Yes the simple sequence control structures (centromeres and telomeres) are unresolved. There are also some large scale duplications > 1MB that are unresolved. This is as good as it gets with all current technology until we can magically peer at DNA and see the base pairs. However, saying "they are even lacking in the Celera version of the genome" is unintentionally funny cause of course they aren't in the Celera version, the Celera version leaves out all of the hard parts and many of the easy parts already, why would you expect the hardest parts to be resolved in the Celera (unfinished) version? If you NEED to use sequence in areas which has not been assembled or has not been sequenced then perhaps you might consider doing the work yourself for those areas as the funding and interest exist to work on them, instead of relying on someone else to generate the data for you. The HGP was meant to accelerate disease research and it has done so and will continue to do so for most diseases. Do you rememeber when you needed to spend 5 years mapping a trait loci, another 3 years sequencing (poorly) the local region, another 3 years mapping a finer interval, and then being able to construct a linkage SNP study on a population to attempt to identify the disease mutation? Before the sequence was available identifing and cloning a simple Mendalian disease trait was someone's and many grad students life work! You think you could always look up a candidate interval and identify all of the genes in that interval so you could postulate about the disease mechanism? No, 10 years ago you would sitting there with a disease phenotype and no other information, heck, before the HGP you didn't even have a decent human linkage map- these were built as part of the HGP. Before those ~15 years ago, you couldn't even map a disease interval. Anyway I'm all for finished and complete sequence as I push this viewpoint for a living, but there are limits for a reason. So congrats to Sanger and UWASH for finally publishing Chrom1, we've been waiting for a while:)
Most of the sequence in the trace archive is from large genome sequencing projects where we intentionally oversample genomes to 6-10x or more. Also each trace while averaging 864 characters only contains about 600 bp of real data. What this means is that the trace archive currently represents about 60 billion basepairs of unique sequence. In human genome numbers thats 20 genomes worth of data. The approximate output in traces right now is max about 30 million per month so the 1 billion traces represents 33 months at the current output of the world's sequencing centers. As the trace archive was started after the human genome project, most of the traces related to the human genome aren't in the repository. It is not difficult to produce massive amounts of sequencing data- the trick is in turning it into something that one can use to answer scientific questions.
The findings are being reported in Nature, which has exceedingly high standards. There is absolutely no reason to make such accusations.
While not disagreeing with the responses to the initial posters trollish remarks, to suggest that a scientific fact is a fact because it was published in Nature "which has exceedingly high standards" is a shall we say misinformed opinion about how scientific literature works. Each article if deemed appropriate for publication in Nature is reviewed by 2 or 3 scientists in the field. If they say the work merits publication, then Nature will probably publish it. The mistake here is assuming that Nature is unbiased in what they would like to publish- and let me tell you a new Homo species is something Nature will really want to publish. It is high profile, generates news and revenues, and convinces people like this poster that Nature is above reproach. Truthfully, articles like this get second chances and for this article there is probably an extremely limited reviewer pool leading to less of a thorough review than other scientific topics may receive. Publishing scientific journals is as much of a business as any other and while we have may have great trust in Nature, I predict there were still Nature editors wetting their pants over this one!
How can anyone call this serious doubts when the article is posted on spectator.org?
Please read their article in the same exciting issue on the deadly effects of not letting people use DDT which says "You heard it talked about today on Rush Limbaugh..."
Anti-Science? Nope. Not at all. I hear about all of homo saps technological breakthroughs on Rush.
Why post this trash?
Is it just to annoy people like me who might have some work to do on the human genome and are instead posting to slashdot?
Or so we can all laugh about how the author of this piece tells us that there is no code, and then proceeds to tell us that there is a meta-code?
It's nice that the genome has been "sequenced in its entirety" and is presently
undergoing "error checking" which
should "continue for the next year".
Last time i checked at ncbi the genome was at 30.4% finished. and the rough draft assembly is in 148307 pieces according to the golden path.
And of course the finished target for the human genome is three years from now!
First, this thread is almost as interesting to me as listening to a bunch of fortran programmers debate the pluses and minuses of JAVA. That said, Doubletwist is pangea or was anyway. They have, like so many companies added a.com in their name to attempt to increase their popularity. As for identifying most of the human genes, it's fairly trivial. In this case, dt says they id'ed 65,000. This is how it is done so all you slashdot readers can do it in your spare CPU cycles:
1. you suck off all the cDNA data from NCBI. 2. you suck off all the HTGS and Finished Human Sequence. 2a.(only if you are really good) You create your own virtual assembly of all the human data and use RH maps to assemble whole chromosomes. (but you don't have to) 3. you look for a lot of gapped aligments with the cDNAs on the human sequence.
Slashdot Mirror
Yes the simple sequence control structures (centromeres and telomeres) are unresolved. There are also some large scale duplications > 1MB that are unresolved. This is as good as it gets with all current technology until we can magically peer at DNA and see the base pairs. However, saying "they are even lacking in the Celera version of the genome" is unintentionally funny cause of course they aren't in the Celera version, the Celera version leaves out all of the hard parts and many of the easy parts already, why would you expect the hardest parts to be resolved in the Celera (unfinished) version? If you NEED to use sequence in areas which has not been assembled or has not been sequenced then perhaps you might consider doing the work yourself for those areas as the funding and interest exist to work on them, instead of relying on someone else to generate the data for you. The HGP was meant to accelerate disease research and it has done so and will continue to do so for most diseases. Do you rememeber when you needed to spend 5 years mapping a trait loci, another 3 years sequencing (poorly) the local region, another 3 years mapping a finer interval, and then being able to construct a linkage SNP study on a population to attempt to identify the disease mutation? Before the sequence was available identifing and cloning a simple Mendalian disease trait was someone's and many grad students life work! You think you could always look up a candidate interval and identify all of the genes in that interval so you could postulate about the disease mechanism? No, 10 years ago you would sitting there with a disease phenotype and no other information, heck, before the HGP you didn't even have a decent human linkage map- these were built as part of the HGP. Before those ~15 years ago, you couldn't even map a disease interval. Anyway I'm all for finished and complete sequence as I push this viewpoint for a living, but there are limits for a reason. So congrats to Sanger and UWASH for finally publishing Chrom1, we've been waiting for a while :)
Most of the sequence in the trace archive is from large genome sequencing projects where we intentionally oversample genomes to 6-10x or more. Also each trace while averaging 864 characters only contains about 600 bp of real data. What this means is that the trace archive currently represents about 60 billion basepairs of unique sequence. In human genome numbers thats 20 genomes worth of data. The approximate output in traces right now is max about 30 million per month so the 1 billion traces represents 33 months at the current output of the world's sequencing centers. As the trace archive was started after the human genome project, most of the traces related to the human genome aren't in the repository. It is not difficult to produce massive amounts of sequencing data- the trick is in turning it into something that one can use to answer scientific questions.
The findings are being reported in Nature, which has exceedingly high standards. There is absolutely no reason to make such accusations.
While not disagreeing with the responses to the initial posters trollish remarks, to suggest that a scientific fact is a fact because it was published in Nature "which has exceedingly high standards" is a shall we say misinformed opinion about how scientific literature works. Each article if deemed appropriate for publication in Nature is reviewed by 2 or 3 scientists in the field. If they say the work merits publication, then Nature will probably publish it. The mistake here is assuming that Nature is unbiased in what they would like to publish- and let me tell you a new Homo species is something Nature will really want to publish. It is high profile, generates news and revenues, and convinces people like this poster that Nature is above reproach. Truthfully, articles like this get second chances and for this article there is probably an extremely limited reviewer pool leading to less of a thorough review than other scientific topics may receive. Publishing scientific journals is as much of a business as any other and while we have may have great trust in Nature, I predict there were still Nature editors wetting their pants over this one!
How can anyone call this serious doubts when the article is posted on spectator.org?
Please read their article in the same exciting issue on the deadly effects of not letting people use DDT
which says "You heard it talked about today on Rush Limbaugh..."
Anti-Science? Nope. Not at all.
I hear about all of homo saps technological breakthroughs on Rush.
Why post this trash?
Is it just to annoy people like me who might have some work to do on the human genome
and are instead posting to slashdot?
Or so we can all laugh about how the author of this piece tells us that there is no code,
and then proceeds to tell us that there is a meta-code?
PLEASE!!!!!!
It's nice that the genome has been "sequenced in its entirety" and is presently undergoing "error checking" which should "continue for the next year".
Last time i checked at ncbi the genome was at 30.4% finished. and the rough draft assembly is in 148307 pieces according to the golden path.
And of course the finished target for the human genome is three years from now!
First, this thread is almost as interesting to me as listening to a bunch of fortran programmers debate the pluses and minuses of JAVA. That said, Doubletwist is pangea or was anyway. They have, like so many companies added a .com in their name to attempt to increase their popularity. As for identifying most of the human genes, it's fairly trivial. In this case, dt says they id'ed 65,000. This is how it is done so all you slashdot readers can do it in your spare CPU cycles:
1. you suck off all the cDNA data from NCBI.
2. you suck off all the HTGS and Finished Human Sequence.
2a.(only if you are really good) You create your own virtual assembly of all the human data and use RH maps to assemble whole chromosomes. (but you don't have to)
3. you look for a lot of gapped aligments with the cDNAs on the human sequence.
Eureka! You found the genes!