The Gene Is Having an Identity Crisis

gollum123 writes "New large-scale studies of DNA are causing a rethinking of the very nature of genes. A typical gene is no longer conceived of as a single chunk of DNA encoding a single protein. It turns out, for example, that several different proteins may be produced from a single stretch of DNA. Most of the molecules produced from DNA may not even be proteins, but rather RNA. The familiar double helix of DNA no longer has a monopoly on heredity: other molecules clinging to DNA can produce striking differences between two organisms with the same genes — and those molecules can be inherited along with DNA. Scientists have been working on exploring the 98% of the genome not identified as the protein-coding region. One of the biggest of these projects is an effort called the Encyclopedia of DNA Elements, or 'Encode.' And its analysis of only 1% of the genome reveals the genome to be full of genes that are deeply weird, at least by the traditional standard of what a gene is supposed to be and do. The Encode team estimates that the average protein-coding region produces 5.7 different transcripts. Different kinds of cells appear to produce different transcripts from the same gene. And it gets even weirder. Our DNA is studded with millions of proteins and other molecules, which determine which genes can produce transcripts and which cannot. New cells inherit those molecules along with DNA. In other words, heredity can flow through a second channel."

Memory RNA

A thread on DNA and its relationship to RNA gives me a chance to ask: what ever happened to the idea that memory was encoded in RNA? In 1970s science fiction novels like Niven's A World out ot Time [amazon.com] , you had people learning new skills through the injection of RNA. When did it become clear that RNA had nothing to do with memory?

Re:Memory RNA

The answer lies in the RTFW (Read The F'in Wikipedia) article about Memory RNA [wikipedia.org]:

One experiment that was purported to show a chemical basis for memory involved training planaria to solve an extremely simple "maze", then grinding them up and feeding them to untrained planaria to see if they would be able to learn more quickly. The experiment seemed to show such an effect, but it was later determined that the original planaria had left chemical tracks inside the maze itself that were not properly cleaned away before the next set of planaria were run.

It's not a complete explanation, but it implies that pathfinding behavior(e.g. getting out of a maze) had much more to do with following a chemical "bread crumb" trail than using memory alone.

Re:Memory RNA

>>The way I understand it is that a large part of evolutionary theory ASSUMES that memory can't be inherited.

Maybe not memory per se, but certain phenomena have demonstrated Lamarckian style inheritance.

Mm, DNA can be methylated, which modifies its behavior. A fat pregnant mother will methylate the genes in the fetus, resulting in a kid much more genetically prone to being fat. Experiments with dutch prisoners of war during WWII showed that even when raised under similar conditions, kids from mothers who ate more when they were pregnant were much more prone to obesity.

There's also effects like http://en.wikipedia.org/wiki/Genomic_imprinting [wikipedia.org] which modify an offspring's genome on the fly between generation and generation.

Re:Memory RNA

Well, I've not learned about RNA holding memory in any of my classes, and even Wikipedia has little to say [wikipedia.org] on the subject.

I'd venture a guess that it's not correct (simply not enough evidence supporting it, but that has not yet been ruled out either [nih.gov].

The bottom line is that we do not yet fully understand memory, in much the same way that we do not fully understand synapse formation in the brain. We should just wait and see before jumping to any conclusions (and maybe write a grant proposal or two along the way).

Re:

A very simple answer is that RNA degrades *extremely* rapidly. Injecting RNA could feasibly give a short change in phenotype, but it is hard to imagine that RNA would be able to encode something as long-lasting as memory.

Re:Memory RNA

RNA is a copy of DNA created by an enzyme called RNA Polymerase [wikipedia.org]. All RNA Polymerase does is a simple copy. There is no mechanism for creating "new" RNA that contains data that is not already present in your genes. That is, your body does not contain any device that can write memory information to RNA strands.

Re:Memory RNA

Raising both a two year old and a seven month old, there appears to be precious little instinct. Maybe dancing to music.

Re:Memory RNA

I think what you are referring to is more closely related to instinct than memory. Instinct is related to sets of behaviors that are performed naturally, learned. These are to some extent controlled by genes, but can be overridden by learning. I.E. genes will encode for certain basic neural pathways to be formed, but the brain's development will then be left to augment or diminish that pathway's strength.

Memory is an entirely different system, in which patterns simulating previous stimuli are stored and available to be replayed or compared against. Calling the effect of instinct "ancestral memory" or "genetic memory" is at best a poetic interpretation, at worst a logical flaw similar to Lamarkian Evolution [wikipedia.org] wherein giraffes have long necks because their ancestors stretched out trying to graze from tall shrubs, then trees, rather than the Darwinian idea that giraffes have long necks because short necked giraffes did not live to reproduce as well as long necked ones.

So, what we REALLY need is . . .

. . . A Human Genome Interpreter Project.

Re:So, what we REALLY need is . . .

How long does it take to compute 42?

Oh wait, we already have 42. What was the question again?

I Knew It

Not only does God code in machine language, but it is all spaghetti. Thats probably why eventually malfunction and die.

Re:I Knew It

Thats probably why eventually malfunction and die.

Some faster than others, apparently.

Re:I Knew It

You have it backwards. God doesn't code in spaghetti machine language. The FSM itself coded God.

Re:I Knew It

Well, what do you expect when you knock off a major project in under a week?

How is any of this new?

Epigenetics [wikipedia.org]
RNA Splincing [wikipedia.org]
siRNA [wikipedia.org]

Re:How is any of this new?

http://www.pbs.org/wgbh/nova/sciencenow/3411/02.html [pbs.org]

I'm waiting longer before hitting submit.

Re:How is any of this new?

Some of it is new, but none of it is surprising to anyone who has been paying attention for, I dunno, the past decade.

It's been clear since the mid-90's when we learned that there were only 44k "coding" genes but at least ten times that many proteins that more was going on than simple templating.

Things like methylation of double-stranded DNA have been known to be important in oncogenesis for at least ten years. miRNAs have been considered important for five years or more. Other conserved non-coding regions have been known for almost as long or longer.

This story is going to be like that "green" story that reports breathlessly every year or so that some company has instituted green policies because they save money! Just like Interface did fifteen years ago, and hundreds of others have done since.

The interesting thing is that these persistently "new" stories give us a measure of how slowly what can fairly be called "general knowledge" changes. Based on evidence from the "green" story we can expect to be hearing the AMAZING NEWS that there's more to genes than template coding until at least 2020.

Following the details as we learn them is fascinating. Being told that an uncontroversial fact we've known for a decade or more is "news" is very, very irritating.

Inteligent Design

Who would have thought God coded DNA using Perl...

...or not

Don't take my word for it, take the word of a cellular biologist [scienceblogs.com].

How many genes does it take to invent a light bulb

I recall people freaking out when the human genome project revealed that Humans only have about 30,000 genes rather than the previous estimate of 150K.

It always seemed to me that measuring Human complexity based on the number of our genes is a little like judging a book by the number of words it contains. It completely ignores the fact that words have Meaning.

Poetry is both the most compact and the most subtle form of written expression.

This latest finding suggests to me that something similar applies to our genetic heritage.

-S

 

Re:How many genes does it take to invent a light b

I recall people freaking out when the human genome project revealed that Humans only have about 30,000 genes rather than the previous estimate of 150K.

It always seemed to me that measuring Human complexity based on the number of our genes is a little like judging a book by the number of words it contains. It completely ignores the fact that words have Meaning.

Uh, I remember when they discovered that too, and I don't recall any scientists "freaking out" because the low number of genes implied we had low "complexity". Instead, I remember them being very excited, because they already knew there are far more than 30,000 proteins generated from our DNA, meaning that the 1:1 gene:protein mapping theory had to be wrong, and the mechanism was far more complicated than previously thought.

This sounds to me like a continuation of the line of inquiry opened by that discovery years ago, where now they're gaining a better idea of how the genes really code for proteins. With the extremely interesting aspect that some of this is controlled by things not part of the DNA itself, yet which can still be inherited.

To (ab)use your analogy, if the human body is a work of literature then proteins are the words, and genes are characters. The number of words hasn't changed, it's just that before we thought the language was like Chinese, where a single character mapped to a single word. Now we realize it's more like English, where the interactions between characters create different words. Oh and now we've discovered that there's also punctuation like apostrophes and hyphens which can significantly alter the meaning of the resulting words.

DMCA Takedown

RE: NOTICE OF COPYRIGHT INFRINGEMENT

Ladies and Gentlemen:

We act on behalf of God (the "Owner").

As required under Sections 512(c)(3) and 512(d)(3) of the Digital Millennium Copyright Act (17 U.S.C. ??512(c)(3) and 512(d)(3)), we are instructed to place you on notice that:

1. The Owner is the exclusive owner of the copyrights in and to the human DNA, RNA, and all other information contained therein

2. Decryption of aforementioned encrypted information constitutes an unlawful cicumvention of encryption technology

Please cease and desist from further decryption of stated copyright information and publication of previously acquired DNA information.

Re:Surprise, surprise!

It has been known for a long time that junk DNA wasn't junk. However its one of those catchy memes that has persisted it the general public far longer then it was believed to be true.

Re:Surprise, surprise!

Computer memory is actually a pretty good analogy for this: the "unused" DNA is not reachable by any "pointers" and thus wasn't important when eucaryote evolution began. Some of these areas are obviously non-coding ever-repeating nonsense sequences, others appear to be random information - exactly like unused RAM in a computer system. Of course, nothing in there is really random, it's just a product of whatever process happened to use the areas before.

Here's the catch, however. Just like a programmer who develops against an ancient API with a lot of well-known bugs and workarounds, some transcription mechanisms actually began to rely on the presence of the "useless" areas in order to work.

It's all a huge mess, the deeper you look, the less elegant it all becomes. For example, epigenetic mechanisms modify the meaning of DNA code depending on different contexts, as the article mentioned. But that's still not the whole picture. In order to create a protein, DNA is first transcribed into RNA, which then in turn gets executed in order to assemble the protein. However, the intermediate RNA information is modified beyond recognition before it is used. Then, after the protein is finally assembled, it too can be modified extensively. All of these steps are hopelessly interwoven, and they use zillions of chemical messenger signals in order to tweak an manipulate each other.

Genetics really is the worst spaghetti code project ever and I assume that more advanced (=complex) organisms really paint themselves into an evolutionary corner eventually, because the whole system - while beautifully specialized - is essentially becoming more and more difficult to alter meaningfully when radical change needs to happen.

Re:So, as a car analogy...

Think of it this way- if your protein-coding genes are the blueprints for a car, then epigenetics are the blueprints, operating procedures, and logistics for a mass production automobile factory. By reading your genes, you can find out the kinds of proteins that make you up. Similarly, car blueprints tell you how to make a car. A car, just one car. However, your cells are not putting out handbuilt cars. It's a modern Toyota factory going on in there, with continuous production and assembly. It's a marvel of mass production, with transcription, splicing, translation, post-translational modification, and relocation to the site of use all going on in multiple sites constantly. Production has to be carefully coordinated to make sure you have the right amounts of the right proteins delivered at the right times.

Epigenetics is the guy at the factory who knows how many cars to build this month, and the guy who makes sure that 10,000 cars have 10,000 steering wheels available to put in. Epigenetics is the guy who tells the line to hold up on building doors, because there's a surplus of doors in the warehouse already and we should use those first. Epigenetics is not the stuff you are made of, but rather a system of production control of that stuff.