Structure Prediction Protein Design Biclustering + cMonkey Network Inference Publications          Links + Collaborations People                   Opportunities             Software + Code Teaching            

 

WCG Status Update

WCG Post

HPF1 -> HPF2 transition
[May 26, 2006 1:18:03 PM]

The HPF1 -> HPF2 transition AND the fact that HPF1 continues to run on the grid seems to be causing confusion. Since the grid is all about crunchers crunching for a cause I thought I'd repost a response and an explanation:

First of All:
All work done on HPF1 is real and will go into the database for biologist!!

that said I can see why you might wonder if we are killing time, So I'll explain.

we set out to do 100 genomes (human and other organisms that humans interact with like pathogens)

while we spent a year calculating/crunching the gene-finding guys got better at what they do. Also they sequenced a bunch of new mammal genomes (like elephant and latipus and cat) and they found new human genes (new genomes to compare to plus better techniques == new genes in what we thought was nonfunctional DNA).
(see this guy's lab to get a flavour of why new genes are being found:
http://genes.cs.wustl.edu/BrentLab/MB-Lab-index.html
http://genes.cs.wustl.edu/BrentLab/MB-Lab-research.htm
)

So this M. Brent guy gave us a bunch more genes and possible genes to run.

Also other scientists from around the world came up to us and said "what about this virus, what about rice, people eat rice and people are starving, what about this new relative of malaria, what about armadillo, which is the only other animal that gets this disease ..." on and on. All these genes are mucho important, so I begged IBM to let me keep running ...

Also the sequencing people are sequencing new organisms all the time and interesting / high importance genes are hitting the databases all the time.

So in the end we've finished our initial goal, but IBM has been nice enough to let me widen the focus to include more genes that are either new-human-genes OR new genes from high interest organisms.

All work done on HPF1 is real and will go into the database for biologist.
No work currently on the grid is benchmark OR code testing OR in anyway funny.

HPF2 will be different(more tightly focused) in that we'll be trying to focus on human secreted and malaria genes (and that will be that) (HPF2 tries so many structures per gene that it takes so long to run that even 5000 genes represents a whole project).

Thanks for crunching.

monkey monkey monkey monkey monkey monkey monkey monkey monkey monkey monkey monkey monkey