"‘The gene does not lead,’ she says. ‘It follows.’"

December 3, 2013 7:13 AM

The selfish gene is one of the most successful science metaphors ever invented. Unfortunately, it’s wrong.
posted by overeducated_alligator (79 comments total)

At first I thought, well, of course, everyone knows this. What makes a cell in your retina different from a skeletal muscle cell, different from a muscle cell in your heart, different from a white blood cell, different from a skin cell? Surely this is obvious?

But he's right, it really doesn't get the attention it deserves, but I think a large part of that is because it's not well enough understood yet. The evolution and development group of researchers had some great popular science books about this a while ago, but they were all fairly small and not super exciting examples that could capture imagination like the Selfish Gene's writing could. It's a more complicated story, and requires more explanation.

And a lot of this comes down to the definition of "gene" which is a point of debate. Various definitions, such as protein coding region, or transcribed RNA element, are common, but neither matches the original definition of "unit of heritability". Individual DNA bases in a protein coding gene can be the units, meaning that a single protein coding gene can have many "genes". And far more important that the protein coding genes in the evolution of species are the regulatory elements that control the "reading", as it's termed in this piece. And this was known by King et al back in the 1970s!

I think a lot of the knee-jerk criticism of the ENCODE project (and perhaps I'm alone in thinking that that criticism was incorrect and unjust) was due to people not yet wanting to accept this post-Dawkins view of cellular biology and genetics.
posted by Llama-Lime at 7:41 AM on December 3, 2013 [4 favorites]

And a lot of this comes down to the definition of "gene" which is a point of debate. Various definitions, such as protein coding region, or transcribed RNA element, are common, but neither matches the original definition of "unit of heritability".

Yep, and this original definition is question-begging—because what if heritability doesn't have "units"?
posted by overeducated_alligator at 7:45 AM on December 3, 2013 [8 favorites]

but what does this sort of statement even mean:

"To repeat: you are 80 per cent cow."

80% in what sense?
Like if I take two books both written in the same language containing exactly the same words (but in a different order) are they 100% the same?

In a sense, for me the problem with this "different reading" metaphor is that it suggest that the readings are infinite and not in some sense "genetically determined".

The Grasshopper / Locust thing just shows that you can genetically encode for a kind of "dual-system". But the duality is itself genetically determined. Its like a Optimus Prime can be either Humanoid or Truck but he still has only two possibilities.
posted by mary8nne at 7:53 AM on December 3, 2013 [7 favorites]

The article seems to suggest that just because something can change, implies it can't be structurally determined.

But a structure can be architecturally determined / (specified by genetic means) to have a set of possible "configurations". But the number of configurations are still limited by the actual genetic specification.
posted by mary8nne at 8:02 AM on December 3, 2013 [2 favorites]

This kinda gives Darwin more credence, no? An advantage - a mutation- arises that the gene "locks in" and perpetuates. Am I missing something?
posted by Benny Andajetz at 8:03 AM on December 3, 2013 [1 favorite]

A fascinating article that shows once again that we're just beginning to understand how little we know; thanks for posting it.
posted by languagehat at 8:08 AM on December 3, 2013 [3 favorites]

Link bait, good article sure, but still link bait.
posted by Cosine at 8:09 AM on December 3, 2013 [2 favorites]

Like if I take two books both written in the same language containing exactly the same words (but in a different order) are they 100% the same?

It would be more like the same sentences in about the same place in the book, and you're sort of contradicting your own argument. Dawkins is the one placing the emphasis on individual words/sentences as the only thing worth thinking about. It's not that sentences aren't important or don't convey meaning, but there's a lot between the lines, between a book and a culture, an author and an audience, all sorts of contextual reasons sentences are reused, an impossibly large myriad of other things worth thinking about.

The article hit the nail-head, hard, in saying that the selfish gene is a selfish meme, an idea that's crowded out other important ideas which have valuable, complementary insights into evolution and life.

See also: the debate on group selection, also starring Richard Dawkins as the man prepared to die on the hill that his and only his idea is worth half a shit.
posted by crayz at 8:12 AM on December 3, 2013

Why do you say that, cosine?
posted by Jacob Knitig at 8:13 AM on December 3, 2013

This kinda gives Darwin more credence, no? An advantage - a mutation- arises that the gene "locks in" and perpetuates. Am I missing something?

Not that I can find. They're not suggesting that there is any new method for passing traits on.

Basically they're saying that the ability to re-read your own genome in multiple ways is an evolutionary advantage. Which it is.

(dear god, I'm agreeing with Dawkins about something. although this actually is an area he's qualified to open his yap about.)
posted by Tell Me No Lies at 8:14 AM on December 3, 2013 [2 favorites]

80% in what sense?

80% in the sense that if you define "gene" as a protein coding gene, i.e. a section of DNA that is translated into a sequence of amino acids, about 80% of them are the "same" as in humans. Each protein is a sequence of typically 100-500 amino acids, and there are 20 different amino acids, for example:

MEESQAELNVEPPLSQETFSDLWNLLPENNLLSSELSAPVDDLLPYTDVA
TWLDECPNEAPQMPEPSAPAAPPPATPAPATSWPLSSFVPSQKTYPGNYG
FRLGFLQSGTAKSVTCTYSPSLNKLFCQLAKTCPVQLWVDSPPPPGTRVR
AMAIYKKLEHMTEVVRRCPHHERSSDYSDGLAPPQHLIRVEGNLRAEYLD
DRNTFRHSVVVPYESPEIDSECTTIHYNFMCNSSCMGGMNRRPILTIITL
EDSCGNLLGRNSFEVRVCACPGRDRRTEEENLRKKGQSCPEPPPRSTKRA
LPTNTSSSPQPKKKPLDGEYFTLQIRGFKRYEMFRELNDALELKDALDGR
EPGESRAHSSHLKSKKRPSPSCHKKPMLKREGPDSD

Humans and cows each have about ~22,000 protein coding genes, and when they sequenced the cow, they could find about 80% of human genes that match up with a cow gene, by aligning these amino acid sequences together. It turns out that things either match up quite well, or not at all, for the most part, so it's easy to place the dividing line at 80%. The order of each of these elements is not the same, but there are occasionally regions where the order of the genes matches up (which is called synteny).

This is what most people call a "gene" in molecular biology. However, none of these "genes" matches up at all with Dawkin's definition of gene. Dawkin's definition shifts to whatever it needs to be, and does not appear to have a physical basis that is coherent.
posted by Llama-Lime at 8:14 AM on December 3, 2013 [11 favorites]

You could even say flexibility is the essence of being a primate.

As many multitudes of Garrus instances are well aware.
posted by Slackermagee at 8:14 AM on December 3, 2013 [4 favorites]

I think the article is unfair to Dawkins - it says that he "resists" the gene expression theory "by accomodating it". That doesn't sound much like resistance - that sounds like someone fitting new data into existing data - you know, like scientists are supposed to do.
posted by corb at 8:15 AM on December 3, 2013 [4 favorites]

I'm afraid this article doesn't really nail down a working definition of how gene expression "changes." It seems to me that it's better to say that novel gene expression happens.

Now as far as I can tell, the critique of the selfish gene theory is that since organisms develop in environments, there is no canonical phenotype, because there is no canonical environment. Cool organisms' genomes respond to a wide variety of stimuli in adaptive ways, and cool gene expression -- potentially novel in a historical sense, but not novel in a genetic sense -- can separate populations without changes in the genome.

So it's just another axis of understanding? That's fair.

But from the article:

As you do, certain genes ramp up expression to build more muscle and fire the muscles more quickly. You get faster. You’re becoming a different animal.

No. I'm really not. I'm just responding to changes in the environment, albeit not by the conventionally expected behavioral or physiological channels.

This is interesting, and I agree that there's a marketing challenge to getting the idea across, but doing some work on the phrase "gene expression changes," which sounds a tiny bit handwavy (but totally isn't!) might be a good place to start. From my perspective, it's largely a question of vocabulary. When I say "genetic," instantly you think of heritability and such.

"Organisms can respond to environmental stimuli in a number of ways: physiological responses, behavioral changes, epigenetic response..." Something like that.
posted by lumensimus at 8:19 AM on December 3, 2013 [3 favorites]

The evolution and development group of researchers had some great popular science books about this a while ago

Is there a current "best", book on this subject? One that's approachable by a non-bio person?
posted by sammyo at 8:19 AM on December 3, 2013 [1 favorite]

So, essentially, our DNA is a program, and gene expression is the command line arguments?
posted by ymgve at 8:19 AM on December 3, 2013 [3 favorites]

Provided your program writes its own command line arguments based on environmental factors, and we don't know all the command lines it can produce... Sounds like it.
posted by lumensimus at 8:24 AM on December 3, 2013 [3 favorites]

I LOVE sending things like this to my Kindle. Brilliant! Juicy stuff here, and remember, in 5 years, we will probably be confident that the author got many things wrong but that's science, baby!
posted by Mister_A at 8:24 AM on December 3, 2013 [1 favorite]

>> but what does this sort of statement even mean:

>"To repeat: you are 80 per cent cow."

It means it wouldn't kill you to just try the salad, or some pasta, or even just a chicken fillet sandwich, instead of always getting a god-damned hamburger every time we eat out.
posted by benito.strauss at 8:26 AM on December 3, 2013 [16 favorites]

i honestly don't understand the difficulty in parsing that statement, but I am a biologist by training (lapsed now).
posted by Mister_A at 8:27 AM on December 3, 2013

"In other words, the gene-centric model survives because simplicity is a hugely advantageous trait for an idea to possess. People will select a simple idea over a complex idea almost every time."

Suck it, Occam. Lightweight.
posted by klarck at 8:36 AM on December 3, 2013 [2 favorites]

My thoughts after reading this is that gene expression is more nuanced and more accurate way of explaining how genes result in the incredible variety of life, rather than proving that the selfish gene idea is wrong, wrong, wrong. The emphasis on gene expression as a paradigm shift might be a result of the journalist, although the quotes from the proponents don't really support that.

The main problem is that article doesn't really give a good example of how it's different. My (not particularly knowledgable) interpretation is that genes will respond different in different environments and the different ways they respond (plasticity) can be selected for in the standard way. My intro bio class really stressed how useful mutations are very rare and it's much more common for genes to be used in different ways (which seems like gene expression?).
posted by raeka at 8:40 AM on December 3, 2013 [1 favorite]

corb, what I find really damning is that by shifting his definition, a "gene" simply becomes a convenient term for a multitude of distinct human concepts, and many physical phenomenon. By becoming so broad, it loses meaning, and the definition is no longer a useful tool for communication or thinking about modern biology. And all the debates that Dawkins fought so heard about gene selection vs. group selection seem rather incredible now, because the definition of 'gene' may as well extend to the entire group if it's going to be extended as far as he has already extended it. In the past he was quite unrelenting in a specific definition of gene and selection, and how selection only happens on genes, but it seems unintelligible now (though it seemed a rather pointless debate to me when I first read it anyway). In many ways, the term "gene" has been stolen out from under the classical geneticists. It's been a really long time since I read the Selfish Gene, so others please correct me if I'm unfairly treating Dawkins arguments there, or in his later books that I did not read.

For example, one of the examples of a selfish gene is the phenotype of altruism. However this phenotype is not the result of a single bit of DNA that gets selected. We now know that this "gene" must be the result of a huge team of bits of DNA, all expressed at the right time in many different cells. This includes both protein coding genes that form mechanical parts that organize the right neural pathways, and the regulatory elements that control the "expression" of a protein coding gene, or whether a particular mechanical part gets included in the cellular machine. So there are hundreds to thousands of bits of DNA, and hundreds of mechanical parts, spread across millions-billions-trillions of cells, all timed correctly in their expression, that consist of this the altruism gene. Each of which would be considered a gene itself to modern researchers, and each of the sub-genes of the altruism gene does a multitude of other things besides the altruism phenotype, so the sub-gene can be selected as part of a different team of higher level genes.

I think the view that Dawkins synthesized is a useful view for the time, but it's becoming inadequate as we understand more and more biology and have better views into the workings of cells. Kind of like how Newtonian physics works well for things we encounter everyday at human scales, but if we want to look at really really small things we need quantum physics, or if we want to look at really really distant/fast/etc things we need relativity. At a high conceptual level, e.g. breeding animals or doing early genetics, this broad definition of gene is quite useful and should not be forgotten. But if we want to understand the physical basis for complex traits like altruism we're going to need to abandon it and dig deeper.
posted by Llama-Lime at 8:52 AM on December 3, 2013 [6 favorites]

Why do you say that, cosine?


Because the headline and intro are all "Selfish Gene is Wrong, Total Bullshit, Get Rid of It, Must Go Away"

Then a page in comes the "well, selfish gene is mostly right, we are just fine tuning it."

That is link bait.
posted by Cosine at 8:57 AM on December 3, 2013 [7 favorites]

I think the article is unfair to Dawkins

Because it is, the writer's opinion of Dawkins comes through pretty clearly, and too me immaturely. Colours an otherwise very interesting piece.
posted by Cosine at 8:59 AM on December 3, 2013 [1 favorite]

Hey, my dissertation is pretty much all about applying this kind of thinking! How it fits in can be found at the tail end of this rap about my PhD.

There is a deeper divide within genetics that this piece ignores, its already pretty wonderfully in depth for a popular piece, but illuminating it I think will help it all make a lot more sense.

In the 1920s and '30s there was a mass movement of out of work physicists who, having suddenly run out of things to do when we figured out to much of classical physics, came to biology. They brought with them a mechanistic view of how the universe works that they used to cause massive transformations in how we understand and interact with biology, and incidentally most used phages. One of the most influential of these scientific interlopers was a charismatic guy named Max Delbrück who quickly reasoned that, if we were ever going to understand how life works, we would need to start with the simplest organism possible and work our way up. He isolated seven bacteriophages against E. coli B, originally just his lab strain, and named them in a series T1 (previously) through T7. The central idea was that he and his growing number of colleagues¹ would focus on truly understanding how these phages worked and use that knowledge to generalize to Escherichia coli, then the mouse, and then us. An essential component of this was the "Phage Treaty" among researchers in the field, which Delbrück organized in order to limit the number of model phage and hosts so that folks could meaningfully compare results. What came out of their original focus on these phages, in many respects encapsulated in Erwin Schrödinger's What is life?, has shed light on so much as to truly redefine our self-understanding as a species, much less medicine:

The Luria–Delbrück experiment elegantly demonstrated that in bacteria genetic mutations arise in the absence of selection, rather than being a response to selection. These are mutations both to the genetic elements that could be plausibly described as genes and the many that can't. Evolutionary biology has made so much more sense ever since.

The Hershey–Chase experiment showed once and for all that nucleic acids were in fact the heritable molecule.

The two guys who discovered the model for the structure of the B form double helix were phage folk. Incidentally they published it in easily the snarkiest, most badass, and likely most important published scientific paper ever, written as an accessible single page. The structure of DNA, and its relationship to function that they discovered, is true for all of life.

Most of the central dogma, was also figured out using phage, from most of the functions of RNA to the triplicate nature of codons

Delbrück turned out to be absolutely right to start simple, and his branch of Biophysics turned into molecular genetics and then split off into the molecular biology, protein biology, molecular physiology, bioengineering, as well as genomics and the various other –omics that we know today. It did however immediately bump into and conflict with the then already well established field of classical genetics. Where for fifty years people has already been studying genetics in a way much more analogous to stamp collecting than physics. These classical geneticists had no idea how what they studied worked, and many couldn't care less. They were mostly concerned with building complex mathematical models that demanded the simplistic unit of inheritance we still call a gene and using those models to map traits to what we now understand to be loci on chromosomes using linked inheritance. While they built what we now know to be astonishingly accurate maps in mice and fruit flies, and did do a significant amount of important theoretical work - the study of biology never really took off until the physicists came in to start looking at how things actually worked and why by tinkering with the parts.

Dawkins very much comes from the remnants of the classical genetics perspective, building models that require simplistic understandings of the underlying biology rather than trying to reduce the mechanisms to their simplest parts, that has been constantly cut off at the knees over the last 70 years with newer models based on deeper empirical understandings of the mechanics of biology. The gene centric understanding of evolution was already outdated as a genuinely useful concept when Dawkins swiped the idea from Hamilton and Williams, indeed we now know that the Central Dogma understanding of molecular genetics that it relies on is to simplistic to work much if not most of the time. Heritable traits do not even usually exist as distinct coding sequences with clear start and end sites that turn into proteins that always have the same function. They exist as regulatory elements that expand or contract the expression of functional elements either entirely or in response to specific stimuli, those functional elements are not always proteins, they exist as elements that have both regulatory and functional effects, they exist as a function of the location elements are found in relation to others, and there are even important traits that are difficult to describe as heritable at all when the molecular basis is properly understood. The reality is a hell of a lot more wibbly wobbly than a gene centric model can support.

The Selfish Gene concept, nowadays, is something that can easily be a lot more pernicious and destructive than wrong, its stupid and useless. It is less analogous to Plato's geocentric model, which was very wrong but at least is still useful to this day for things like pointing telescopes, than it is to Tycho Brahe's geocentric model, which was pretty correct but entirely useless. Where Brahe, motivated by his own understanding of the Bible, worked desperately to incorporate his data into a model that would keep the Earth at the center. He used his data and Kepler's mathematics to construct something that was many orders more complex and that ended up being more or less just the same thing - just spinning around on an Earth centered axis for no empirically discernible reason. Sure it could still be used to predict the motions of the celestial bodies in more or less the same way, at least once the mathematical errors were fixed, and it neatly incorporated all of the contemporary known reliable data - but the inherent distortion would have made, for example, the Law of Universal Gravitation impossible to discern from their equivalent to Kepler's laws.

Those few population geneticists, stamp collectors who do statistics, who are still enamored with the selfish gene concept just need to catch up with the times and learn more about what has happened in the many descendants of molecular genetics over the last thirty years. There may be little hope for Dawkins himself, being pretty entrenchedly enamored with himself, but the rest of us can surely move on.

¹Frank Stahl famously wrote: "The Phage Church, as we were sometimes called, was led by the Trinity of Delbrück, Luria, and Hershey. Delbrück's status as founder and his ex-cathedra manner made him the pope, of course, and Luria was the hard-working, socially sensitive priest-confessor. And Al (Hershey) was the saint."
posted by Blasdelb at 9:15 AM on December 3, 2013 [55 favorites]

"To repeat: you are 80 per cent cow."

80% in what sense?

Imagine if you have a recipe for a cake, exactly 100 words long. Print a copy. Now go back and change the word 'flour' to 'cement.' Print a copy. Make two cakes.

These are two very different outcomes, but they are 99% the same in that their recipes are 99% identical. The genome is the instruction book or the recipe for building an organism. Two or more can have a very high amount of overlap and still produce dramatically different outcomes.
posted by K.P. at 9:15 AM on December 3, 2013 [6 favorites]

... in gene expression so widespread and powerful they alter not just the hopper’s behaviour but its appearance and form. Legs and wings shrink. Subtle camo colouring turns conspicuously garish. The brain grows to manage the animal’s newly complicated social world...

Ok, so how does this work in real life? I'm not so much worried about jumping, but how do we about growing the brain?

Since I've missed that trick, I'm missing just how the mechanism of "gene expression" actually works? Are there different codes? Are there "if then else" elements to the genetic code? Is the "code" very clever that it functions differently in freezing water vs boiling water? Did I miss something in the article or is there a whole field of "expression studies"?
posted by sammyo at 9:19 AM on December 3, 2013 [2 favorites]

The Selfish Gene concept is one of the more pernicious stupid ideas in evolution, in my opinion. It's not a thing that real evolutionary scientists talk about, in my experience – much of it is wrong, and what's right is tautological. Nevertheless Richard Dawkins (the most over-rated man in biology, perhaps?) wrote a very provocative book promoting the idea and it's gained a lot of traction with educated laypeople. It's one of those situations where the author claims to have come up with some exciting new concept that will create a whole new paradigm in their field, but which is actually just an imperfect analogy describing a bunch of things that everyone in the field already knows.

Genes don't have intent. Yes, they can be acted on by natural selection, and yes they can sometimes be selected for even if they reduce the fitness of the individual in which they reside. It's still just a stochastic process mediated by outside events though, a process which is simple and well understood (at a broad level), and which has many interesting emergent properties but which has nothing to do with selfishness (or any other volition) on the part of genes. Individuals can be selfish, and behavioral evolution predicts that selfishness will be selected for under some circumstances, but extending that by analogy to genes is inappropriate and misleading.
posted by Scientist at 9:22 AM on December 3, 2013 [4 favorites]

"This is what most people call a "gene" in molecular biology. However, none of these "genes" matches up at all with Dawkin's definition of gene. Dawkin's definition shifts to whatever it needs to be, and does not appear to have a physical basis that is coherent."

Seriously, Dawkin's "gene" is very much like a True Scotsman¹ its impossible to pin down to any real molecular basis, you know basis in fact, because the definition is constantly changing to suit whatever point he is trying to make at the time. Its not really Dawkin's failing, its one he has inherited from a long tradition of classical genetics that is pretty analogous to using the terminology of people who studied stamps before it was clear what a postal service was. We now know better and only really talk about "genes" very carefully with caveats, but not everyone has moved on.

¹"Imagine Hamish McDonald, a Scotsman, sitting down with his Glasgow Morning Herald and seeing an article about how the "Brighton Sex Maniac Strikes Again". Hamish is shocked and declares that "No Scotsman would do such a thing". The next day he sits down to read his Glasgow Morning Herald again; and, this time, finds an article about an Aberdeen man whose brutal actions make the Brighton sex maniac seem almost gentlemanly. This fact shows that Hamish was wrong in his opinion but is he going to admit this? Not likely. This time he says, "No true Scotsman would do such a thing"
posted by Blasdelb at 9:23 AM on December 3, 2013 [2 favorites]

IANAS, but I am familiar with the Selfish Gene idea. Until I read this discussion I always thought it was more like a thought experiment or a koan, I didn’t realize that people took it as The Way Things Are.

Interesting article.
posted by bongo_x at 9:34 AM on December 3, 2013 [2 favorites]

There's the genome, representing an organism's genetic information, and then there's the proteome, representing the proteins currently being expressed by a given cell/tissue.

The genome is hard-coded. It does not change (we'll ignore methylation, mutation, etc). The proteome varies constantly based on the cell's environment, activities, interactions, etc. Which, incidentally, is why you should laugh at any article that reports that X chemical has been shown to alter gene expression. Everything alters gene expression. Eating a PBJ alters gene expression. Wearing a hat alters gene expression. Your mom alters gene expression.

It's not limited to dramatic, system-wide changes, like the transformation of a grasshopper to a locust. It's completely fundamental to normal, everyday cellular activity.

Note, however, that the ways in which gene expression change in response to the environment is itself encoded in the genome in one way or another. Like a game of chess, all of the possible configurations of the pieces are encoded in the structure of the rules, but the actual configuration changes from moment to moment, via complex interactions occurring between the board and the players, and the players with each other.
posted by dephlogisticated at 9:46 AM on December 3, 2013 [1 favorite]

Since I've missed that trick, I'm missing just how the mechanism of "gene expression" actually works? Are there different codes? Are there "if then else" elements to the genetic code? Is the "code" very clever that it functions differently in freezing water vs boiling water? Did I miss something in the article or is there a whole field of "expression studies"?

There is more than just a field of expression studies, there are fields of gene interactions, systems biology, etc. etc. that try to bring this all together. At a very abstract level, you can think of the expression regulatory code as a very complex, deeply nested set of if/else statements about boolean expression states of other proteins, as well as proteins that signal other bits of information (heat shock, chemical sensing, etc.). More realistically on the physical level, it's the combination of a bunch of non-linear cascades of signal information, and more simple thermodynamic models of how well two complex biomolecules bond to each other.

Responding to your earlier question, the evo-devo book that was most popular is probably Endless Forms Most Beautiful which provides a view of expression from the perspective of the evolution of body (and brain) development). Larry Moran also has a critique of Carroll's hype that I don't disagree with, but it shouldn't stop you from reading Carroll's book as I think Carroll provides a fairly clear introduction to more about how the cell works and gives it a nice evolutionary and multi-cellular perspective that you don't get from most molecular biology primers. Also, there are lots of good pictures in the book, so I'd be suspicious of the Kindle version, as I've never encountered a Kindle book that had legible figures.

The mechanism of gene expression is the copying of DNA into RNA by polymerase, and translation from RNA into protein by ribosomes, the so-called central dogma. Part of the way that gene expression is regulated is through the initiation of copying. This requires a number of factors. First, the general region of DNA must usually be in the euchromatic state, or "open" state rather than in the heterochromatin state. This means that the DNA is not densely packed up for storage, but is open for use, and all the other parts of the polymerase can recognize the DNA sequence that indicates the start of the protein coding gene. The polymerase itself is made of proteins (which are in turn expressed from DNA in the genome). And there are transcription factors that recognize the start of the gene, which are proteins, and which must also be expressed from the rest of the genome. So there's a significant amount of what a programmer would call "variable state" in which proteins are expressed, and what parts of the DNA are euchromatin vs. heterochromatin.

This is an extremely simplistic and cartoonish view of what's going on, but I believe it's a fair representation of our current understanding, and there are teams of experts and textbooks on most parts of this. I spend my time figuring out how to merge expression data with information about how proteins signal to each other, and how proteins turn on and off expression of particular parts of DNA, along with novel mutations that occur in individual proteins that result in a different amino acid sequence.
posted by Llama-Lime at 9:47 AM on December 3, 2013 [10 favorites]

Very interesting, thanks!
posted by jeffburdges at 10:00 AM on December 3, 2013

"Since I've missed that trick, I'm missing just how the mechanism of "gene expression" actually works? Are there different codes? Are there "if then else" elements to the genetic code? Is the "code" very clever that it functions differently in freezing water vs boiling water? Did I miss something in the article or is there a whole field of "expression studies"?"

The googleable term d'art for 'expression studies' is epigenetics, but researchers don't really use the term when we're talking to each other because its so absurdly broad as to be meaningless. You do, however, have it more or less right. I'm going to be using bacteria as an example, mostly because I do microbiology, but human gene expression works more or less the same way just with a few more steps.

Genomes are really just big long stretches of DNA, the wikipedia article has some astonishingly gorgeous depictions of what it looks like that are really helpful, and information is encoded in sequences that are based on one of four chemical letters. These letters can be read by enzymes that recognize their unique shapes from inside face, which is done to remake the code as either more DNA for replication like this or as RNA for the next step that I'll talk about in a bit, or by enzymes that recognize the unique shape of the face that they present to the bigger grove between the two backbones that you can see pretty clearly here. There are many proteins that attach to DNA that recognize specific sequences from the outside of the helix, and most of the are involved in guiding when and where the DNA is read. In this way specific sequences of DNA are responsible for recruiting factors involved in its own function. Those sequences can recruit more strongly or more weakly, the things they recruit can bind under some conditions and not others or work under some conditions and not others, and the regulation of all of this tends to work in Rube Goldberg type mechanisms that are somehow both fantastically beautiful and stupid.

The primary what in which DNA creates function is through the transcription of its sequences into RNA. RNA is a very similar compound to DNA that is simply missing a single functional group that prevents it from forming helices in the same way but allows it to form all sorts of other more complex structures. It has more or less the same code, only one of the letters is usually but not always missing a methyl group, and it can have a bunch of different kinds functions. It is made when an RNA polymerase enzyme gets recruited by sequence specific factors attached to the DNA at the start sites of "coding sequences" and it unzips the DNA helix along a short stretch, the RNA polymerase then progresses along the DNA while reading it and making a complimentary stretch of RNA based on the code it passes over. Eventually it reaches a sequence intended to kick off the RNA polymerase through one of a few different mechanisms releasing both the RNA polymerase and the new stretch of RNA.

This RNA strand might be designed by the DNA that encoded it to fold into an enzyme that has some function, or be designed with a sequence at the start of it that recruits one of those RNA enzymes (the ribosome) to start translating the sequence into an entirely different molecular language. The ribosome reads the RNA three letters at a time and attaches one of various amino acids to a growing chain based on which three letters are next. The amino acid sequence that results then folds into a specific shape that then has function. There are various steps in this that can also be messed with in order to regulate when proteins are made and when they aren't. For example, the sequence that recruits the ribosome can do so in a stronger or weaker way, or there can be a sequence that makes a knot that stops everything above or below astonishingly specific temperatures, or a complimentary strand of RNA that matches the opposite sequence can be made - creating double stranded RNA that is immediate degraded in modern cells (Many viruses use double stranded RNA genomes, and many more ancient viruses likely did).

This is the classical example of how it works
posted by Blasdelb at 10:08 AM on December 3, 2013 [8 favorites]

Genes don't have intent. Yes, they can be acted on by natural selection, and yes they can sometimes be selected for even if they reduce the fitness of the individual in which they reside. It's still just a stochastic process mediated by outside events though, a process which is simple and well understood (at a broad level), and which has many interesting emergent properties but which has nothing to do with selfishness (or any other volition) on the part of genes. Individuals can be selfish, and behavioral evolution predicts that selfishness will be selected for under some circumstances, but extending that by analogy to genes is inappropriate and misleading.

Everything you just said, Dawkins says repeatedly in the very book you are criticizing.
posted by empath at 10:10 AM on December 3, 2013 [4 favorites]

mary8nne: "but what does this sort of statement even mean:

"To repeat: you are 80 per cent cow."

80% in what sense?
Like if I take two books both written in the same language containing exactly the same words (but in a different order) are they 100% the same?"

I think this would be an interesting programming project of some sort. I don't think mere random letter arrangement in themselves are the same case, as there is semantics behind this as well (otherwise we'd all be 100% cow if we're just rearranging letters). I'm sure there's a way to parse out what we mean when we say this and how to implement a method to apply this to other data forms such as books. More work than I'd bother.
posted by symbioid at 10:11 AM on December 3, 2013

And this article reads like lamarkism, to me.
posted by empath at 10:12 AM on December 3, 2013

"And this article reads like lamarkism, to me."

Thats the thing though, Lamark wasn't entirely wrong. There very much are heritable traits that are influenced directly by the environment rather than than through natural selection. We inherit a lot of our DNA methylation state from the stem cells in our mothers that produced the egg that produced us, and that is if anything as affected by the lives of our maternal grandmothers. Once you start paying more attention to data than dogmatic orthodoxy, all sorts of weird shit starts making sense.
posted by Blasdelb at 10:18 AM on December 3, 2013 [5 favorites]

Scientist: "Genes don't have intent. Yes, they can be acted on by natural selection, and yes they can sometimes be selected for even if they reduce the fitness of the individual in which they reside. It's still just a stochastic process mediated by outside events though, a process which is simple and well understood (at a broad level), and which has many interesting emergent properties but which has nothing to do with selfishness (or any other volition) on the part of genes. Individuals can be selfish, and behavioral evolution predicts that selfishness will be selected for under some circumstances, but extending that by analogy to genes is inappropriate and misleading."

empath: "Everything you just said, Dawkins says repeatedly in the very book you are criticizing."

No he really doesn't. In the book Dawkins does a fancy dance around the molecular basis for it, hopping from one definition for gene to the next, specifically to avoid the natural conclusions of what Scientist is saying. The Gene is itself a shitty term that has precious little useful meaning when applied to real biology. In order to coherently call it selfish he has to distort our current understanding of molecular biology around one that is more than a hundred years old, and it has stretched too much to really succeed for a long time now.
posted by Blasdelb at 10:35 AM on December 3, 2013

> Because the headline and intro are all "Selfish Gene is Wrong, Total Bullshit, Get Rid of It, Must Go Away"

Then a page in comes the "well, selfish gene is mostly right, we are just fine tuning it."

That is link bait.

Have you never seen a headline before?
posted by languagehat at 10:51 AM on December 3, 2013 [1 favorite]

Genes, in the classical sense, are a useful enough fiction before people move on to the dizzying complexity of actual regulatory systems.

An organism's state is an emergent property of its DNA (and modifications), the cellular context of the DNA, and the external environment up to that point in time. Everything else is a simplification, the question is which models are useful under what circumstances. If I'm doing plant breeding in my backyard, I can't control the environment, but the stretches of DNA that behave as semi-independent heritable traits are the ones that I can actually transfer. If I can use DNA editing or rewrite a new genome from scratch, I can start to alter the traits that are not so simply transferred by conventional breeding.

There is no such thing as a gene in reality. This does not invalidate the concept's usefulness in certain contexts. Newtonian mechanics is provably wrong, but still a convenient simplification for many cases.
posted by benzenedream at 10:53 AM on December 3, 2013 [4 favorites]

"So, essentially, our DNA is a program, and gene expression is the command line arguments?"

This is a really great analogy in a lot of ways! Where the command line arguments are built into the program itself and alter the operation of the program in fundamental ways without altering the program itself. However, it is also a really seductive analogy that fails in subtle but incredibly important ways.

We tend to abstract the 'program' as being a code of letters written in Base-4, as I did in my comment above, but this only really loosely reflects the reality. The 'program' is really written in the chemical structure, three dimensional shape, and charges of biological molecules that, while they do follow patterns that are reliable enough to make our abstractions useful, will frustrate the hell out of anyone who is only thinking of this as a computer code.

Similarly, design by natural selection doesn't happen in the logical, ordered, hierarchical ways that programmers at least attempt to build their systems in. If there is a God of creation that went around designing the genomes of all of the living things on Earth, they are the sloppiest, most frustrating, terrible programmer you could possibly imagine. Intelligent Design proponents are particularly frustrating to me as a biologist having seen how fundamentally unintelligent the design of living critters actually is when you get down to the real moving parts. At least it is designed according to a sort of logic so fundamentally alien to our own that by any human standard we couldn’t help but call it stupid even as it is incredibly beautiful. Stupid things like fundamentally unrelated systems being piled on top of each other such that one can’t be manipulated without messing up the other – necessitating otherwise functionless patches to the paired system whenever the other is modified, or Rube Goldberg-esque fragile systems of regulation that respond to all kinds of wrong stimuli, or systems of global regulation that are pretty analogous to reading the same giant program in either Python or C++ to produce one of two desired global results, or the kinds of systems that you can just tell are 99.9% amateur patch jobs are everywhere.

Here is a talk that Drew Endy gave to a bunch of computer science oriented people that describes the problems he has had trying to exploit this kind of analogy to allow non-biologists to manipulate biological systems as well as the hopes he has for what he can do if he succeeds.
posted by Blasdelb at 11:02 AM on December 3, 2013 [12 favorites]

Cosine: "Because the headline and intro are all "Selfish Gene is Wrong, Total Bullshit, Get Rid of It, Must Go Away"

Then a page in comes the "well, selfish gene is mostly right, we are just fine tuning it."

That is link bait."

If you really need a narrative simple enough that everything is either consistently presented as a "good thing" or a "bad thing" and never a "complex thing," you would love news television these days.
posted by Blasdelb at 11:09 AM on December 3, 2013

Seriously, this article is pretty fucking impressive. It elegantly explains some fucking complex and subtle graduate level ideas to the middle school level understanding most of us only sort of half remember. I wish more science journalism were like this.
posted by Blasdelb at 11:20 AM on December 3, 2013 [4 favorites]

So, essentially, our DNA is a program, and gene expression is the command line arguments?

As a programmer I would say that if DNA is a program, gene expression is more like the way that programs branch differently given different data. And DNA, like most old production code, is filled with a lot of way old out of date code branches that are never used. But sometimes if the type of data being input changes (changes in environment), it makes sense to use a GOTO to jump to some weird subroutine that isn't generally used that might handle the data better (e.g. grasshoppers switching to locust mode due to overcrowding). And genetic accommodation is like putting in a hack to make that subroutine active no matter what the data inputs are. So basically DNA is a huge program that could theoretically do almost anything, and evolution is mostly about fine-tuning and hacking the program to work better given changes to the input data, to the point where a series of clever hacks make one copy of the program in one species behave almost unrecognizably different than another almost identical copy in some other species.
posted by burnmp3s at 11:30 AM on December 3, 2013

We humans, for instance, share more than half our genomes with flatworms; about 60 per cent with fruit flies and chickens; 80 per cent with cows; and 99 per cent with chimps.

With apologies to those who have broached this subject already, the above sentence is maddening to me. It doesn't have any sensible meaning.

#1. This meme has been kicked around forever (alright, the last 20 years at least). But we have only just recently completed sequencing the full human genome and haven't done it yet for a lot of creatures. And it's not like we have a consensus full genome sequence for a broad set of humans as of yet to enable us to say what a representative human is.

#2. Most of human genome sequences are vast desert territories. It is doubtful these are same for flatworms or chickens, and probably not even with chimps. If there is no evolutionary selection, there is evolutionary drift. (There can be selection here, but not rigorous).

#3. What are you measuring? Since it isn't the full genome or even the full set of expressed genes and proteins in each species, is it representative genes? You can compare cow hemoglobin to human hemoglobin and get a figure (like 90%).

#4. What are you measuring? Even if you limit it to a single protein are you measuring amino acid changes or nucleotide and silent changes? Is a silent change equal to an expressed change? (I would argue no.)

#5. How do you align? A technical matter, but one which there is not an agreement among geneticists. If you are lucky enough to have two protein expressing genes of the same length, the matter is somewhat easier. You still have to be concerned about multiple deletion/insertion events. Once you start getting different lengths, do you assume a two amino acid change is a single event? Is it an event equal to a nucleotide change?

#6. How do you align given introns and exons and the vast unused areas of the genome?

#7. Really, you have 6.4 billion nucleotides in human DNA (diploid, more in women, less in men), about 6.5 pg.
Flatworms? A 2005 paper gives a 340-fold range in the size of their genomes. I don't see how they could be said to be 50% related to each other. (0.06 pg to 20.52 pg)
posted by dances_with_sneetches at 11:35 AM on December 3, 2013 [1 favorite]

Oh boy, I love talking about this stuff. To expand upon dances_with_sneetches comment, which are great biological and technical computation questions/critiques that I personally believe researchers in these fields have mostly adequately answered:

We actually do have a great idea about representative human genomes. In terms of public data outside of the reference genome assembly, we also have the 1000 genomes project, and hundreds of normal human genomes that we sequenced alongside cancers. It's been a couple years since I've visited these numbers but in general:

• 50-75% of the human genome's sequence can be traced back to repetitive DNA that self-replicates (with the 75% only when you try to reconstruct ancestral sequences using human and chimp and others). This is discarded from the cross-species similarity numbers because it's thought that nearly none of it has biochemical function. However, it's been known since the middle of last century that it is a cauldron of evolution, because it's how new regulatory motifs can evolve.
• 5% of the non-repetitive human genome can be traced to a common ancestor with mouse. This is a much higher number than was expected, because of this 5% only:
• 1% of the human genome codes for proteins. This is the typical "stuff" that constitutes the machinery of the cell, that we examine in molecular biology. These are the mechanical parts of the cell (along with the RNA gene parts)

It's this 1% that is usually analyzed in detail, because it has a clear signal that we can pick out of the rest of the DNA, and it has clear biochemical function in cells. Of this 1%, 80% of the genes have orthologs (genes with a common ancestral sequence) in cow. Of this 1%, 99% is similar to chimpanzees.

And I think that different precise measures are used in the 60% chicken, 80% cow, and 99% chimp measures, they all have different metrics, IIRC.

However, to revisit the Claire & King paper I mentioned in my first comment, we've known since 1975 that these proteins genes are not the primary difference between humans and chimps, it's the regulatory elements that are the primary difference! But talk to many evolutionary biologists, and they'll get all huffed up if you say that it's not the protein coding genes, it's the regulatory stuff that matters! There are no definitive, agreed upon answers on how the measure quantity of evolution, so it's almost pointless to say it's this or it's that. But one thing is clear, Dawkin's definition of gene provides no insight or conceptual tools to think through these things.

#3. What are you measuring? Since it isn't the full genome or even the full set of expressed genes and proteins in each species, is it representative genes? You can compare cow hemoglobin to human hemoglobin and get a figure (like 90%).

For chicken and cow we have genome assemblies that are in a decent enough state that we can answer these questions. I haven't looked at flatworm ever, but I assume it would be in a good enough state that the >50% number makes sense.

#5. How do you align? A technical matter, but one which there is not an agreement among geneticists. If you are lucky enough to have two protein expressing genes of the same length, the matter is somewhat easier. You still have to be concerned about multiple deletion/insertion events. Once you start getting different lengths, do you assume a two amino acid change is a single event? Is it an event equal to a nucleotide change?

The UCSC genome browser uses a wide array of metrics, but I usually prefer the Multiz method. For example the TP53 gene in human shows homology in mammals for most of the exons, but only some exons in other vertebrates. I would guess that even frog (X tropicalis in that figure) has enough similarity for the TP53 gene to be considered orthologous. But this is a guess, as I haven't looked at this particular example in detail.posted by Llama-Lime at 12:10 PM on December 3, 2013 [12 favorites]

Here we see a transition in the understanding of the fundamental mechanisms of biological change where phenotypic plasticity replaces mutation and environmental responsiveness replaces selection. Add polyphyly replacing descent, and biology has a new set of mechanisms that evolves it beyond the theory of evolution.
posted by No Robots at 12:55 PM on December 3, 2013 [1 favorite]

Isn't there a big hole in the explanation of Genetic Accomodation that you could drive a truck through?

To quote from the article:

"Genetic accommodation involves a three-step process.

"First, an organism (or a bunch of organisms, a population) changes its functional form — its phenotype — by making broad changes in gene expression. Second, a gene emerges that happens to help lock in that change in phenotype. Third, the gene spreads through the population.

"For example, suppose you’re a predator. You live with others of your ilk in dense forest. Your kind hunts by stealth: you hide among trees, then jump out and snag your meat. You needn’t be fast, just quick and sneaky.

"You get faster. You mate with another fast hunter, and your kids, hunting with you from early on, soon run faster than you ever did

"Then a big event — maybe a forest fire, or a plague that kills all your normal prey — forces you into a new environment. This new place is more open, which nixes your jump-and-grab tactic, but it contains plump, juicy animals, the slowest of which you can outrun if you sprint hard. You start running down these critters. As you do, certain genes ramp up expression to build more muscle and fire the muscles more quickly. You get faster. You’re becoming a different animal. You mate with another fast hunter, and your kids, hunting with you from early on, soon run faster than you ever did. Via gene expression, they develop leaner torsos and more muscular, powerful legs. By the time your grandchildren show up, they seem almost like different animals ..."

So we now have grandchildren in the discussion, but at this point, according to the explanation, there has been no change in the DNA itself yet, just the way that it gets expressed. But these grandchildren - who are being described as 'almost like different animals" - were born with the same DNA you were, so how did they 'jump' so far ahead? They still had to start from exactly the same starting line that you did. This whole thing starts to sound like Lamarckism ...
posted by woodblock100 at 1:11 PM on December 3, 2013

I think this only appears to be a hole from another terminology switch around 'gene'. The 'genes' that are being selected for in these accommodation examples are the regulatory elements in DNA that allow expression to change. Whereas the field has adopted "gene" to mean only the protein segments.

Also, DNA that is in the "open" euchromatin state, i.e. in the expressed state, is much more susceptible to changes. For example in cancer, where we can watch highly-accelerated human evolution in practice, the more expressed a gene is, the more quickly that region of DNA mutates. I would speculate that there are studies out there that show, but I have not seen, that transposable elements (a type of repetitive DNA) are going to be much more likely to be active in areas of actively expressed DNA than in regions of closed up heterochromatin, and transposable elements also drive a lot of the evolution in gene regulation.

I wouldn't let an aversion to Lamarckism prevent you from reevaluating new data. This is not Lamarckism, and a binary divide between Lamarck/non-Lamarck is not a useful way to think about the actual physical processes that are going on.
posted by Llama-Lime at 1:26 PM on December 3, 2013

Apologies, my last comment was nonsense! Germ line versus soma strikes me again (the worst kind of lamarckism!) I spend too much time in cancer, where somatic mutations are the ones driving the evolution. For evolution of species, only the mutations in the germ line matter!

So the genetic accommodation story has nothing to do with changes in gene expression regulation in muscles. But it may have something to do with the plasticity of the gene expression regulation programs; those that are better able to adapt with a different gene expression program will be selected. And I interpret the "gene following" to mean that later protein evolution fixes that regulatory program more firmly in the population.
posted by Llama-Lime at 1:55 PM on December 3, 2013

But these grandchildren - who are being described as 'almost like different animals" - were born with the same DNA you were, so how did they 'jump' so far ahead? They still had to start from exactly the same starting line that you did.

I think the point is not that they have the same DNA, but that they have no new genes. Let's say that tomorrow that humans as a species decided that only 50% of the population could reproduce and that the 50% would be decided by who from each gender could lift the heaviest amount of weight over their head.

Out of the initial population, people who randomly had the sorts of advantages for weightlifting would succeed and pass on those traits to their offspring. Their children would also have an additional advantage of knowing their whole lives that they need to train in weightlifting, so even a child with less than average genetic predisposition towards lifting heavy things would be better at it in adulthood than the average person from the previous generation where lifting weights was not such a big deal. The next generation after that would be even better, because aspects that help specifically with constantly training since birth to become weightlifters (which rarely happened before) would be selected for.

It's at that point where an actual mutation might happen to make a particular person freakishly good at lifting weights beyond what was possible before, and that mutation would get spread through future generations as a brand-new gene. So overall, the idea is that DNA has a huge amount of flexibility built-in above and beyond simple on-off switches of traits, to the point where the environment itself along with natural selection can produce drastic changes in the physical traits of a species over just a few generations without needing a long trial and error process of random DNA mutations to provide a way for those changes to happen.
posted by burnmp3s at 2:12 PM on December 3, 2013 [1 favorite]

"So we now have grandchildren in the discussion, but at this point, according to the explanation, there has been no change in the DNA itself yet, just the way that it gets expressed. But these grandchildren - who are being described as 'almost like different animals" - were born with the same DNA you were, so how did they 'jump' so far ahead? They still had to start from exactly the same starting line that you did."

Concepts like ahead and behind, or more advanced and less advanced, will only ever cause confusion when studying evolution. There is only really more adapted and less adapted, which is incredibly context dependent. Its also pretty non-intuitive how massive phenotypic changes can be caused by incredibly subtle differences to coding DNA and how its expressed. While locusts and grasshoppers are almost indistinguishable on a molecular level they are still pretty ridiculously different on a phenotypic level in a way that challenges our traditional definitions of what a species is. Is it really the same critter? Has it 'evolved' in a meaningful sense of the word? We have a pretty decent sense of whats going on but what that is cannot really be adequately described by the kinds of traditional nomenclature we teach school kids.

It's also, incidentally, a misnomer that there are no changes to the DNA itself exactly. Many changes to DNA expression happen when sections of DNA are modified such that the part of the base pairs that sticks out of the major groove gets read from the outside of the helix in a different way.

"This whole thing starts to sound like Lamarckism ..."

This is something people who deal with real data have been marveling about for thirty years now, that textbooks are only just beginning to catch up is not the author's problem. The conflict between Lamarckism and Darwinian evolution has been outdated for more than a hundred years and it would indeed be pretty remarkable if it were somehow still directly relevant to our modern understanding of evolution. Regardless, there have indeed been molecular mechanisms described that have been demonstrated to allow phenotypically acquired traits to be inherited through generations.
posted by Blasdelb at 2:20 PM on December 3, 2013 [2 favorites]

So they are saying that humans and cows don't share 1760 ortholog genes? I don't buy that. If they are saying the 99% for chimpanzees and humans then there are still 220 non-orthlog genes. I don't buy that either.
posted by dances_with_sneetches at 3:44 PM on December 3, 2013

dances_with_sneetches: "56So they are saying that humans and cows don't share 1760 ortholog genes? I don't buy that. If they are saying the 99% for chimpanzees and humans then there are still 220 non-orthlog genes. I don't buy that either."

You know, all of this data is available online, I share your skepticism toward shitty journalism that makes pronouncements without citing either sources or methods, but if you think you can do a better job of analyzing genomic data than the various people who have come up with different kinds of percentages according to rigorous and clearly stated methods, then you are free to do so. If you do it right the NCBI will even do the bulk of the calculations for you so that you don't even need a nice computer.
posted by Blasdelb at 3:52 PM on December 3, 2013 [1 favorite]

The article sure makes the case once again that, as a culture, we need to be more immune to the winds of scientific fashion, to the same sort of dogmatic stance that led to plate-theory being unthinkable for a half-century. We've been through this over and over for centuries, each era thinking it finally had been handed the handle to the answer. Nature is not as simple-minded as our 'cleverness'.

The article makes the case with a considerable excess of words and handwaving.

A less rhetorical, more to-the-point example of the new territory showed up the other day in Hughes' Mice inherit specific memories, because epigenetics?

How could a smell sensitivity, formed in an adult animal’s olfactory bulb in its brain, possibly be transmitted to its gonads and passed on to future generations?
The researchers are nowhere near being able to answer that question, but they have some data that points to epigenetics.

It's great that observation eventually beats back indoctrination each time, but it's an unnecessarily rough ride for the culture every time.

I strongly suspect cosmology is head for the same rubicon.
posted by Twang at 3:53 PM on December 3, 2013 [1 favorite]

The 80% cow figure comes from the original cow genome sequence paper, and is based on orthologous genes. Orthology can be resolved even when the DNA sequence is quite diverged, (I believe as far as only 30% identity in the amino acid sequence?). Why does this not seem plausible? When I was an undergrad the chicken genome sequence was starting to become available, and it was extremely useful for predicting where human (protein-coding) genes were in the genome because it was far enough diverged that any similarity was almost certainly protein-coding sequence.

The 99% (98.7% in this paper) figure for chimp is an entirely different metric because chimp genomes are nearly identical to human genomes. 98.7% of the actual genomic DNA is of identical sequence. As in they line up exactly. There are rearrangements of segments (e.g. the famous chromosome 2 fusion in humans), and duplications of segments, but for the most part they are incredibly similar. I don't know how many or how few orthologous genes map between the two.

As Blasdelb says, if you have contradictory information the scientific community would love to be convinced of that. And I am definitely curious as to why these seem implausible.
posted by Llama-Lime at 4:02 PM on December 3, 2013

Blasdelb, you directed me to the entire freaking Genbank? That's like saying read the Encyclopedia Britannica and then find a better way to re-order information.
posted by dances_with_sneetches at 4:06 PM on December 3, 2013

Blasdelb, you directed me to the entire freaking Genbank? That's like saying read the Encyclopedia Britannica and then find a better way to re-order information.

So what's the problem? Blasdelb does that twice a week from what I can gather.
posted by Golden Eternity at 4:49 PM on December 3, 2013

Thank you - good article!
posted by socks-o-neurosis at 5:39 PM on December 3, 2013

Thanks Blasdelb & Llama-Lime for specific replies (if not the actual one I really really need to comprehend your answers (so.. no practical intelligence expansion expression? sigh)). As a non-bio person I fluctuate between awed fascination and overwhelmed stupification whenever I bump into these topics.

My impression was that the Selfish Gene concept was to genetics as the Bohr "solar system" image of an atom was to physics. A useful visualization learning image but not in any way a functional model.
posted by sammyo at 7:25 PM on December 3, 2013 [1 favorite]

Why Life Does Not Really Exist
posted by homunculus at 7:57 PM on December 3, 2013 [1 favorite]

Ya ya ya, life is like porn, you know it when you see it (and if you don't see it in time, it eats you.)
posted by sammyo at 8:10 PM on December 3, 2013

So they are saying that humans and cows don't share 1760 ortholog genes? I don't buy that. If they are saying the 99% for chimpanzees and humans then there are still 220 non-orthlog genes. I don't buy that either.

Definition of paralogs and orthologs

Note that orthologs and paralogs are only really relevant to simple cases of defined modular protein/long RNA coding entities. There is no binary quality of orthology and paralogy, and there are plenty of messy cases where there is no clear "closest relative". Look at any multi-species protein family tree and you'll notice all kinds of weird exception cases, e.g. stolen or duplicated exons.
posted by benzenedream at 9:23 PM on December 3, 2013 [1 favorite]

Great post and even better comments, thanks for all the links and leads
posted by abecedarium radiolarium at 5:58 AM on December 4, 2013 [1 favorite]

I did not like the article's use of the grasshopper -> locust thing. They seemed to be saying "grasshoppers have this incredible ability to change shape into locusts, and therefore DNA is not important to evolution". Like, I have this incredible ability to change shape from a person that is sitting down to a person that is standing up. That doesn't change the fact that DNA is what I use to reproduce.

I guess I'm supposed to be impressed because their shape changing is brought about by epigenetic stuff, but I guess I missed the part where they made a strong case that these epigenetic changes are heritable. The change from a child to an adult is also a very striking one, but when the adult reproduces, the result will be a child -- their offspring will not inherit the new "adult" shape. My children would not inherit my new "standing up" position, either.

People in this thread have made a strong case that some of these epigenetic states are indeed heritable, though. I'm just not sure if the article actually gave evidence of this and I missed it.
posted by Galaxor Nebulon at 1:20 PM on December 4, 2013

On the importance of DNA vs non-DNA state in inheritance: What would happen if you took DNA out of my cells and stuck it in some random eukaryotic cell? Is the rest of the cell's machinery sufficiently different that my DNA wouldn't function in there? How different does it have to be before my DNA doesn't work anymore?
posted by Galaxor Nebulon at 1:22 PM on December 4, 2013

All of the children of grasshoppers are grasshoppers, until the activation of what turns them into locusts, and at least almost all of the children of locusts are locusts. The change is indeed heritable without being something that could be meaningfully called a mutation. I guess its kind of weird to argue the point as the case has been made for decades now, long before I was born even, but here is a review talking about the 100+ cases we've found so far,

Transgenerational Epigenetic Inheritance: Prevalence, Mechanisms, and Implications for the Study of Heredity and Evolution
This review describes new developments in the study of transgenerational epigenetic inheritance, a component of epigenetics. We start by examining the basic concepts of the field and the mechanisms that underlie epigenetic inheritance. We present a comprehensive review of transgenerational cellular epigenetic inheritance among different taxa in the form of a table, and discuss the data contained therein. The analysis of these data shows that epigenetic inheritance is ubiquitous and suggests lines of research that go beyond present approaches to the subject. We conclude by exploring some of the consequences of epigenetic inheritance for the study of evolution, while also pointing to the importance of recognizing and understanding epigenetic inheritance for practical and theoretical issues in biology.

"What would happen if you took DNA out of my cells and stuck it in some random eukaryotic cell? Is the rest of the cell's machinery sufficiently different that my DNA wouldn't function in there? How different does it have to be before my DNA doesn't work anymore?"

If you did that with a whole lot of cells, for the most part, you would get a whole lot of dead cells. DNA has function as part of a dynamic interaction with the other biological molecules of the cell it instructs, you would end up all sorts of switches flipped in crazy ways they aren't built to handle causing chaos. Putting DNA from any cell into a red blood cell, which ordinarily has no DNA, would probably not do much other than crowd out the haemoglobin they are meant to transport. Putting the DNA of a fast replicating cell like a endothelial cell or a hair cell or an immune cell into a slowly replicating cell like a nerve cell or a muscle cell would probably have a pretty decent chance of causing cancer as DNA set to make new cells conflicted with an environment not equipped to handle that safely.
posted by Blasdelb at 2:19 PM on December 4, 2013 [1 favorite]

Scientist: "Genes don't have intent"

Have you done more than just read the title of "The Selfish Gene?"

He is quite clear that he is not supposing that genes have intent on page 4: "Before going any further, we need a definition. An entity, such as a baboon, is said to be altruistic if it behaves in such a way as to increase another such entity's welfare at the expense of its own. Selfish behavior has the opposite effect...It is important to realize that the above definitions of altruism and selfishness are behavioral, not subjective. I am not concerned here with the psychology of motives ...My definition is concerned only with whether the effect of an act is to lower or raise the survival prospects of the presumed altruist and the survival prospects of the presumed beneficiary." (emphasis mine)

Dobbs issued a clarification today that states that "The Selfish Gene" - the text - is mostly right about how genes and evolution work. As a meme (the idea of "the selfish gene" for those who don't get past the title) it can obscure some cool biology. For what it is worth, I think Dobbs is right, but I don't think that the Modern Synthesis needs to be completely thrown out because of "-omics".

If the Modern Synthesis needs to be upended, then you don't need to go after Dawkins, you need to go after Fisher, Mayr, Haldane, and Dobzhansky.
posted by cnanderson at 11:12 AM on December 5, 2013

Jerry Coyne on "Die Selfish Gene, Die"
posted by cnanderson at 11:40 AM on December 5, 2013

Maybe its that I've gotten to used to the crazy mindfucks of describing science, but Dobbs doesn't really seem to be saying anything different in his clarification at all. Like all the theories that are both wrong and useful, and how taking Gen and O Chem is all about learning this and then that rule that is always the case except for the times it isn't, I don't really get the contradiction in saying that Dawkins' book is both mostly right in what it says and entirely wrong in how it says it.

We're not really attacking Modern Synthesis at all here, just the specific ways in which a lot of models that rely on it also rely on heritable traits being transfered in something that can be meaningfully described as a descrete unit that could be called selfish.
posted by Blasdelb at 12:01 PM on December 5, 2013

Dawkins responds to "Die Selfish Gene, Die"
posted by cnanderson at 7:01 AM on December 6, 2013 [1 favorite]

This story has blown up on my twitter feed today. Here is a link to a Storify I made of tweets by Joel Mcglothlin :I think my views line up pretty close with his
posted by cnanderson at 3:30 PM on December 6, 2013

cnanderson, thanks for the links.
posted by polecat at 4:20 PM on December 6, 2013

This story has blown up on my twitter feed today. Here is a link to a Storify I made of tweets by Joel Mcglothlin :I think my views line up pretty close with his

Where the Selfish Gene goes wrong as metaphor is when we focus on the word "gene." This means so many things that confusion is inevitable.

The other place it goes wrong is when we focus on the word "self." There is no need to anthropomorphize statistical processes that are relevant to evolution, even if there were more meaningful "units" they could be applied to.
posted by Golden Eternity at 4:28 PM on December 6, 2013

Dawkins responds to "Die Selfish Gene, Die"
posted by cnanderson at 3:01 PM on December 6

Ah, someone beat me to it. Good. Slapping these devious jerks down is a constant battle.
posted by Decani at 12:42 PM on December 7, 2013

David Dobbs published a revision to this article. I particularly like that he left a link to the original article, has highlighted a PDF noting the changes, and adds hyperlinks to sources.
posted by cnanderson at 6:21 AM on December 19, 2013 [2 favorites]

Epistasis is hardly a new concept. In fact, geneticists have been arguing about its importance ever since R.A. Fisher and Sewall Wright bickered about it in the 1920s. Dawkins acknowledges a role for gene- gene interactions in The Selfish Gene, noting that ‘the effect of any one gene depends on interaction with many others.’ But research since then show that these interactions take place in non-linear, non- additive ways of a complexity impossible to understand at the time Dawkins wrote his book. Casey Greene and Jason Moore of Dartmouth, for instance, recently found that in some cases epistatic interactions seem to warp conventional gene-trait relationships so profoundly that they can often negate the gene as a trait’s reliable carrier.

If anybody had any doubts about the initial version of the article, and is still reading the comments down here, please check out the highlighted sections in his PDF. It's a perhaps clearer explanation of all the stuff we went over in this thread.
posted by Llama-Lime at 10:00 AM on December 19, 2013 [2 favorites]