able to predict a protein's structure reasonably accurately, if you can by-pass the majority of the prohibitably intensive MD calculations with Monte Carlo Minimization and just do the end bit (full atom relax) you should get the best of both worlds.

I think that this would give you a bigger depression in which to fall to the lowest energy, I'm thinking of a golf green that has a sunken dish shape around the hole.

If you already do this sorry, but it just occurred to me a while ago, and I'm not sure what "NMR structure prediction" is.

Edit: for formating

I'm thinking of a golf green that has a sunken dish shape around the hole.

I do that a lot as well...

---

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

nuclear magnetic resonance (NMR). So... you start with a really big magnet like the Mayo Clinic did... ok ok, I'm not positive on that either. But I think it works like an ultra high resolution MRI machine. Where the magentic field is used to excite atoms, this is detected and rendured on a computer to view structure non-invasively. For example, using an MRI to "see" a tumor, or joint damage.

Such a technique would avoid the cumbersome crystilization process used in X-ray crystallography.

---

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/

Why don't you use Molecular Dynamics in the full atom relax?

I would of thought, that with the Monte Carlo Minimization, you would have a random (averaged) possible structure (if I understand what I have read), that would lend itself to MD, in that, if the MD was taking too long you could chuck it and go to another ab-inito structure.

Let me say this another way for clarity:

MD should be able to predict a protein's structure reasonably accurately, if you can by-pass the majority of the prohibitably intensive MD calculations with Monte Carlo Minimization and just do the end bit (full atom relax) you should get the best of both worlds.

I think that this would give you a bigger depression in which to fall to the lowest energy, I'm thinking of a golf green that has a sunken dish shape around the hole.

If you already do this sorry, but it just occurred to me a while ago, and I'm not sure what "NMR structure prediction" is.

Edit: for formating

A good question. Monte Carlo Minimization appears to be significantly more powerful than MD for eary stage sampling as energy barriers that would trap MD trajectories can be effortlessly hopped over. It is possible that in the late stages MD would be useful as well, but in our admittedly limited tests we did not find this to be the case. even in the end game, there are discrete jumps, even for sidechains, that are relatively slow in MD. The power of MD is in simulating actual motions I think, more than conformational sampling.

---

A good question. Monte Carlo Minimization appears to be
significantly more powerful than MD for eary stage sampling as energy
barriers that would trap MD trajectories can be effortlessly hopped over.
It is possible that in the late stages MD would be useful as well, but in
our admittedly limited tests we did not find this to be the case. even in
the end game, there are discrete jumps, even for sidechains, that are
relatively slow in MD. The power of MD is in simulating actual motions I
think, more than conformational sampling.

Thanks for the explaination.

I think I need to read more on MD as I didn't realise that it could be
trapped, I thought it was just prohibitivly CPU intensive, though I
suppose/assume that could be one-in-the-same thing.

Feet1st "NMR structure prediction" is (I think) what R@H uses for
"full atom relax", I don't know how this ties in with NMR (which it must do somehow)
here

Incorporates NMR data into the basic Rosetta protocol for rapid structure determination at moderate to high resolution and speeds up the process of NMR structure prediction.

Edit: BBCode gave me problems, as usual :)

Feet1st "NMR structure prediction" is (I think) what R@H uses for
"full atom relax", I don't know how this ties in with NMR (which it must do somehow)
here

Incorporates NMR data into the basic Rosetta protocol for rapid structure determination at moderate to high resolution and speeds up the process of NMR structure prediction.

Actually, that's not quite correct. NMR stands for Nuclear Magnetic Resonance, which is the absorption of energy by atoms when held in a magnetic field of a certain strength. This phenomenon is exploited in NMR spectroscopy, which is technique that is used to experimentally determine the structure of proteins. NMR spectroscopy (usually abbreviated as just NMR) gives a series of constraints of which atoms are in close contact with each other. Structure prediction by NMR consists of elucidating these constraints and using them to make a structure representation that makes sense.

We can use these constraints from NMR in our full atom relax protocols in order to shrink the size of the conformational space. This means that if we can fix two different parts of the protein together, there are fewer possible conformations for the protein.

---

So... NMR gives you sort of a "fuzzy" picture? And then you use that, along with the amino acid sequence to form a more precise prediction on Rosetta?

Like if I give you a fuzzy picture of a house, you can SEE it's a 2-story, and where the windows and doors are, but you still have to figure out if it is brick or stucco. And whether the screen door is plexiglas or screens.

...and so knowing ahead of time that it's a 2-story, helps Rosetta narrow the search. And so that would be how Rosetta uses NMR information to aid in structure prediction... have I got it right?

---

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/