Genome@home

[CTho9305]

What exactly does it do? Folding@home is supposed to take DNA/(m)RNA and find the protein shape, right? Seti/rc5 are obvious. But I don't understad what the genome@home client does.

 

[Poof]

Actually, Folding@Home takes known proteins (small ones to start) and tries different algorithms to simulate the steps necessary for a particular fold of the protein. The folded area is usually an active binding site for that protein, which allows it to perform certain functions in the body. If it misfolds (like what happens with the protein Prion, which inturn causes Mad Cow Disease), then you can have troubles. If scientists can determine what steps a protein goes through to fold, then they might be able to determine what might cause a mis-fold.

Genome@Home, using a known database of proteins, has you actually "design" the genes that would actually be responsible for creating that protein (that's what genes do, by the way... ). Scientists know that several genes can be responsible for the formation of a single protein, however they don't yet know all the genes that could make that protein (you might recall that the Human Genome Project was pretty much announced as being completed only recently). The project actually has you take a known protein and go through a reverse-engineering process to determine what the gene would look like that would have created that specific protein. The way the client is configured, you could actually get a WU (which consists of the info for a certain number and type of amino acids), that can be re-run over and over if you like (a sortof "built-in caching&quot, but each time it's run, it has a different "seed" that will trigger the creation of a variation of the gene from the first time you run it. Thus each time you run it on that one WU, you'll get a different gene at the end. Normally, you'd just upload your completed gene data and download a new one. The re-running of the same WU is good for those who are on a dialup or have problems connecting.

The results from Genome@Home will become part of a Pulic Domain database for scientific use.

Hope this helps?

 

[Assimilator1]

Cool
I think its a little clearer to me too

 

[CTho9305]

Ok, so genome is taking amino acid sequences and plugging in the codons to get the RNA/DNA that made it up?

 

[Poof]

Basically I think that's what's happening (what I call "reverse-engineering&quot. When you do that, you would expect to get known genes and I would think, "unknown" ones (or ones that don't really exist, at least as part of human chromsomes).

Here's some info that might help illustrate it better. It's kinda cool that they're using existing data (eg., X-ray crystallography) to provide info to process the WU.