1000 Genomes Project Releases Pilot Genome Data

eldavojohn writes "Three pilot projects have been completed for the 1000 Genomes Project and as a result, the pilot data has been released. This makes the data of nearly 700 people available for analysis via FTP (Americas mirror, European mirror). Dr. Eric D. Green of the National Human Genome Research Institute said, 'The 1000 Genomes project has a simple goal: peer more deeply into the genetic variations of the human genome to understand the genetic contribution to common human diseases. I am excited about the progress being made on this resource for use by scientists around the world and look forward to seeing what we learn from the next stage of the project.' There's not a whole lot of information on their site about this data, but the repositories have many readme files explaining the data layout."

Re:Would you guys stop messing with the godamn cod

[Yvan256]

I tried to mod you "funny", but the moderation system is indeed non-working. Also, after hitting "reply" I was sent to a new page with a messed-up "Reply to this" button, instead of simply having a comment textarea below your post on the same page.

Someone's messing around with the live website.

Re:Stupid question

[jcmurray]

Although they offer FTP access to the genomic data--including population, alignment and sequences (traces, calls, etc.)--the NCBI has hosted the files with a README [nih.gov] and guide [nih.gov] (aspera_transfer_guide.pdf) about Aspera's "fasp technology" [asperasoft.com] that the NCBI claims to incorporate automated checksum verification for both casual downloaders, via a browser plugin, and bulk downloaders, via a cross-platform command-line application. Aspera is new to me; they claim to have some throughput (bandwidth) advantages as well.

Nevertheless, the sequence data files embed MD5 checksums directly, per NCBI documentation [nih.gov], which I would expect bulk downloaders to take advantage of independent of any third-party "technology."

Re:4% genes of 2% total DNA of 700 people

[Turbio]

Hey, I work with junk regions! (satellite DNA) And I completely agree with you. But I see that centering on the variability of those few regions rather than sequencing a second complete genome will probably be better for health-care research. The project's title is completely misleading. That's for sure.

Re:Double edged sword

[Turbio]

Most genes are used to produce proteins. Samples of all the proteins present inside a living human cell are exposed on the cell's membrane, as part of the immune system machinery. (See Mayor Histocompatibility Complex http://en.wikipedia.org/wiki/Major_histocompatibility_complex [wikipedia.org]) Those proteins can be targeted using homing peptides (think of it as a specific antibody) on a liposome (http://en.wikipedia.org/wiki/Liposome [wikipedia.org]). That liposome can contain anything from drugs to viral RNA. Right now, you can make a solution with liposomes (or polymersomes), that when injected only kills the person if he/she has a certain trait. But I gess that is not a weapon of mass destruction... Anyway, making it a disease is another story. It could involve modifying the HIV virus, which already has a lipid bilayer just like a liposome. And then you have the problem of keeping that disease specific to the target population. And viral genomes tend to mutate at a very fast rate.

Re:How much data is this?

[karlnyberg]

Approximately 3.5TB. That's the math.