An evolution revolution

DannyL

Wild Horses
I don’t know enough about evolutionary science to really write that much informed about it, but this story is blowing my mind:

http://www.guardian.co.uk/science/2010/mar/19/evolution-darwin-natural-selection-genes-wrong

Particularly this bit:

“But Lamarck was scorned for a much more general apparent mistake: the idea that lifestyle might be able to influence heredity. "Today," notes David Shenk, "any high school student knows that genes are passed on unchanged from parent to child, and to the next generation and the next. Lifestyle cannot alter heredity. Except now it turns out that it can . . ."

What?! It turns out Lamarck was right (or at least not totally wrong). The one thing this article doesn’t get over is what a major, major heresy this is – Lamarck is often wheeled out as an example of “bad science” and is generally extremely reviled. And for the article to state that such a key tenet of evolutionary theory (the inviolability of the gene) is now up for questioning – it’s a major revolution in thinking.

The article has lots of other good stuff to say as well. Well worth a read.
 

Mr. Tea

Let's Talk About Ceps
Fascinating stuff, Dan. I've been hearing bits and pieces along these lines for a few years and it's very interesting, especially in as much as it applies to humans.

It's a typically hyperobolic headline, though - clearly not "everything we've been told about evolution is wrong", as this new research hardly overturns Darwinian evolution-by-natural-selection. Rather, it's an adjunct to it. It's important not to overstate or simplify the Darwinian picture; for example, no-one thinks genes are "inviolate" - we've known for decades that genes can mutate at any given cell division and it's these mutations (as well as the mixing of genes during sexual reproduction) that give rise to variation in the first place, which natural selection then acts on.

This sentence is very telling, especially the bit I've highlighted:

What if Darwin's theory of evolution – or, at least, Darwin's theory of evolution as most of us learned it at school and believe we understand it – is, in crucial respects, not entirely accurate?

Thing is, most science that we learn at school isn't entirely accurate. In fact some of it is plain untrue. Atoms do not consist of electrons 'orbiting' nuclei like a miniature solar system; nuclei do not consist of a hard cluster of ('elementary') protons and neutrons like billiard balls that have been glued together - and an organism's biology is not determined uniquely by its nuclear DNA (mitochondrial DNA has been known about for ages, for example). And we've known for a long time that the environment can directly affect genes, rather than just their likelihood of being passed on, by inflicting chemical damage that causes mutations.
 

DannyL

Wild Horses
Yeah, I recognise the hyperbole – and the false claims of the headline. The author writes a kind of summing of his reading in the self-help field every Saturday in the magazine – I quite enjoy it but he's obviously not a hard scientist and he gets a right shoeing in the comments boxes.

Having said that I liked a lot of what he said – criticism of pop-Darwinism and the way that evolutionary psychology is used in a “just so” story sense to apply to the most banal situations.

I]we've known for a long time that the environment can directly affect genes, rather than just their likelihood of being passed on, by inflicting chemical damage that causes mutations[/I]

As I said above what shocked me was the mention of Lamarck who, IIRC, claimed he had bred certain traits into toads which were then passed down to the generations, implying that the genetic structure had been altered by this process. it’s cited in Koestler’s book “The Case of the Mid-Wife Toad”. I thought he was a bit of a scientific bogeyman, much mocked and maligned.
The thing that got me about the article is not a challenge to “Darwinism” as such (I could care less about ideological pissing battles) but just that such a central key idea – genes do not adapt and create positive traits that are passed down due to environment – could now be questioned struck me as remarkably odd. Seems to me to such a radical shift, and very different from my previous understanding of genetics. Very possibly, this is just my ignorance and the idea has been around for awhile.

Also, surely “chemical damage” is very different from positive adaption and the passing on of said trait?
 
Last edited:

Mr. Tea

Let's Talk About Ceps
If you want to hear something that'll freak you out - I read recently about the discovery of a piece of snake DNA found in modern ruminants (cattle/bison/sheep/goats), that seems to have been transferred horizontally by a virus around 40 MY ago (the last common ancestor of modern mammals and reptiles lived hundreds of MY ago, by contrast). I mean, horizontal gene transfer between microbes is one thing, but between eukaryotic (multi-cellular) organisms it's just fucking nuts. And awesomely cool, I think.

Apparently this discovery isn't even that new - I did a quick google and it's mentioned in this article from ten years ago.
 

nomadthethird

more issues than Time mag
Interesting thread... but Lamark is still wrong.

Acquired characteristics are not passed down genetically, which was the focal point of his theory of evolution. A gamete's genes can acquire new characteristics that the parents didn't contribute to it (called "non-parental" or "recomibinant" sequences) through a process called "crossing over" during Prophase I of meiosis. Basically, homologous chromosomes can form "chiasmata" (little crosses) when they meet, and then the arm of the cross can switch to the other chromosome, which recomibines the gene sequences on the chromosomes from the mother and father.

A lot of geneticists think crossing over is an even bigger source of variation in eukaryotes than other forms of mutation (deletion, translocation, inversion, trisomy, aneuploidy).

Edit: In my experience, Lamark actually gets a lot of credit, if for nothing other than the fact that his research greatly influenced Darwin's theories.
 

Mr. Tea

Let's Talk About Ceps
A gamete's genes can acquire new characteristics that the parents didn't contribute to it (called "non-parental" or "recomibinant" sequences) through a process called "crossing over" during Prophase I of meiosis. Basically, homologous chromosomes can form "chiasmata" (little crosses) when they meet, and then the arm of the cross can switch to the other chromosome, which recomibines the gene sequences on the chromosomes from the mother and father.

Interesting...so does this process 'remix' codons into new genes, or just mix existing genes into new sequences? And if it's the latter, how does this differ from normal sexual reproduction, whereby the two parents' genes get mixed up anyway? Is it a different degree of mixing, something like that?
 

nomadthethird

more issues than Time mag
Interesting...so does this process 'remix' codons into new genes, or just mix existing genes into new sequences? And if it's the latter, how does this differ from normal sexual reproduction, whereby the two parents' genes get mixed up anyway? Is it a different degree of mixing, something like that?

I think I just answered this question on my genetics test.

Basically, what's getting mixed up is the sequences of codons. But genes are just sequences of codons. Usually, an entire gene or set of genes "crosses over" during homologous recombination. And most genes have, in humans, something like 3k bp (so 1000 codons).

The method of mixing is different in XX and XY individuals, although I'd wager the degree is similar. Some people may argue that XY individuals get less crossing over in the gametes, because the X and the Y are so asymmetrical in size. I'm not entirely sure if that's the case, but it seems plausible.
The autosomal cells of an XY individual, of course, still gets plenty of crossing over.

Anyway, females get one X from mom, one X from dad. Males get one X from mom and a Y from dad. What's interesting is that, if gene expression of both X chromosomes in females were left unchecked, women would have far more variation than men and an excess of gene expression. So in order to control gene dosage in XX individuals, there's this process whereby one allele on every X chromosome (either the one from mom or the one from dad) gets randomly shut off. The shut off alleles are called Barr bodies.

If you think about it, this means that women are basically combinations of their grandmothers' genes, put through a blender, with each gene containing one activated allele from either grandmother, randomly assorted into a human being. Men, on the other hand, are much more like their maternal grandmothers or mothers than they are their grandfathers, genetically: most of the genes for brain function, and traits related to personality, are on the X. The Y only codes for a few traits, and the rest of the stuff on it is hormone regulatory stuff and a bunch of repeating sequences that don't seem to have a function (tho I'm sure they'll discover one sooner or later).

This makes perfect sense when I apply it to people I know. I look like a combination of my grandmothers. My brother looks much more like my mother's side of the family than he does my father's. But the margin of error here is rather large...
 

Mr. Tea

Let's Talk About Ceps
Interesting...thanks for that.

Yeah, you can't really read too much into familial resemblances I think - my brother and I are startlingly dis-similar in looks, each of us is convinced the other was found in a basket or something.
 

nomadthethird

more issues than Time mag
Sorry, I still didn't address part of that question.

Every gene has one allele from mom and one from dad. So every gene is already "mixed". Crossing over just recombines what's already mixed, to form even more novel combinations.
 

nomadthethird

more issues than Time mag
Interesting...thanks for that.

Yeah, you can't really read too much into familial resemblances I think - my brother and I are startlingly dis-similar in looks, each of us is convinced the other was found in a basket or something.

Yes, and it's important to remember that the genes that code for things you can't see are often a lot just as important as the ones that code for the visible phenotypes, in terms of function and viability.

What's really fascinating is how X-linked traits work. Men are more often colorblind not because it's a Y-linked trait, but because it's X-linked. This happens because of the asymmetry of the X an Y. The Y has no gene for colorblindness. So if your mother gives you one allele for colorblindness, you're going to be colorblind. There's no chance dad's allele can save you from it.
 

Mr. Tea

Let's Talk About Ceps
Ah, that's news to me - I was still working on the high-school-biology basis that you get this gene from ma, that gene from pa...and alleles goes kinda beyond my knowledge altogether.

Still, it's a neat illustration of the school-science-lie principle I mentioned upthread.

Edit: hang on, I'm just being dense here, aren't I - we all have two copies of each gene, of course, except those on the sex chromosomes in the case of men. Two copies which may be the same or different. Sorry, it was the word 'allele' that threw me.
 
Last edited:

nomadthethird

more issues than Time mag
Ah, that's news to me - I was still working on the high-school-biology basis that you get this gene from ma, that gene from pa...and alleles goes kinda beyond my knowledge altogether.

Still, it's a neat illustration of the school-science-lie principle I mentioned upthread.

Edit: hang on, I'm just being dense here, aren't I - we all have two copies of each gene, of course, except those on the sex chromosomes in the case of men. Two copies which may be the same or different. Sorry, it was the word 'allele' that threw me.

Close...it's one copy of each gene, each gene has two alleles, except in the case of men on the sex chromosomes.

I know, it's confusing.

What's worse is the "distinction" between sister chromatids and chromosomes. It took me a few weeks to work that one out. That or when they give you this:

AaBB x aaBb

and make you work out three generations of offspring in the proper mathematical ratios-- once with linkage, once without, once with linkage and crossing over
 

grizzleb

Well-known member
The old punnet square eh. As invented by the same guy who made strawberry punnets I'm let to believe.
 

Mr. Tea

Let's Talk About Ceps
xkcd+date+between+biologists+discussing+Punnett+Squares.png
 

padraig (u.s.)

a monkey that will go ape
I have also been doing this stuff all semester.

Interesting...so does this process 'remix' codons into new genes, or just mix existing genes into new sequences? And if it's the latter, how does this differ from normal sexual reproduction, whereby the two parents' genes get mixed up anyway? Is it a different degree of mixing, something like that?

meiosis is a part of human sexual reproduction, or at least the precursor - it's the formation of the gametes (i.e., sperm or egg, depending) which will eventually fuse during fertilization to form a zygote. as nomad alluded to, crossing over is more the exchange of entire genes (or alleles) than a remixing of the primary DNA sequence, which would be more in the realm of mutation. one important thing to remember is that coding DNA only makes up ~1.5% of the human genome, so a lot of times you're just swapping around non-coding stuff (tho a fair portion of that has other functions, usually something to do w/regulation). when genes are swapped, there's ways that you can actually measure they physical distance between them on the chromosome, depending on how often the traits they code for are expressed in a population.

& to your original point - you're correct, epigenetics are a more of a corollary to Darwin - it goes w/o saying that none of this would give Richard Dawkins a fit as the author of that stupid article suggests - than a refutation. what defines epigenetic is that there is an alteration w/no change to the primary DNA sequence, usually involves some kind of change to the actual structure of the chromosome, i.e. DNA methylation (gene silencing) or modifications to the histones in nucleosomes. where it gets tricky is all the mutual effects that different things have on each other - that's molecular biology, there's a million things & most of them are acting on each other somehow & it can be a real bastard to figure out what's doing what to what. a lot of it's not that well understood. I guess the big idea is that people thought to alter gene expression you had to alter the DNA sequence & then epigentics came along & was like "no, there's this kooky back door, check it out." that's another biology,thing, there's not all hard & fast rules & equations like physics - I mean, we have to follow rules (most of them made up by physicists), but a lot of times if you ask a biologist a question s/he'll be like "I don't know. no one knows."

anyway, epigenetics has all kinds of cool potential applications. take medicine - for cancer, say, if you could figure to out a way to induce epigenetic changes in cancer cells, or reverse the ones that they induce, some means of getting the cancer cells to recognize molecular signals that tell them to stop dividing &/or go into cell death (apoptosis), then you'd have a cure for cancer w/o such deleterious side effects (granted, easier said than done). or just understanding the role that epigenetic changes play in the etiology of various diseases.
 
Last edited:

padraig (u.s.)

a monkey that will go ape
well, if there's a gene that codes for irie, & an epigenetic change has turned it on (or the opposite, that there's a mechanism which inhibits the production of irie, and that inhibition has been turned off) & it's a heritable change, then yes, Zhao, there is indeed a chance that your kids could be born irie. keep in mind that it's also dependent on the irie allele from your partner, & environmental factors. and I have to caution you that the mechanisms by which irie is produced or inhibited are - almost certainly - poorly understood.
 

nomadthethird

more issues than Time mag
I have also been doing this stuff all semester.



meiosis is a part of human sexual reproduction, or at least the precursor - it's the formation of the gametes (i.e., sperm or egg, depending) which will eventually fuse during fertilization to form a zygote. as nomad alluded to, crossing over is more the exchange of entire genes (or alleles) than a remixing of the primary DNA sequence, which would be more in the realm of mutation. one important thing to remember is that coding DNA only makes up ~1.5% of the human genome, so a lot of times you're just swapping around non-coding stuff (tho a fair portion of that has other functions, usually something to do w/regulation). when genes are swapped, there's ways that you can actually measure they physical distance between them on the chromosome, depending on how often the traits they code for are expressed in a population.

& to your original point - you're correct, epigenetics are a more of a corollary to Darwin - it goes w/o saying that none of this would give Richard Dawkins a fit as the author of that stupid article suggests - than a refutation. what defines epigenetic is that there is an alteration w/no change to the primary DNA sequence, usually involves some kind of change to the actual structure of the chromosome, i.e. DNA methylation (gene silencing) or modifications to the histones in nucleosomes. where it gets tricky is all the mutual effects that different things have on each other - that's molecular biology, there's a million things & most of them are acting on each other somehow & it can be a real bastard to figure out what's doing what to what. a lot of it's not that well understood. I guess the big idea is that people thought to alter gene expression you had to alter the DNA sequence & then epigentics came along & was like "no, there's this kooky back door, check it out." that's another biology,thing, there's not all hard & fast rules & equations like physics - I mean, we have to follow rules (most of them made up by physicists), but a lot of times if you ask a biologist a question s/he'll be like "I don't know. no one knows."

anyway, epigenetics has all kinds of cool potential applications. take medicine - for cancer, say, if you could figure to out a way to induce epigenetic changes in cancer cells, or reverse the ones that they induce, some means of getting the cancer cells to recognize molecular signals that tell them to stop dividing &/or go into cell death (apoptosis), then you'd have a cure for cancer w/o such deleterious side effects (granted, easier said than done). or just understanding the role that epigenetic changes play in the etiology of various diseases.

It's funny you say that about physics, I have biology professor who, whenever you ask him a question he can't answer, he just says, "it's a physics thing", even when it clearly isn't, as a joke.
 

padraig (u.s.)

a monkey that will go ape
well yeah, for example all the biochemical reactions in the body have obey the laws of thermodynamics. all bodies have to obey some kind of mechanics, whether classical or quantum (& I know special relativity lurks somewhere in there as well). biology doesn't really have the laws in the same sense that chemistry & physics do, but we make use of their laws in biology.

I read a story about this professor of neurobiology at Princeton - apparently he was originally going to become a physicist. as an undergrad at Caltech, he was taking a mechanics class & another course in molecular & cellular bio. he went up to the physics prof w/some question & the prof was like "oh yeah that's already been thought of" & wrote down a bunch of equations on a sheet. then he went to his bio professor & asked dude a question about something with synapses, can't remember what, & the bio guy, in classic fashion, said "I have no f**king clue." upon which the Princeton guy decided to switch fields.
 
Last edited:
Top