There is an approximate 8% excess of Adenine and Thymine above random in the DNA of humans. This suggests mutational bias and/or non-random mutation. If 3 billion coins were found to be 58% heads vs. 42% tails, then the chance hypothesis of a random unbiased coin flip would be easily rejected. The odds of such an event happening are astronomical according to the binomial distribution.
But such an imbalance is reflected in the human genome where:
(Guanine + Cytosine) / (Total Number of Bases) = CG ratio= 42%
(Adenine + Thymine) / (Total Number of Bases) = AT ratio= 58%
Is this due to natural selection?
Well let’s ask it a different way, can selection maintain this if the mutations weren’t biased or non-random? If the mutation rate is hypothetically about 126 point mutations per individual per generation (about Moran and Gruar’s figures), and they all had an a priori probability of 29 Adenine, 29 Cytosine, 29 Guanine, 29 Thymine — that would mean selection would have to select against 8 Guanine and 8 Cytosine, to leave C+G = 42 and A+T = 58. One can of course scale the numbers, but I chose 126 to make the math easier to see.
That’s a whopping 16 mutations per individual that selection has to select against for each generation. Given the Muller limit of mutations is about 1, this falsifies natural selection as the cause for maintaining the low CG ratio unless selection can act in a synergistic sort of way.
What is peculiar however is the CG ratio is different depending on the species. Woese noticed this even in 1961:
THE ratio of guanine plus cytosine to adenine plus thymine (GC/AT) in the deoxyribonucleic acid (DNA) of a given species of a micro-organism is constant, but the GC/AT ratio of different kinds of microorganisms can vary by a factor of as much as five times
and this from wiki:
https://en.wikipedia.org/wiki/GC-content
GC content is found to be variable with different organisms, the process of which is envisaged to be contributed to by variation in selection, mutational bias, and biased recombination-associated DNA repair.[17] The species problem in prokaryotic taxonomy has led to various suggestions in classifying bacteria, and the ad hoc committee on reconciliation of approaches to bacterial systematics has recommended use of GC ratios in higher level hierarchical classification.[18] For example, the Actinobacteria are characterised as “high GC-content bacteria”.[19] In Streptomyces coelicolor A3(2), GC content is 72%.[20] The GC-content of Yeast (Saccharomyces cerevisiae) is 38%,[21] and that of another common model organism, Thale Cress (Arabidopsis thaliana), is 36%.[22] Because of the nature of the genetic code, it is virtually impossible for an organism to have a genome with a GC-content approaching either 0% or 100%. A species with an extremely low GC-content is Plasmodium falciparum (GC% = ~20%),[23] and it is usually common to refer to such examples as being AT-rich instead of GC-poor.[24]
And a curiosity, the Chargaff parity rules:
First parity rule
The first rule holds that a double-stranded DNA molecule globally has percentage base pair equality: %A = %T and %G = %C. The rigorous validation of the rule constitutes the basis of Watson-Crick pairs in the DNA do.
Second parity rule
The second rule holds that both %A ~ %T and %G ~ %C are valid for each of the two DNA strands.[3] This describes only a global feature of the base composition in a single DNA strand.
So we have the DNA that is duplicated via Okazaki fragments the antisense way and the DNA read straight up. I suppose if the 2nd parity rule were violated, we could see an excess amount of G’s or C’s depending the direction we were reading. I suppose parity rule 2 is somewhat unsurprising, except for this fact, that it is occasionally violated:
https://en.wikipedia.org/wiki/Chargaff%27s_rules
In 2006, it was shown that this rule applies to four of the five types of double stranded genomes; specifically it applies to the eukaryotic chromosomes, the bacterial chromosomes, the double stranded DNA viral genomes, and the archaeal chromosomes.[5] It does not apply to organellar genomes (mitochondria and plastids) smaller than ~20-30 kbp, nor does it apply to single stranded DNA (viral) genomes or any type of RNA genome. The basis for this rule is still under investigation, although genome size may play a role.
Is the violation because there is a mutation bias or natural selection or both?
Chargaff parity rules only emphasizes the lack of parity between CG and AT and yet this other strange feature:
Wacław Szybalski, in the 1960s, showed that in bacteriophage coding sequences purines (A and G) exceed pyrimidines (C and T).[6] This rule has since been confirmed in other organisms and should probably be now termed “Szybalski’s rule”.
A valid experimental observation is to see whether there is a real time change in the present day from generation to generation of these various ratios or parities. If for example, Plasmodium falciparum CG content doesn’t change from generation to generation even if the genome mutates, then it would seem to me mutations are biased/non-random or something.
NOTES:
1. the 42% figure was inexact since it the average wasn’t weighted by the size of the chromosomes, but I felt 42% was in the Ball Park. The unweighted average is here:
http://blog.kokocinski.net/index.php/gc-content-of-human-chromosomes?blog=2
“Random” does not necessarily imply “equi-probable”.
The hypothesis that both were equally probable would be rejected. But it could still be random.
Agreed.
I would have just supposed bias rather than non-random if the ratio was universal among all species, but each species, if bias is implicated, would have a separate bias (like a loaded die).
So it’s hard to just pull non-random off the table.
However one wishes to label the phenomenon, it does exist and it doesn’t seem reducible to simple first principles since each species has their own non-equiprobable bias.
I would think chemistry would yield a more predictable bias nearly constant across species.
Funny how afraid you are to say designed.
This is a rather esoteric topic for discussion here. Were you going to eventually suggest these observations can only be explained by ID…or ID is the best explanation?
Sling enough shit against the wall…
Is the CG ratio that esoteric? It is a feature of our genome.
In fact it might be a feature where it is really easy to imagine transitionals and small gradual steps as more and more Adenine and Thymine get added one nucleotide at a time to get an excess of 480,000,000 Adenine and Thymine nucleotides from the evolutionary starting point on the way to building a human from some ancestor.
(3.3 gigabases x 16%) = 480,000,000 DNA nucleotides on a single strand
ID? No mention of ID in the OP. 🙂
This is The Skeptical Zone, and I was hoping to see some skepticism over the power of natural selection to explain the CG ratios.
Lots of people say the coin flip issue isn’t relevant to biology or that selection can effectively work to do things analogous to making a buzzillion fair coins all heads through cumulative selection. I was just putting out scientific reasons to be skeptical of natural selection in such cases.
But if you think ID is the best explanation, and you and Petrushka come up with such inferences all on your own, I think that would be awesome.
Sal, I think we’re once again seeing the issues of treating nature with coin-flip statistics when you don’t have a clue about he mechanisms involved. You think A=T=G=C should be true, and that deviations are what? Mysterious? Designed?
Some examples: Cytosine can be methylated. When 5-methylcytosine loses its amino group, thymine results. Uracil repair systems don’t fix that base. Inefficient repair->mutational bias, and organisms that methylate C are less likely to have high GC content.
Cytosine deamination is pH and temperature dependent. T-dimers get formed by UV light. Organisms have different repair systems with different efficiencies. I could go on-but you get the point.
🙂
What is it you think natural selection fails to explain? When Erwin Chargaff demonstrated in 1950 that the four bases were present in DNA in other than 25% ratio and yet that guanine/cytosine and adenine/thymine were equal, this was a very strong hint about the double-stranded nature of the molecule.
Let me propose a little hypothesis: that the differing ratios follow phylogeny. Of course, with DNA sequencing now routine, I suspect this is largely of historic interest.
The OP said BIASED mutations. Thanks for confirming my point. Sheesh!
stcordova,
Sal:
1) Its hard to understand what your point is. What does this post do for you? What do you think the implications of that mutations aren’t A=T=C=G introconversions aren’t equiprobable in all organisms?
2) Considering you’ve been writing on biological ID/Creationism for what seems like a decade, I find it amazing that you don’t know this about cytosine deamination, and posit OMG “what if” original posts. Read Savada’s “Life, The Science of Biology.” Most Freshmen and AP bio students read it.
Why is that funny?
Well, this is the site for slinging guano.
Non adaptive features. Do you think EVERYTHING is an adaptation?
That doesn’t matter. Even if it’s not about ID it’s still about ID, or God. Those are the only things folks here are interested in being skeptical about.
I really don’t understand the point of the OP. Has evolutionary theory ever claimed that mutations are random with respect to A,T,G and C?
Because if he comes right out and says he sees design, counter arguments can be specific and direct. But it he kind of indirectly implies, hints, and otherwise circles his religious doctrine, he thinks he has plausible deniability that he is, as always, beating the same dead horse.
We know that Sal is pimping for ID, and we see that he tries to pretend otherwise until he’s backed into a corner, when he drops the thread and tries to sneak in through yet another back door.
Now, if Sal were honest enough to say “My god did this by pure miracle, that’s my story and I’m sticking with it” then you’d find no such skepticism. Instead, I think you’d find people saying “you could have said that years ago, and saved yourself all this effort to fool everyone, yourself included.”
Things sorely lacking in the specific type of evidence needed for determining the claims made for them to be true.
Actually, Murray’s certainly brought up a number of other matters equally dubious that certainly were met with considerable skepticism.
Glen Davidson
Has anyone asserted than “natural selection”, in and of itself, can explain the observed CG ratios? I ask because evolution does encompass a number of known, identifiable processes other than natural selection, which, in turn would appear to indicate that the “skepticism” you were “hoping to see” would seem to be fully present and accounted for in persons who don’t hold the position that “natural selection” has “the power… to explain the CG ratios”.
Before you go around falsely accusing me of not understanding the mechanism, you might want to take note that I posted more about cytosine methylations in 2015 than anyone else at TSZ since I was promoting the fact of epigenetics in biology of which cytosine methylations are a component. For example:
Furthermore, I never claimed what you falsely attribute to me:
So you are arguing against something I never claimed. Misrpresentation, presumption are not good arguments. And now I’m gonna learn you something. 🙂
So how do you propose UV light gets into someone’s ovaries to form T-dimers in the gametes. How about the testicles, does he walk around in the sunlight trying to get UV deep into his testicles. Hahaha! And you present yourself to criticize me? Hahaha!
Ha, most Cytosine methylations have to occur when Cytosine connected to a Guanine on the same strand to form a CpG dimer. Those only occur in about 1-1.5% of the genome, so that’s not much of an explanation.
Added to that, these CpG dimers which allow cytosine to be methylated to create epignetic marks are highly important to health, they can’t be mutated to Thymine without effect of the attendant epigenetic machinery.
I think I see what you’re getting at. You’re skeptical of the power of natural selection to select for ( or against) a change of one base pair out of 3.2 billion? An explanation of this is that there is strong selection in small regions. This would give a bias. If I remember correctly you’d find different ratios in gene rich vs gene poor areas. Since the 2 base pairs are not equivalent with respect to function selection is possible. For example A-T base pairs have a lower melting temp and are favored in regions where transcription or replication is initiated.
It could also be that the bias is due mostly to the non-equivalence in terms of chemistry and has nothing do due with selection.
Sal, I have to ask if you’re being a bit coy in claiming ID isn’t an issue here. If I remember correctly you hold to a YEC idea…at least in terms of life being a recent creation. If life is less and a few million years old drift could not have changed the GC ratio much and what we see must be pretty close to what the Creator started with. That would imply a definite purpose to the non-50% rations
Sal, Really? You have a bunch of people asking what point you’re even trying to make, and you come at me with what, this:
“So how do you propose UV light gets into someone’s ovaries to form T-dimers in the gametes. How about the testicles, does he walk around in the sunlight trying to get UV deep into his testicles. Hahaha! And you present yourself to criticize me? Hahaha!”
Your original post mentions the differing GC content of Actinobacteria, Streptomyces, Saccharomyces, Arabidopsis. Do they have testes, Sal? Are they exposed to light? Your stupid statement makes my point: different organisms have different environments, different mechanisms acting on their DNA and different DNA repair activities in their germlines. We’ll find different, but phylogenetically related GC and AT content.
If you think there’s a point to be made about these differences, please state it.
Isn’t it obvious? God made the organisms with different GC content in order to give chemists in the late fifties something to study.
Streptomyces coelicolor A3(2), GC content is 72%.
Uh Rob do you have reading comprehension problem, in the case of Streptomyces it has an more cytosine not less.
Just admit it, I caught you making stupid remarks, and the the UV explanation doesn’t explain the decreased CG of creatures like humans.
In case you don’t understand 72% > 42%. Hahaha!
Sal:” increased CG of creatures like humans. ”
That’s um, decreased from 50% We have 41% GC content. On that note:
Sal: “Ha, most Cytosine methylations have to occur when Cytosine connected to a Guanine on the same strand to form a CpG dimer. Those only occur in about 1-1.5% of the genome, so that’s not much of an explanation.”
Right, CpGs are ~1%. But for a A=T=G=C organism, the expected CpG value would be .25*.25=6.25% (or for actual current human GC content, .21*,21=4.41%). Quite a gap between expected and actual, no? So GpC erosion could account for upwards of 50% of the difference.
Sal, I never said one mechanism accounts for ALL the differences in ALL organisms, did I. Read what I wrote: environment, DNA repair mechanisms, modifications to DNA, etc.
“Uh Rob do you have reading comprehension problem, in the case of Streptomyces it has an more cytosine not less.”
Yes. Some UV exposed organisms are thought to have higher GC content because more GC=less UV indued dimers, higher fitness, where AT is more prone to mutation, and erode away.
” the UV explanation doesn’t explain the decreased CG of creatures like humans. ”
Did I say that?
Saccharomyces (yeast) has virtually no methylation, so your cytosine methylation as a cause for Thymine explanation fails for that species too.
Some explanations for point mutations anyway are here:
http://www.nature.com/scitable/topicpage/dna-replication-and-causes-of-mutation-409
UV radiation does DNA damage but for different reasons most likely than those you cite:
Sal, you talk about humans as if we are a species in isolation from all others. Have you looked into the variation in GC content within apes, or primates, or mammals? It’s possible you’re seeking an explanation at entirely the wrong taxonomic level.
Also, base compositional bias can definitely be due to selection in at least some cases. In avian mitochondrial genomes that does seem to be the case. Each codon position has a different compositional bias, the bias is driven by composition of the sense strand, and the one gene whose sense strand is the L rather than the H strand (ND6) shows the biases on that strand.
But right in the OP it is quoted:
Uh, Rob, the explanations you gave for UV are for creation of low GC content not high.
If Actinobacteria are high GC they can’t be low GC at the same time. Therefore your UV explanation is invalid for Actinobacteria. How did that UV create high GC content? 🙂
There, I just taught you some elementary logic.
Well thanks to RobC anyway despite some of his misstatements, he sparked my interest with this comment:
GpC erosion? As in genetic entropy? You don’t say.
Anyway from a professor from my undergrad alma mater who used to attend our ID meetings (he didn’t like ID, but wanted to hear what we had to say):
http://mbe.oxfordjournals.org/content/22/3/650.full
Whoa! If this is the case, and we’re still mutating the CpG portions of the DNA, how come the human race is still alive! Natural selection ain’t arresting the melt down.
“Uh, Rob, the explanations you gave for UV are for creation of low GC content not high.
If Actinobacteria are high GC they can’t be low GC at the same time. Therefore your UV explanation is invalid for Actinobacteria. How did that UV create high GC content?”
Because you’ve got it exactly backwards. Once again. Previously, I said “Yes. Some UV exposed organisms are thought to have higher GC content because more GC=less UV indued dimers and higher fitness, where AT (rich regions are) more prone to mutation, and erode away.’
“UV radiation does DNA damage but for different reasons most likely than those you cite”
Followed by a picture of T-T dimers, and the statement: “Relatively flexible areas of the DNA double helix are most susceptible to damage..” Flexible like….AT rich regions? Sal, come on.
You aren’t even reading. You’re stating either your misunderstanding or deliberate lies about what I said, and calling them my misstatements. Stop it.
I’ve said several times that these are examples, and that no one mechanism accounts for ALL the GC content differences among species. You just keep making the point I made in comment 24: “different organisms have different environments, different mechanisms acting on their DNA and different DNA repair activities in their germlines. We’ll find different, but phylogenetically related GC and AT content.” And now you’re like: hey, Saccharomyces is different than a bacterium. And we have testicles. See the word “different” I used Sal? It has meaning.
And again, the elephant in the room….what is your point, Sal?
Sal:
Um, Sal, slow down and think. Rob wrote:
That is an explanation of higher GC content, not lower.
Actually, you just made a fool of yourself.
Also from Dr. Fyxell’s paper:
CpG’s are only 1% of the genome, yet cause 1/3 of the heritable diseases, that’s a factor of 30 times compared to the strictly 1 to 1 proportional hypothesis of cause and effect!
Told you guys, mutating the Cytosine is bad juju:
See, told ya so! So I reject the CpG mutation as a good explanation for low GC content. Added to this, why should there be such intense GC islands in the genome in warm blooded vertebrates when they aren’t in cold blooded vertebrates. The structure is non-random!
This directional evolution was noted here:
Bernardi, G. 1995. The human genome: organization and evolutionary history. Annu. Rev. Genet. 29:445–476.
So the problem is even though the human genome is lower in GC, it has concentrated regions of GC. This violates the bionomial distribution (coin flip distribution). Genome organization is biased and non-random.
Though Bernardi’s isochore theory has been mostly refuted (as far as I can tell) there is a fact of CG islands.
http://genesdev.cshlp.org/content/25/10/1010.full
Did they thank Him?
Don’t like the competition?
Says the guy who just told us that fine-tuning isn’t a creationist claim.
Keiths:
But RobC confused the issue with this statement:
UV does cause cytosines to mutate:
So which UV mechanism is RobC talking about? One that destroys T or one that creates T? Sheesh!
RobC may be thinking of studies like this:
But anyway the issues are not as clear as one supposes:
So RobC gives an explanation for some high GC Actinobacteria but then it doesn’t explain low GC Actinobacteria. Further he’s left some ambiguity about what he expects UV to actually do. Does it make more C or less?
Sal,
Yes, which directly contradicts your earlier claim:
You wrote:
It is you, not Rob, who needs to slow down and employ some “elementary logic”.
The Actinobacteria Streptomyces is a soil bacteria and thus usually hides from UV. I think your hypothesis is suspect. Thanks anyway for reading and responding.
Sal,
Your opinion of the hypothesis is separate from the fact that it is an argument for higher GC content, not lower.
If you presume to lecture others on logic, it pays to get your own logic right.
“daylight UV induces a characteristic UV-specific mutation, a UV-signature mutation occurring preferentially at methyl-CpG sites”
Sal, I thought you were the big expert on CpG methylation, as you praise yourself as one in comment 21. You’re now using CpG UV mutations and skin cancer to disprove GC bias in bacteria—most of which lack CpG methylation.* Try again.
*Though interestingly, some that do-Mycoplamas, for example, show CpG depletion similar to vertebrates.
Anyway, I think you’ve done enough damage to yourself for one night. Its been a hoot, but maybe in the future, try 1) making a point 2) reading for comprehension and 3) not getting in over your head.
The closest thing to a point you’ve made is: “This violates the bionomial (sic) distribution (coin flip distribution). Genome organization is biased and non-random.” This is so remarkably stupid I don’t feel the need for further comment.
Again, its fun you’re finding these papers, but as I’ve said a half dozen times….different organisms have different environments (mutational spectra), different mechanisms acting on their DNA and different DNA repair activities in their germlines. So they evolve different GC contents.
I just love TSZ, where logic is highly overrated, until it becomes important to the locals.
Oh, Sal, since you cited the paper on the low GC actinobacter, which I’m sure indicates you read it, why don’t you quote the line about what was found in their genome regarding UV damage?
Special protective mechanisms you say? Photolysases? No…
That’d be to similar to what I said: “….different organisms have different environments (mutational spectra), different mechanisms acting on their DNA and different DNA repair activities in their germlines. So they evolve different GC contents.”
Why, you’re right. We only care about logic when it works against theists, cdesign proponentsists, and Sal. Otherwise, nobody cares about it at all. Why did nobody else ever notice this? Why has nobody ever even hinted at this obvious pattern? It’s a mystery that can be explained only by the hypocrisy of atheism.
I didn’t say methylated CpG is or isn’t a cause for high CG in bacteria. You’re the one who put the UV and methylated C on the table as an explanation of low CG. And an explanation of low CG is not an explanation of high CG, and you got caught making an utterly stupid remark.
Your assertion of UV creating high CG in soil bacteria where UV can’t reach them is right up there with the rest of your statements.
I think we need a “humorous graphic” of an imploding Sal rather than an exploding weasel.
Oddly enough, I can’t find any such assertion. You have mastered the art of seeing what you wish to see, but you’d be advised to keep what you see to yourself, lest others notice your special ability and consider its ramifications to all you write.
The Actinobacteria Streptomyces has a genome size of approximate 10 mega bases. As seen above Actinobacteria go from low GC to high.
From:
http://onlinelibrary.wiley.com/wol1/doi/10.1111/j.1758-2229.2011.00274.x/full
the GC content goes in Actinobacteria as low as 37% to as high as 75%
I now show why these extreme deviations from an unbiased mutation scenario are astronomically improbable, thus mutational bias in the genomes is most certainly established.
Taking the formula for the binomial approximation for the normal distribution for standard deviations of an unbiased binary outcome as presented here:
let
p = 0.5 = probability of being CG in an unbiased mutational scenario
n = 10,000,000 = approximate actinobacterial genome size
sigma = standard deviation
= SquareRoot ( n * p * (1-p) )
= SquareRoot ( 10,000,000 * 0.5 * 0.5) = 1581 nucleotides
So a 5% deviation from the expected value of 50% CG in a 10,000,000 bp genome is 500,000 bases.
500,000 / 1581 = 316 standard deviations from the mean
Given that about 27 standard deviations is beyond the Universal Probability Bound suggested by Seth Lloyd.
So the Actinobacteria clearly have a biased mutation in their various GC ratios. The cause has been argued to be biased mutation.
But why is there such strong bias in one Actinobacteria (with 37% GC, 822 standard deviations) and another Actinobacteria (with 75% GC, 1581 standard deviations).
If one wants to argue differing mutational mechanisms or differing DNA repair mechanisms, that’s fine, but I think I’ve established the mutations are biased in a variety of species.
If Kimura is right that most molecular evolution is neutral, it follows the GC ratios are due to biased and/or non-random mutations rather than natural selection.
I would argue against randomness centered on a simple bias. I think non-randomness is indicated because purely random walks are not expected to create functionality.
One can’t, for example, just mutate CpG dimers in DNA and expect good results in light of the fact 1/3 of heritable diseases are associated with mutations at these sites.
So if these organism arose via mutation, I think the mutations are baised and also non-random.
NOTES
26 or 27 standard deviations being the UPB was computed here: