21 replies to this topic

[ihatesnow]

http://www.eurekaler...e-tao122204.php

Edited by maestro949, 08 June 2008 - 12:18 PM.
Enhanced Title from "Research News" to current

[Mind]

Thanks for the post ihatesnow. Interesting, as it relates to the Folding@home project and the F@H prize sponsored by Imminst.

[Athanasios]

LCS results are generated for a set of increasing RMSD [Root Mean Square Deviation] cutoffs--0.1, 0.2, and 0.5 nanometer; for GDT results, the cutoffs range from 0.05 to 1.0 nanometer. These cutoffs are chosen because of the level of certainty in knowing a protein structure with complete accuracy: With x-ray crystallography, the level of certainty is about 0.05 nanometer, and with NMR, it is within 0.10 to 0.15 nanometer. For computer modeling, it may vary more than 0.4 nanometer.

Nice, the accuracy is improving and is currently effective enough to compare a simulated protein with known proteins:

FR methods compare a specific sequence with all of the known folds in the PDB [Protein Data Bank] and estimate the probability of the unknown protein sequence having the same fold as that for a known sequence....FR-category results from the best metaservers were competitive with the best humans.

Edited by cnorwood, 08 June 2008 - 02:21 AM.

[niner]

This was a nice way to get up to date on protein fold prediction. I was at CASP1, but got into a different line of research not long after that. Here is the connection to F@H:

Protein molecules--only 3 to 10 nanometers across--can self-assemble quickly, some as fast as a millionth of a second. But this brief period is long for computers to simulate. Two difficulties arise in mimicking the protein-folding process with a computer. "First, the number of possible conformations a protein chain can adopt is too vast to analyze even with today's most powerful computer," says Fidelis. "Second, the estimates of molecular interactions that we use in simulations are simply not accurate enough to render a successful prediction."

It is generally true that most proteins fold too slowly to use molecular dynamics simulation as a folding tool. The CASP predictions are mainly done using various heuristics and the like. However, those proteins that can fold in a microsecond or less, if small, are amenable to attempts to fold using MD. F@H was talking about doing one of those quick-folding proteins some while ago. Does anyone know if it's been done and how it worked out?

The claim that the estimates of molecular interactions (a set of analytical equations and a parameterization that collectively are known as a force field) used in MD are not accurate enough for fold prediction is open to debate. F@H should (or may already have) shed some light on this.

One correction, if you read the article- a 300 residue protein with the standard 20 amino acids has 20 to the 300th possible sequences, not 20300(!) (It has been shown that the vast majority of possible sequences will not fold. You only have to look at the hydrophobicities of the residues to show this. This was elegant work by Ken Dill's group at UCSF.)

[maestro949]

I don't think it's possible to overstate the importance of this research. Nearly all 21st century medicine will be based on molecular therapies that emerge from this type of work. I suspect that many of our longer-term aging treatments will come from the genomic/proteomic.

I was at CASP1

Does it look like Casp 8 is much different than from when you attended? It would be interesting to get a perspective in how much has changed in this space.

I have a project due in 6 5 hours for finding motifs(paircoils, etc), calculating hydrophobicity, predicting what organelles proteins function in based on AA sequence alone. The algorithms for these are quite robust given the sparse amount of data researchers have had to work with over the past decade but the software tools and documentation for them are in a pretty sad state. It's the one thing that has surprised me the most in this informatics program I'm working through. There are gobs of tools, websites, journal articles, algorithms and theories but they are scattered all over the place and mining the relevant and useful information is fairly challenging for the neophyte.

[ihatesnow]

I don't think it's possible to overstate the importance of this research. Nearly all 21st century medicine will be based on molecular therapies that emerge from this type of work. I suspect that many of our longer-term aging treatments will come from the genomic/proteomic.

I was at CASP1

Does it look like Casp 8 is much different than from when you attended? It would be interesting to get a perspective in how much has changed in this space.

I have a project due in 6 5 hours for finding motifs(paircoils, etc), calculating hydrophobicity, predicting what organelles proteins function in based on AA sequence alone. The algorithms for these are quite robust given the sparse amount of data researchers have had to work with over the past decade but the software tools and documentation for them are in a pretty sad state. It's the one thing that has surprised me the most in this informatics program I'm working through. There are gobs of tools, websites, journal articles, algorithms and theories but they are scattered all over the place and mining the relevant and useful information is fairly challenging for the neophyte.

i suggest going over to the rosetta message board http://boinc.bakerlab.org/rosetta/ im sure someone will get back to you fast ,with useful suggestions

Edited by ihatesnow, 08 June 2008 - 09:11 PM.

[niner]

I was at CASP1

Does it look like Casp 8 is much different than from when you attended? It would be interesting to get a perspective in how much has changed in this space.

OMG, it is now huge! It's much more automated, which is essential, given the number of targets and number of participants. It's a very impressive scientific endeavor. Protein folding is still a hard problem. We're getting better, but there's a long way to go.

[Mind]

New insights into protein folding.

Stanford researchers have begun prying open the lid, literally, on the inner workings of chaperonin molecules by deducing the mechanism by which the lid operates on a barrel-shaped chaperonin called TRiC.

"Understanding how the lid opens and closes really helps us understand how everything moves inside the chaperonin," said Judith Frydman, associate professor of biology and one of two senior authors of a paper published online recently in Nature Structural & Molecular Biology.

"This is just the beginning, but now we can start to understand how the protein is pushed inside the cavity of the chaperonin and what this folding chamber looks like," Frydman said. Learning how a protein is manipulated inside TRiC while it is being folded is a crucial step in Frydman's larger plan.

"Our goal is to eventually exert control," she said. "If we could re-engineer the chaperonin to either fold better misfolded proteins or alternatively to remove them from circulation, then we could prevent those proteins from being harmful to cells."

Misfolded proteins have been implicated in a number of diseases, including some cancers, as well as ailments related to aging, such as Alzheimer's and Parkinson's diseases.

"Folding is one of the key steps for the health of the cell," Frydman said.

Imminst F@H prize.

[VictorBjoerk]

A little off topic but does progeria result from misfolded proteins affecting different structures in the body or is it a lack of the protein?

[ihatesnow]

http://norfolk.cs.wa...ails.cgi?id=449

[ihatesnow]

http://norfolk.cs.wa...ails.cgi?id=449

http://www.scienceda...80611161044.htm http://www.nature.co.../nsmb.1436.html

Edited by ihatesnow, 19 June 2008 - 05:36 PM.

[maestro949]

A little off topic but does progeria result from misfolded proteins affecting different structures in the body or is it a lack of the protein?

Progeria is caused by mutations in the Lamin A protein. Usually this means that the protein is not transcribed into a functional protein or not at all.

I found this by googling progeria and selecting the first link.

[dnamechanic]

A little off topic but does...

...I found this by googling progeria and selecting the first link.

There is a funny site about 'Google is your friend'. I initially posted it but decided to remove the link.

Edited by dnamechanic, 19 June 2008 - 10:49 PM.

[VictorBjoerk]

I know there are also many scientists on this board which may give even better answers in the field of bioscience and aging than google.....
However I've now found some information about this........

[Shepard]

I know there are also many scientists on this board which may give even better answers in the field of bioscience and aging than google.....

The problem is a lot of people don't want to waste time answering questions that can be easily answered with a little initiative.

[Mind]

New success in protein process simulation

In the new analysis, the researchers developed a computer simulation of the interaction of a single molecule of ADP with the carrier protein. Thanks to better simulation software and larger and more sophisticated computer arrays than were available for previous studies, this simulation tracked the process by which ADP is drawn into the carrier. It also showed how ADP orients itself as it travels to the site where it binds to the carrier.

In the simulation, the researchers observed for the first time that ADP disrupts several ionic bonds, called salt bridges, when it binds to the carrier protein. Breaking the salt bridges allows the protein to open -- in effect unlocking the door that otherwise blocks ADP's route into the mitochondrion.

The simulation included every atom of the carrier protein and ADP, as well as all of the membrane lipids and water molecules that make up their immediate environment -- more than 100,000 atoms in all. It tracked the interaction over a period of 0.1 microseconds, an order of magnitude longer than what had been possible before. "Until two years ago 10 nanoseconds was really pushing it," Tajkhorshid said. "Now we are reaching the sub-microsecond regime, and that's why we are seeing more biologically relevant events in our simulations."

More computing power=better simulations=more revelations. This is why efforts such a folding@home and the F@H Prize are so critical.

[Mind]

Billionaire David E. Shaw creates specialized supercomputer for protein folding simulation.

The new supercomputer is distinguished from other molecular dynamics computing tools like I.B.M.’s BlueGene/L supercomputer and the Stanford Folding@home distributed computing project in that the machine is designed to simulate a very narrow set of problems on biological processes that take place over a millisecond or longer. Molecular simulations are now done as a series of tiny intervals that may be as short as a femtosecond, one billionth of one millionth of a second, and may last no longer than a microsecond, or one millionth of a second.

Vijay Pande, the director of the Folding@home project at Stanford, which computes its problems by distributing portions of the work over the Internet to graphics processors in the desktop computers of volunteers, said his group had talked about sharing research with the Shaw lab. He noted that both approaches had pros and cons and were potentially complementary.

[maestro949]

Billionaire David E. Shaw creates specialized supercomputer for protein folding simulation.

It's awesome to see billionaires pouring their resources into the advancement of technology, particularly that which will provide exponential gains in medical research some day.

[Mind]

Water is "Designer" Fluid that Helps Proteins Change Their Shape.

"While it is well known that water plays an important role in the folding process, we usually only look at the motion of the protein," said Gruebele, who also is the director of the U. of I.'s Center for Biophysics and Computational Biology, and a researcher at the Beckman Institute. "This is the first time we've been able to look at the motion of water molecules during the folding process."

Using a technique called terahertz absorption spectroscopy, Gruebele and his collaborator Martina Havenith at the Ruhr-University Bochum studied the motions of a protein on a picosecond time scale (a picosecond is 1 trillionth of a second).

The technique, which uses ultrashort laser pulses, also allowed the researchers to study the motions of nearby water molecules as the protein folded into its native state.

The researchers present their findings in a paper published July 23 in the online version of the chemistry journal Angewandte Chemie.

Terahertz spectroscopy provides a window on protein-water rearrangements during the folding process, such as breaking protein-water-hydrogen bonds and replacing them with protein-protein-hydrogen bonds, Gruebele said. The remaking of hydrogen bonds helps organize the structure of a protein.

In tests on ubiquitin, a common protein in cells, the researchers found that water molecules bound to the protein changed to a native-type arrangement much faster than the protein. The water motion helped establish the correct configuration, making it much easier for the protein to fold.

[ihatesnow]

http://boinc.bakerla...ead.php?id=1177

[Ghostrider]

http://boinc.bakerla...ead.php?id=1177

D.E. Shaw recruits heavily through Stanford. Not surprised to hear talks of collaboration.

[ihatesnow]

A little off topic but does progeria result from misfolded proteins affecting different structures in the body or is it a lack of the protein?

Progeria is caused by mutations in the Lamin A protein. Usually this means that the protein is not transcribed into a functional protein or not at all.

I found this by googling progeria and selecting the first link.

you can ask the scientists on the Rosetta message board, proteins is what the project is all about http://boinc.bakerlab.org/rosetta/