Slashdot Mirror
Function of 80% of the Human Genome Charted
ananyo writes "In what is likely to be a historic moment in science, ENCODE, the Encyclopedia of DNA Elements, has published 30 papers in Nature, Genome Research and Genome Biology today, assigning some sort of function to roughly 80% of the genome, including more than 70,000 'promoter' regions — the sites, just upstream of genes, where proteins bind to control gene expression — and nearly 400,000 'enhancer' regions that regulate expression of distant genes. The project was designed to pick up where the Human Genome Project left off. Although that massive effort revealed the blueprint of human biology, it quickly became clear that the instruction manual for reading the blueprint was sketchy at best. Researchers could identify in its 3 billion letters many of the regions that code for proteins, but those make up little more than 1% of the genome, contained in around 20,000 genes. ENCODE, which started in 2003, aims to catalog the 'functional' DNA sequences between genes, learn when and in which cells they are active and trace their effects on how the genome is packaged, regulated and read. Nature has set up an ENCODE site with an explorer, that groups the papers by topic, and collects all the papers, which are available free."
Just happened to hear an NPR interview on the way back to the office. The researcher described most of the 80% as regulating the expression of the protein codes. Brace yourself Slashdot: he called it the 'operating system'.
My God, it's Full of Source!
OUTSIDE_IP=$(dig +short my.ip @outsideip.net)
Wait for the lawsuits when they come across a gene some company holds a patent for. They "invented" it you know.
In what is likely to be a historic moment in science
I'm not knocking the achievement, but wouldn't it be a more truly historic moment when they've nailed down the function of 100% of the genome? Where was the big celebration when they got to 64.576%?
systemd is Roko's Basilisk.
Imagine you were given a slightly buggy 3.2Gig piece of software used to run a group of factories and to managing communications between them. You were told to debug this piece of software, but you had no source code, only the machine code only. You could of course observe it's behaviour and set up situations in various factories to see how it behaved. To complicate matters further, you realised that the factories were all performing different functions and making different things but they were all running the same software, but it wasn't immediately obvious why they behaved differently. This is pretty much a huge oversimplification of the challenge that faces modern genetics. We have the assembler language, but we're still not sure entirely what bits are coding sections, which are data sections and which sections are marked up to act as configurations sections. It's a huge task and I love it.
It's time you wised up (it's RIGHT THERE). Junk DNA is non-encoding. Meaning it doesn't get used to make proteins. This was thought to be the entire purpose of DNA, so junk DNA was like commented out code or dead code. But we learned that this genes have other uses. Like some DNA affects the trans-coding of nearby genes. And it appears some non-encoding DNA is still selected for, so it probably has some unknown function.
But nature most certainly deals in junk. And there is DNA in you right now that's non-encoding, non-functional, and not selected for. IE, it's junk. It just happens to be the random uninitialized value or a previous mismatch of old code that was cut long ago. Deal with it.
Uh, it was called into question over a decade ago. No scientific paper or journal article ever referred to it as "junk" DNA, it was called non-encoding regions and it was understood fairly early on that at least some of the area held a regulatory function. What wasn't realized until the human genome project concluded was just how little of the genome was encoding and how massively important the regulatory regions were (the human body creates a heck of a lot more than 10,000 different proteins so the regulatory regions must be more than simple on/off switches but must also have the ability to affect structural changes in the protein encoding sequence),
There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
The sky will be the limit.
The understanding of how DNA works, ( and correspondingly, how to hack it ) is the ultimate reverse-engineering accomplishment.
Life is a textbook, full of worked examples. We are at the stage we realize there is an alphabet, the letters mean something, and have the definition of a few words. Kindergarten stuff.
If we play our cards right, and don't spend all our resources fighting amongst ourselves, the future is incredibly bright. We have worked examples of damn near everything we need... photosynthesis ( solar powered CO2 sequestration and energy storage ) for starters. We have bioluminescence, electric eels, and all sorts of sensor examples.
I figure we have been given a huge shipment of arduinos with all sorts of accessories, and we have now figured out how to make the light blink.
We don't know how its wired yet, how the compiler works, and just now figuring out some of what makes the hardware work.
If our society will value knowledge above greed and accounting, if there is anything limiting our potential, I have yet to see it. However if greed and accounting is all we know, we will soon run into all sorts of limits, imposed only by our inability to adapt. First of these will be exhaustion of the earth's fossil fuels, followed by food and water famines. We will be like the chick that hatched, but failed to scratch, find food, and thrive, living off the energy stored in the egg - until it is depleted.
The earth is our egg.
I value highly the knowledge our species acquires. It is our survival.
"Prove all things; hold fast that which is good." [KJV: I Thessalonians 5:21]
Strictly speaking, it is still universally believed that 'many' non-coding genes are historical junk with no current function, and experiments on organisms with much simpler DNA than ours bears this out (in short, they will scramble suspected 'useless' sections of DNA and look for changes in function; impossible to do in humans, but relatively simple for c. elegans). If you are talking about the antiquated view of _all_ non-coding genes as "junk DNA:" This is not the usual belief, at least not in the molecular biology/bioinformatics community. You may just be seeing the release of tons of ENCODE work today, but this research (and a vast amount of related research) has been ongoing for several years. It is well known, and accepted, that many non-coding genes play important regulatory rules in DNA transcription.
Am I misinterpreting this, or is the usual belief that many genes are obsolete sequences that have no current function being called into question?
I don't think any serious molecular biologist ever thought the majority of DNA had no function, just as no neuroscientist ever believed that we only use 10% of our brain, but that's precisely the sort of sound bite that, when uttered in a press release somewhere, echos around the public consciousness forever and never dies because it provides a conveniently sciency premise for the next batch of rebooted superhero origin stories. The distinction is that before this study, we knew non-coding DNA was involved in regulation but not to what extent; i.e. there were plenty of specific anecdotal findings but nothing this systematic and large scale.
As for the significance of this sort of work, yes, it exactly like release day for a major software package, it's an anticipatory excitement and not a "we finally found the Higgs Boson after decades of searching" type of achievement. Molecular biologists and geneticists everywhere can now do a simple web search see how this affects the system they are working on without needing to perpetually beg the labs that possess the specialized high-throughput instrumentation to do a one-off experiment just for their favorite gene. . .
> It's also got fairly good licensing terms - I mean the OS can be replicated (it is billions of times - once for ever cell), Yeah, right. But use the wrong process for replication, and you'll end up paying for it for almost two decades! How's that for vendor lock-in?
Now none of that rules out the existence of regions of DNA that do little or nothing and can be mutated without consequence. There are definitely such regions in the genome.
i hope that cleared something up
Uhm, no. "Junk" DNA is really junk. It consists of repeating regions, transposons, inactive and decaying retroviruses, etc. It has no direct functions, and its indirect usability is questionable. For example, pufferfish has almost no junk DNA while some plants (corn, for example) have lots of it without any visible effects on mutation rates or cell viability.
Actually, most molecular biologists KNOW that the majority of _eukaryotic_ DNA has no function. It's junk, deal with it. Fairly small parts of non-coding DNA perform useful functions: gene expression regulation (less than 0.1% of total DNA), mechanic 'handles' for cell replication machinery (about 5% of DNA), various RNA enzymes (less than 1%), etc. But most of it is still junk.
Nope. We HAVE confirmed it. Like, we've analyzed them it turns out that about 33% of the genome consists of simple repeating sequences (SINEs and LINEs). Yep, your genome consists mostly of copy&pasted crap. Oh, then there are about 8% of viral remains and lots of useless LTRs.
They serve no function at all and can be safely deleted without any effect on viability. In fact, pufferfish genome has almost none of them and it has no discernible effect at all. On the converse, plants have in general much more repeated fragments (with great variation between species) and also no discernible effects on their cell viability. But what do I know? After all, my company just produces a novel sequencing method, so it's not like I know what I'm speaking about.
Gee after decades of Darwinists proclaiming junk DNA to be key evidence of Darwinian evolution,
It still is. Also, the impact of non-encoding DNA has been suspected for over a decade. These papers just kind of hammer it home. Furthermore, there is STILL junk DNA. As in non-encoding, non-selected, no-purpose sections of your DNA that are remnants from long ago. It's dead code, commented out code, or simply gibberish. (There's just less of it than we previously thought)
(before really knowing what those non-coding segments of DNA were for)
There are noncoding segments that appear to be selected for which we still don't understand
I wonder what they'll say now.
We're still right. And all of this still helps us prove our case.
There are more robust theories of evolution than the Darwinian model
Well, sure. Darwin laid out the basic premise that species change over time and become separate species. That's still very much true. He got some details wrong, which later models improved upon, so I guess that Darwin's specific model is out of date. I mean, he didn't even know DNA existed. But "Darwinian", "Darwinist", and "Darwiniest" are kinda loaded terms only used by fundie creationists. I'd avoid them if I were you.
but they all suggest design to one extent or another.
Bwahahahahah! sure kiddo, if you say so.
the other theories are honest and don't assert a naturalistic origin for things like DNA.
I'm actually intrigued by this. What would be the origin of DNA be other than nature? Anything that isn't man-made is natural. Kinda by definition.