Posted by
timothy
on from the spin-the-wheel-of-fortune-dial dept.
GuyFromAccounting writes: "Tomorrow's Economist has a new article
describing research that shows that the number of human genes is more than twice
the estimate made a few months ago.The
article describes why it is so hard to estimate the number of genes."
God is a prima donna programmer
by
Zombie
·
· Score: 5
This makes me think of the item on "prima donna" programmers the other day. If we for a second assume that the preposterous be true, and there is a God and he created man, then
he's a spaghetti coder
he doesn't document any of his code
he supports the "embrace and extend" method of evolving functionality
he's clearly into this whole job security through obscurity thing
lots of people die because we can't figure out his code, so he doesn't give a rodent's behind about us
Actually... it looks like God works for Bill Gates!
Re:God is a prima donna programmer
by
joto
·
· Score: 5
Ah, no. But what you see, is only the binary. It's only after it is compiled with gcc (genetic code compiler) that it seems messy. It's like viewing a binary in texteditor. As I remember, God used to be especially picky on intendation and variable naming, and listed preconditions and postconditions for all his functions, as well as invariants for all of his datastructures.
IANAG (I am not a geneticist), but one argument I've heard about the so-called "meaningless drivel" in the DNA is that the current estimates for the distribution of gene sizes is heavily biased towards small genes. Apparently, the cost to sequence a gene scales with the size of the gene, and in the interest of actually finishing a sequencing project, geneticists have favored the smaller, and therefore cheaper genes. The estimates for amount of "meaningless drivel" simply take the estimate for average gene size multiplied by the estimate for the number of genes, and find that this falls short of the total number of base pairs. The problem, then, is that the average size of genes is severely underestimated.
-- ***
Work like a king, command like a slave, create like a dog.
Re:'Genes' vs. 'Instructions'
by
krmt
·
· Score: 5
blockquoth eraserbones
And yet, the DNA spans between genes are generally referred to as 'useless' or, in this case, 'meaningless drivel.' Am I missing something, or is this exactly where the good stuff is?
Not necessarily. You're caught in they typical/. trap of thinking of DNA as computer code. Granted, code is probably the best analogy we have, but it's still an analogy at best. You're very correct that living systems can operate by branching, looping, etc. just like programs. However, your mistake lies in looking too hard at the DNA and not hard enough at the whole system.
The fact is that DNA itself is pretty useless. It can't do anything without proteins, and it's the proteins that are actually acting on each other, on the DNA, and on the RNA. That said, it's probably the proteins that allow these functions in terms of things like splicing out introns (alternate splicing is a form of branching) and DNA replication via DNA polymerase and other helpers (a form of recursion).
While I personally don't believe that the intervening DNA sequences are complete garbage, I don't think they hold the processes you're looking for as much. I agree with the idea that they provide a lot of raw genetic material for evolution, and I also think they play a role in gene regulation by chromatin bundling and such.
However, the idea of DNA as a program is only a small part of the picture, and in reality even when we have the genome and the proteome, we're still going to have to figure out how everything works together. A living system is big and complex, with tons of parts we don't understand yet. It's going to be a fun time figuring it out.
A perfect example of this is cloning. When Dolly was cloned, the media went all-out in seing cloning as "Its here!!" Then, when problems started showing up, people started on the "cloning is horrible, everyone gets defects" bandwagon. When, in reality, neither were true.
The technique used in creating dolly was just awful. The scientist who worked on it has become a cloning opponent, largely due to seing his failures. On the contrary, other cloning researchers, like the ones that did the honolulu experiment with mice, have become its biggest proponents.
What was the dolly experiment like? Well, first off, they chose sheep because there is a very long period of time from when the egg is fertilized to when it divides for the first time; this length of time was assumed to (and likely does) help the odds of the cell thinking things were normal. However, his techniques were awful. After denucleating the ovum, it would go into a dormant state. So, he had to get the new nucleus to be in a dormant state, too. He did this by starving the cell for hours until it almost died. Then, instead of transplanting the nucleus directly, he applied a powerful electric shock through the solution the two cells were in, which usually caused them to merge, and act like a just fertilized egg. Now, I'm sure everyone reading this is just going, "this is going to cause serious problems". And, that it does. Dolly was a lucky sheep. Most of the embryos weren't near as lucky - that shock does a lot of damage to the cell (and starving the nucleus until it shuts down is bad too).
The honolulu experiments, for contrast, used mice. Mice are an even harder subject to deal with, because they have notoriously fast divisions in fertilized egg cells. But, they used them anyways, because they were not only convinced they could clone them, but they wanted to see results several cloned generations down the line. The technique used there involved, like before, denucleating the egg cell - but, doing this *right before* having the new nucleus implanted. To get a dormant nucleus, he took cells that were always (or almost always) dormant, such as certain nerve cells. He extracted the nucleus, and implanted it immediately into the egg cell. Then, he put the new egg cell inside a solution which prevents it from forming a polar body (and throwing away half the genes), so that it would think it was now fertilized. It then began to divide. They had a level of success that looks like, once the technique is perfected, will approach normal external fertilization techniques. Signs of premature aging didn't occur until 5 generations of clones - this due to the fact that genes slowly change over time, and usually for the worse; we basicly extended the mouse's lifespan beyond what its DNA was designed to handle (this could be fixed by making a "DNA backup" from the original mouse, and then reconstructing that DNA each time you want a clone - an important reason for fast DNA sequencers).
The media locked onto the first story. When problems started to arise, they completely switched gears, and made it look like all cloning is dangerous. Bad media! No cookie!:)
-= rei =-
P.S.: BTW, I have yet to see a project where they actually transfer the mitochondria - that's over 1% of most animals' DNA.
-- "Well, then fire it up and show me what this..." (sigh)... "coccoon can do."
The reason is that the genes don't exactly have a marker at each end to delineate them. The genes are to some extent a matter of definition. They're different lengths, they can sometimes be found in different places, and sometimes 'two genes' do exactly the same thing even though they read quite differently.
Some bacteria have been found with two or three sets of genes sort of ontop of each other- starting with different offsets. Its a bit like code that you can jump to at 0x2000 or 0x2001 and it does different, but useful things!
Anyway reverse engineering this lot will take a while...
Month-old stories used to get posted, but no we are getting links to stories that aren't even available yet! Maybe Taco finally got his time machine that was backordered at ThinkGeek.
Nothing is remotely firm yet...
by
update()
·
· Score: 4
The article (and the writeup here) makes it sounds like one presumably very accurate estimate has been supplanted by a very different, presumably very different, estimate. The reality is that identifying genes in raw sequence is very much a work in progress. At the annual Genome Sequencing meeting at Cold Spring Harbor in May, a bunch of groups presented different methods that resulted in widely divergent numbers. Everyone's numbers were increasing over the estimates of last year, though.
It'll sort itself out over the next couple of years as the sequence gets better assembled, more non-human sequence is available for comparison and the groups adopt one another's good ideas. In the meantime, it looks like a good PR person at Ohio State managed to make their findings seem more revolutionary than they are.
By the way, if you want to bet on the number, see the GeneSweep page. (Note that bets must be placed in person!) I put my $5 on 44,000 and change.
It's psychology. While most scientists tend to regard the first few studies on a topic as little more than a theory until its been confirmed by a few other people, the Media and general public tend to take these as absolute answers. "Well, then" they say "Thats taken care of". If the next study indicates the first one is wrong, then somethings changed. Thats news. If it simply confirms that yes, we are little more complex than your average nematode, thats not news. If suddenly we have way more, then the number of genes in the human genome has changed(well not really but you understand what I mean) thats news and shows up in the popular media. Slashdot works the same way. Taco and company aren't going to post 14 stories confirming the number of genes - it's not new, and (to most people) not exciting. But if it challenges current beliefs, its exciting and gets posted.
As for a health skepticism about science, you should be skeptical about science. Skepticism is (or should be) an integral part of science. Nothing should be taken for granted, nothing should be accepted as true until a good number of people have had a chance to kick it around in every way they can think of without finding a problem with the theory/study/whatever.
And when evidence does surface proving that the last theory was wrong, a new one should be created to fit the data and then that should be put under the microscope for flaws. This isn't religion. You don't take things on faith. Everything should be questioned and tested before acceptance.
-- Why?
URL of the article in Genome Biology
by
mcockerill
·
· Score: 4
The research article is available from the Genome Biology web site here.
Did Venter and Collins celebrate to early???
by
andres32a
·
· Score: 4
When Venter and Collins announced some months ago their "success" at sequencing the human genome and that according to their conclusions humans did not have much more genes that worms this did seem kind of odd to me.
If this new paper published by "Genome Biology" is right it could ridicule all past celebration... i mean... if it turns out that Venter only mapped 30 000 genes but there were actually 60 000 genes in humans, wouldnt that mean that were not even half way to a complete gene map for humans??
Moreover, most genes are split up into segments, known as exons, that are separated by long stretches of meaningless drivel.
Kind of like the comments on slashdot!
I suppose you could make it into a Katz joke too.
-PYves
Isn't that what the census is for?
by
Anonymous+DWord
·
· Score: 4
Personally I don't know any human Genes, although my friend had a hamster named Gene. It died, unfortunately. You'd think it would be harder to count the non-human Genes, but I guess that's far in the future. Didn't they count the mouse Genes a while back though?
-- "If he thinks he can hide and run from the United States and our allies, he's sorely mistaken." Bush on bin Laden
'Genes' vs. 'Instructions'
by
eraserbones
·
· Score: 4
Genes, I'll grant you, are the exciting bits of a chromosome, because they (generally) correspond to proteins that can be identified and detected. But, I'm not entirely clear on why they are the primary focus of genetic research.
We all seem pretty comfortable discussing DNA as though it were computer code, so let's follow that metaphor a little further. If I point at a big mess of C code (say, a console app) and ask 'What does this code do?' an amateur might be tempted to scan it for printfs, puts's, and other 'output signifiers.' But really, if that's all you look at, you don't have a clue what the actual funtion of the code is. All those boring scanf's, if/thens and operators are really important.
My rudimentary education in genetics has me convinced that DNA in a living cell has the ability (like C code) to switch, jump, branch, and (most importantly) operate recursively on its own resultant proteins. And yet, the DNA spans between genes are generally referred to as 'useless' or, in this case, 'meaningless drivel.' Am I missing something, or is this exactly where the good stuff is?
And, viewed from this angle, isn't counting genes as pointless as counting KLOCs?
Slashdot Mirror
Actually... it looks like God works for Bill Gates!
*** Work like a king, command like a slave, create like a dog.
The fact is that DNA itself is pretty useless. It can't do anything without proteins, and it's the proteins that are actually acting on each other, on the DNA, and on the RNA. That said, it's probably the proteins that allow these functions in terms of things like splicing out introns (alternate splicing is a form of branching) and DNA replication via DNA polymerase and other helpers (a form of recursion).
While I personally don't believe that the intervening DNA sequences are complete garbage, I don't think they hold the processes you're looking for as much. I agree with the idea that they provide a lot of raw genetic material for evolution, and I also think they play a role in gene regulation by chromatin bundling and such.
However, the idea of DNA as a program is only a small part of the picture, and in reality even when we have the genome and the proteome, we're still going to have to figure out how everything works together. A living system is big and complex, with tons of parts we don't understand yet. It's going to be a fun time figuring it out.
"I may not have morals, but I have standards."
"I may not have morals, but I have standards."
"Uh...yeah...turns out there are plenty more of them. If you want to see the data, though, you need to pay us royalties!"
Got Rhinos?
A perfect example of this is cloning. When Dolly was cloned, the media went all-out in seing cloning as "Its here!!" Then, when problems started showing up, people started on the "cloning is horrible, everyone gets defects" bandwagon. When, in reality, neither were true.
:)
The technique used in creating dolly was just awful. The scientist who worked on it has become a cloning opponent, largely due to seing his failures. On the contrary, other cloning researchers, like the ones that did the honolulu experiment with mice, have become its biggest proponents.
What was the dolly experiment like? Well, first off, they chose sheep because there is a very long period of time from when the egg is fertilized to when it divides for the first time; this length of time was assumed to (and likely does) help the odds of the cell thinking things were normal. However, his techniques were awful. After denucleating the ovum, it would go into a dormant state. So, he had to get the new nucleus to be in a dormant state, too. He did this by starving the cell for hours until it almost died. Then, instead of transplanting the nucleus directly, he applied a powerful electric shock through the solution the two cells were in, which usually caused them to merge, and act like a just fertilized egg. Now, I'm sure everyone reading this is just going, "this is going to cause serious problems". And, that it does. Dolly was a lucky sheep. Most of the embryos weren't near as lucky - that shock does a lot of damage to the cell (and starving the nucleus until it shuts down is bad too).
The honolulu experiments, for contrast, used mice. Mice are an even harder subject to deal with, because they have notoriously fast divisions in fertilized egg cells. But, they used them anyways, because they were not only convinced they could clone them, but they wanted to see results several cloned generations down the line. The technique used there involved, like before, denucleating the egg cell - but, doing this *right before* having the new nucleus implanted. To get a dormant nucleus, he took cells that were always (or almost always) dormant, such as certain nerve cells. He extracted the nucleus, and implanted it immediately into the egg cell. Then, he put the new egg cell inside a solution which prevents it from forming a polar body (and throwing away half the genes), so that it would think it was now fertilized. It then began to divide. They had a level of success that looks like, once the technique is perfected, will approach normal external fertilization techniques. Signs of premature aging didn't occur until 5 generations of clones - this due to the fact that genes slowly change over time, and usually for the worse; we basicly extended the mouse's lifespan beyond what its DNA was designed to handle (this could be fixed by making a "DNA backup" from the original mouse, and then reconstructing that DNA each time you want a clone - an important reason for fast DNA sequencers).
The media locked onto the first story. When problems started to arise, they completely switched gears, and made it look like all cloning is dangerous. Bad media! No cookie!
-= rei =-
P.S.: BTW, I have yet to see a project where they actually transfer the mitochondria - that's over 1% of most animals' DNA.
"Well, then fire it up and show me what this..." (sigh)
They didn't per chance say this because they found 46 chromosomes in a human instead of 23 did they?
icqqm [ICQ:11952102]
The reason is that the genes don't exactly have a marker at each end to delineate them. The genes are to some extent a matter of definition. They're different lengths, they can sometimes be found in different places, and sometimes 'two genes' do exactly the same thing even though they read quite differently.
Some bacteria have been found with two or three sets of genes sort of ontop of each other- starting with different offsets. Its a bit like code that you can jump to at 0x2000 or 0x2001 and it does different, but useful things!
Anyway reverse engineering this lot will take a while...
-WolfWithoutAClause
"Gravity is only a theory, not a fact!"Uninnovate - Only the finest in engineering.
It'll sort itself out over the next couple of years as the sequence gets better assembled, more non-human sequence is available for comparison and the groups adopt one another's good ideas. In the meantime, it looks like a good PR person at Ohio State managed to make their findings seem more revolutionary than they are.
By the way, if you want to bet on the number, see the GeneSweep page. (Note that bets must be placed in person!) I put my $5 on 44,000 and change.
Unsettling MOTD at my ISP.
It's psychology. While most scientists tend to regard the first few studies on a topic as little more than a theory until its been confirmed by a few other people, the Media and general public tend to take these as absolute answers. "Well, then" they say "Thats taken care of". If the next study indicates the first one is wrong, then somethings changed. Thats news. If it simply confirms that yes, we are little more complex than your average nematode, thats not news. If suddenly we have way more, then the number of genes in the human genome has changed(well not really but you understand what I mean) thats news and shows up in the popular media. Slashdot works the same way. Taco and company aren't going to post 14 stories confirming the number of genes - it's not new, and (to most people) not exciting. But if it challenges current beliefs, its exciting and gets posted.
As for a health skepticism about science, you should be skeptical about science. Skepticism is (or should be) an integral part of science. Nothing should be taken for granted, nothing should be accepted as true until a good number of people have had a chance to kick it around in every way they can think of without finding a problem with the theory/study/whatever. And when evidence does surface proving that the last theory was wrong, a new one should be created to fit the data and then that should be put under the microscope for flaws. This isn't religion. You don't take things on faith. Everything should be questioned and tested before acceptance.
Why?
The research article is available from the Genome Biology web site here.
When Venter and Collins announced some months ago their "success" at sequencing the human genome and that according to their conclusions humans did not have much more genes that worms this did seem kind of odd to me. If this new paper published by "Genome Biology" is right it could ridicule all past celebration... i mean... if it turns out that Venter only mapped 30 000 genes but there were actually 60 000 genes in humans, wouldnt that mean that were not even half way to a complete gene map for humans??
Kind of like the comments on slashdot!
I suppose you could make it into a Katz joke too.
-PYves
Personally I don't know any human Genes, although my friend had a hamster named Gene. It died, unfortunately. You'd think it would be harder to count the non-human Genes, but I guess that's far in the future. Didn't they count the mouse Genes a while back though?
"If he thinks he can hide and run from the United States and our allies, he's sorely mistaken." Bush on bin Laden
Genes, I'll grant you, are the exciting bits of a chromosome, because they (generally) correspond to proteins that can be identified and detected. But, I'm not entirely clear on why they are the primary focus of genetic research.
We all seem pretty comfortable discussing DNA as though it were computer code, so let's follow that metaphor a little further. If I point at a big mess of C code (say, a console app) and ask 'What does this code do?' an amateur might be tempted to scan it for printfs, puts's, and other 'output signifiers.' But really, if that's all you look at, you don't have a clue what the actual funtion of the code is. All those boring scanf's, if/thens and operators are really important.
My rudimentary education in genetics has me convinced that DNA in a living cell has the ability (like C code) to switch, jump, branch, and (most importantly) operate recursively on its own resultant proteins. And yet, the DNA spans between genes are generally referred to as 'useless' or, in this case, 'meaningless drivel.' Am I missing something, or is this exactly where the good stuff is?
And, viewed from this angle, isn't counting genes as pointless as counting KLOCs?