Dennis, thank-you for this series. I appreciate that you have attempted to simplify this subject for the lay-person, such as I am. Not everyone is a scientist, but surely there are many others like me, who are interested and only wish to understand more.
It was particularly interesting for me to read about the evolution of language. My mother (she is now deceased) was born in Friesland. and my father was born in Drenthe, and he spoke that dialect. The two dialects (as my parents called them) had enough similarities that they could obviously understand one another. They eventually married and immigrate to Canada after the war.
Friesland and Drenthe are provinces in Holland that are not separated by very many miles, since Holland is a very small country itself, but travel was more limited years ago, which is why the languages differ I would assume.
You seem to want to have it both ways, Preston. Percentage identities of genomes is not important, you suggest in paragraph one. But then in the third paragraph, you go the other way again, “Common descent, recently , is the obvious explanation … for the high degree of identity.”
I agree that whether identity is similar at a 85% level or 90% level is not that important, but when we have a 99% similarity, it is quite quickly touted as very significant in the interpolation. I am quite convinced, even by simply listening to the statements made here, that the level of similarity is definately thought to be important. Interesting.
And I agree that other arguments for common descent would be required.
John, I suspect that the only reason anyone ever made anything of levels of identity of nt sequences of genomes was that even biologists were surprised by the high identities seen between chimp and human when corresponding segments were sequenced in the early days of sequencing. In those days they would have cloned a human cDNA or DNA fragment and then someone would clone a corresponding fragment of chimp DNA, detecting the one of interest by hybridization with the human fragment.
It would have been expected that protein sequences would be quite similar (a few were done before DNA sequencing,) but the fact that the DNA sequence, even that of non-coding DNA segments, was so highly similar was, I suspect, something of a surprise. That high similarity is evidence that speciation was recent in evolutionary terms. That was the point of my last paragraph. The level of identity decays in regions where the sequence has no effect on function. Bits of sequence can remain recognizably similar in pseudogenes between birds and mammals, but if it’s similar at distances greater than that it is taken as a good sign that there is conserved function, even if we don’t know what it is yet.
It sounds to me like he may also have gotten that 70% figure from the young earth creationist Jeffrey Tomkins. Tomkins likes to bandy around that same figure. Unfortunately, Jeffrey Tomkins gave the game away when he used the same flawed method to calculate the differences between humans and Chimps in the GULO pseudogene (most likely junk DNA). He came up with a results of between 84% - 68%.
The fact that he restricted his analysis to one gene, made it fairly easy to test and show that this was complete nonsense.
Where he made the claim that “The 28,800 base human GULO region is only 84% identical to chimpanzees”, the actual result (which can simply be found by counting the differences after alignment) is that they are 97% identical.
He made another claim regarding the 13,000 bases preceding this sequence, stating that they were only 68% identical. The actual result is that they are 98% identical (once again calculated simply by counting the differences after alignment).
Jeffrey has been challenged on this and has refused to correct his errors. Either he is intentionally lying about this or he is extremely incompetent. Either way, Poythress seems to be relying on discredited results (which an 8th grader could showe be be false) to make his case.
Actually 2% or 60,000,000 mutations is exactly what we should expect from neutral theory given the time since our divergence, the generation time and the average number of mutations per generation.
hi aceofspades:
actually evolution doesnt predict anything in terms of changes in the genome. if it was 5% its fine and if it was 10% its fine with evolution prespective. we also dont know how old is chimp or humans. so its another problem here.
It would help your argument if you read up on the science before posting about it. There are very specific predictions made about human-chimp similarity based on divergence times which we estimate from the fossil record.
Currently recent estimates suggest that humans and chimps diverged about 10 million years ago. On average, successive generations for humans and chimps are about 25 years apart. That means that there have been about 400,000 generations separating us from the last common ancestor between humans and chimps and another 400,000 generations separating modern chimpanzees from that LCA. That gives us 800,000 generational changes in total.
We know that there are on average 100 new mutations in each generation. That means that after 800,000 generational changes we should expect there to be about 80 million differences between humans and chimps.
This is almost EXACTLY what we find.
The human genome size is 3.2 billion base pairs. It is 97 - 98% similar to Chimpanzees (if we count up the number of mutations required to change a human to a chimp - acknowledging that most indels are single step mutations). This gives us a total of about 80 million differences between humans and chimps.
This means that the genetics has now confirmed the evidence from archaeology.
Here are some things you can read up on:
Estimating the Human Mutation Rate: Biochemical Method
Estimating the Human Mutation Rate: Phylogenetic Method
Humans and chimps diverged earlier than previously thought
Humans and chimps diverged earlier than previously thought part 2
Humans and chimps diverged earlier than previously thought part 3
Ardipithecus ramidus and the Paleobiology of Early Hominids
In effect, there is now no a priori reason to presume that human-chimpanzee split times are especially recent, and the fossil evidence is now fully compatible with older chimpanzee-human divergence dates [7 to 10 Ma (12, 69)] than those currently in vogue (70)
Strong male bias drives germline mutation in chimpanzees
Under a model in which the mutation rate increases linearly with parental age, the rate of neutral substitution is the ratio of the average number of mutations inherited per generation to the average parental age. We predict the neutral substitution rate to be ~0.46 × 10−9 per base pair (bp) per year in chimpanzees, compared to estimates in humans of ~0.51 × 10−9 bp−1 year−1 (9). These results are consistent with near-identical levels of lineage-specific sequence divergence (12) but surprising given the differences in paternal age effect. In the intersection of the autosomal genome accessible in this study and regions where human and chimpanzee genomes can be aligned with high confidence, the rate is slightly lower (0.45 × 10−9 bp−1 year−1) and the level of divergence is 1.2% (13), implying an average time to the most common ancestor of 13 million years, assuming uniformity of the mutation rate over this time (95% ETPI 11 to 17 million years; table S11).
hi again aceofspades. lets start with the claim that the splitting time was 6-10 my ago. first, human have unique proteins. new protein can need somthing like 10^50 mutations or even more:
so first-how you can push 10^50 mutations into 10 milion years?
secondly- real phylogeny need fossils. but the oldest fossil of chimp we have is about 500,000 years:
when humans fossil is much older and about more then 2 my. so its not a real hierarchyc phylogeny.
3- if its true that one generation=100 new fix mutations its mean that scpecies that splitting older then 100 my will have completely different genome(for example 2 scpecies of sponge)?
First of all, Almost all of our new proteins are made by orthologous genes. They are members of existing gene families that have been duplicated and then altered slightly. This doesn’t require 10^50 mutations. It only requires two.
Secondly, there may be a handful of de-novo genes but whether these are actually functional genes or whether they are simply spurious transcriptions that don’t have a function is currently up for debate. Either way, a paper looking at 60 of these so called de-novo genes found almost identically matching sequences in gorillas and chimpanzees. All that had happened in each of these 60 cases was that a single simple mutation created an ATG start codon or destroyed a premature stop codon which lead to the creation of an open reading frame and a sequence that was once non-coding suddenly started being transcribed, but at very low levels. Once again, these didn’t require 10^50 mutations. They only required 1.
What you won’t find anywhere in the human genome is a gene whose origin cannot be explained because no sequence like it can be found in other great apes.
So your 10^50 number is irrelevant to this discussion.
real phylogeny need fossils
It doesn’t need fossils. Real phylogeny can also be done with genomes and by looking at phenotypic similarities and differences. You don’t need fossils to know that lions and tigers share a recent common ancestor do you? The same is true for great apes. This was obvious before genetic sequencing and now that we have the genomes, these close relationships have been verified.
Also, there are fossils linking humans and chimps that are MUCH older than 500,000 years.
Sahelanthropus - 7 million years old
Orrorin tugenensis - 6 million years old
Just because we haven’t found Chimpanzee remains older than 500,000 years doesn’t prove anything. Fossils are rare but we are uncovering more all the time.
if its true that one generation=100 new fix mutations its mean that scpecies that splitting older then 100 my will have completely different genome(for example 2 scpecies of sponge)?
First of all, the 100 new mutations per generation only applies to humans and chimps. Not sponges. It mostly has to do with how many cell divisions human and chimp male germline undergo.
Secondly, only neutral mutations (most mutations are neutral) will be fixed at the background rate of evolution. Deleterious mutations will quickly be weeded out and advantageous mutations will be fixed faster. We generally find that DNA in coding regions (exons) remains mostly unchanged for many more millions of years.
first, you right that the real number of true unique proteins may be in debate. so lets take the examples that we know they are real:
primate family have unique genes without any orthologous. so we need just one to show that there is no time to evolve those proteins in primate family. the first primate date something like 70-80 my. again- how do will push 10^50 mutations to 80 my?
you said:
"Real phylogeny can also be done with genomes and by looking at phenotypic similarities and differences. "
i agree its also a good way. and this is the problem:
"By contrast, humans share at least 28 unique physical characteristics with orangutans but only 2 with chimps and 7 with gorillas, the authors say. "
so again- no hierarchy.
about the mutations rates- i think sponge rate(or other animals) are not so different from humans rate. so its no so relevant to my claim.
I don’t think you understand the paper you’ve referenced. These orphan genes didn’t just pop out of nowhere with no explanation. 99% of their sequence already exists in other primates like Chimps and Gorillas, the only difference being that in some species (e.g. Gorillas and Chimps) these sequences are non-coding whereas in another species (e.g. humans) they have gained a simple mutation which allows them to be transcribed.
This is exactly what I was talking about when I said:
Secondly, there may be a handful of de-novo genes but whether these are actually functional genes or whether they are simply spurious transcriptions that don’t have a function is currently up for debate. Either way, a paper looking at 60 of these so called de-novo genes found almost identically matching sequences in gorillas and chimpanzees. All that had happened in each of these 60 cases was that a single simple mutation created an ATG start codon or destroyed a premature stop codon which lead to the creation of an open reading frame and a sequence that was once non-coding suddenly started being transcribed, but at very low levels. Once again, these didn’t require 10^50 mutations. They only required 1.
Once again I challenge you to show me a single gene that appears in humans where its origins can’t be clearly traced to sequences found in other primates.
so again- no hierarchy
So what? That was 2009 and has since been overturned with the genetic evidence after the publishing of the Chimpanzee genome. Judging relatedness based on phylogeny (while helpful) is going to be more subjective than genetic evidence when it comes to 4 closely related species.
about the mutations rates- i think sponge rate(or other animals) are not so different from humans rate.
This is ridiculous. There are many things that affect the mutation rate per generation. There is no reason to suspect that the mutation rate per generation in sponges is the same as in humans. If you think you have one, please cite your sources and we can continue that discussion.
Ace, its interesting to me your statement that 99% of the sequence in question already exists in other primates, where the sequence is non-coding. But you say they have gained a simple mutation which causes them to be transcribed into the coding process. So would it be possible for the coding gene to have gained a simple mutation that prevents it from being transcribed, so that now it is non-coding… in which direction is the mutation, and what determines the conclusion? Also interesting that these non-coding genes still have valuable functions, and that the coding genes often have more than one function.
It also occurs to me that your statement “real phylogeny can also be done… by looking at phenotypic similarities…” is contradicted by your other statement that the 2009 comparison of phenotypic characteristics is now invalid, even though its only five or six years old. I guess this kind of verifies to me at least the uncertainty of this whole field. If the phenotypes of known genotypes are invalid for comparison, how do we know if the phenotypes of unknown genotypes are valid? I agree that phenotypes are more subjective.
Yes this is possible, but in many cases we can see that this is unlikely. Have a look at the supporting information from this paper for example which looks at 60 supposedly de-novo genes. (Note that it’s important to say that we don’t even know for sure whether these de-novo sequences are functional genes. All we know for sure is that amino acid sequences matching these have been detected in low levels in some tissues. This could just be spurious transcription - these sequences may not play any meaningful role).
Download Dataset S2 and Dataset S4 (word documents containing the aligned sequences in question)
Have a look at the first gene in S2 (ENSG00000176723)
This sequence exists in Humans, Chimpanzees and Orangutans. That means it probably exists in Gorillas and Bonobos too.
Chimpanzees and Orangutans have an identical non-coding sequence (meaning Gorillas and Bonobos probably have the non-coding version too). It’s non-coding in Chimps and Orangutans because of the presence of an early stop codon, making this sequence too short to transcribe.
Somewhere ion the middle of this gene, humans have gained the insertion of two letters “GT” which destroy the stop codon and create an open reading frame allowing this sequence to be transcribed.
So there are two possible ways to explain this;
-
Either this was originally a functional gene and Orangutans, Chimpanzees, Gorillas and Bonobos all happened to have the same two letters “GT” deleted from their sequences (a remarkable coincidence)
-
Or this was a non functional sequence inherited from a shared ancestor and some early human underwent a mutation gaining the two letters “GT” which created an open reading frame allowing this sequence to start being transcribed.
The simplest explanation is the latter one and that applies to all 60 sequences studied here.
Also interesting that these non-coding genes still have valuable functions
I disagree. How could you possibly know that is true for these 60 sequences? In the other great apes these 60 sequences aren’t genes because they can’t be transcribed. They may have some other function (like acting as a promoter but this is unlikely)
It also occurs to me that your statement “real phylogeny can also be done… by looking at phenotypic similarities…” is contradicted by your other statement that the 2009 comparison of phenotypic characteristics is now invalid, even though its only five or six years old
Science grows and adapts with new evidence. That’s what makes it trustworthy. What I meant when I said that phylogeny can be done by looking at phenotypic similarities was that we can easily tell that humans are great apes or that pandas are bears and that bears are carnivorans by looking at shared features. Obviously I didn’t mean to imply that phenotypic analyses are foolproof. This is why we place greater weight on the genetic evidence.
I guess this kind of verifies to me at least the uncertainty of this whole field.
It sounds to me like you’re looking for a reason to dismiss science. This sounds to me like a classic case of motivated reasoning.
If the phenotypes of known genotypes are invalid for comparison, how do we know if the phenotypes of unknown genotypes are valid?
They aren’t invalid. They are simply indicators of a likely relationship. It shouldn’t take a rocket scientist to note that Ostriches are birds because of a number of shared features. When we start getting down to more closely related species like Orangutans, Humans, Chimps and Gorillas (separated by a mere 15 million years) the phenotypes start becoming a little more woolly and we need to rely more on the genetic evidence to determine relatedness.
To answer your question, we need to do more sequencing of more species and this is being done all the time.
Hi John
I was curious so I ran a BLAT search on that de-novo gene we were talking about to see how it appears in other animals.
- Humans: It is currently transcribed in humans and I found a match with 100% identity as expected: link
- Chimpanzees: Non-functional 98.9% identical to the “functional” human sequence: link
- Gorillas: Non-functional 97.9% identical to the “functional” human sequence: link
- Orangutans: Non-functional 95.6% identical to the “functional” human sequence: link
- Gibbons: Non-functional 95.3% identical to the “functional” human sequence: link
- Rhesus Macaque: Non-functional 93.9% identical to the “functional” human sequence with a large insertion interrupting it: link
- Baboon: Non-functional 93.4% identical to the “functional” human sequence with a large insertion interrupting it: link
Can you see how as we get more distant from humans in relatedness, this “functional” sequence becomes increasingly scrambled?
A creationist that believes that this is a functional gene in baboons and macaques would be hard pressed to explain why it has gained so many more mutations breaking it than a chimpanzee. This pattern is consistent - the further from humans we get, the more mutations we find breaking it.
Can you see how the simplest explanation is that this is a non-functional sequence but a single mutation in humans allowed it to start being transcribed?
ace, first, im now talking about primate family and not just human. if i will show you that evolution cant be happen in general, then we dont need to talk about evolution of human. now, here it again:
“We identified all human proteins from Ensembl (Flicek et al. 2008) that showed significant sequence similarity to chimpanzee (P. troglodytes) and macaque (M. mulatta) gene products but lacked homologues in 13 other complete eukaryotic genomes, including 3 nonprimate mammalian species.”-
so its not just junk that evolve into functional genes. again- at least some new proteins evolve in a 80 my window. how can we push 10^50 mutations in 80my?
about the hierarchy in the primate phylogeny- so if there is no such hierarchy, how can we know who is closer to who? you claimed that phylogeny can built by morphological traits. so from morphological prespective- human is closer to orangutan then to chimp. so genetic level contradict the morphlological one.
by the way- the de novo genes in human may be indeed non-functional proteins or they can be the result of nonsense mutations in both gorila in chimp. the uox pseudogene is a good example:
“One exceptional change is a duplicated segment of GGGATGCC in intron 4 which is shared by the gorilla and the orangutan. However, because this change is phylogenetically incompatible with any of the three possible sister-relationships among the closely related trio of the human, the chimpanzee, and the gorilla, it might result from two independent duplications”.
“The nonsense mutation (TGA) at codon 107 is, however, more complicated than others. It occurs in the gorilla, the orangutan, and the gibbon, and therefore requires multiple origins of this nonsense mutation”
You were actually talking about the impossibility for new proteins to evolve in the 6-10 million years since we diverged from chimpanzees. You’re now moving the goal posts and saying that it’s impossible for new proteins to appear in primates. If I prove you wrong here again will you continue moving the goal posts? If so, I’m going to have to stop wasting my time with you.
Now regarding that paper you referenced. Did you even read the abstract?
- We estimate that around 24% are highly divergent members of mammalian protein families.
- Interestingly, around 53% of the orphan genes contain sequences derived from transposable elements (TEs) and are mostly located in primate-specific genomic regions. This indicates frequent recruitment of TEs as part of novel genes.
- Finally, we also obtain evidence that a small fraction of primate orphan genes, around 5.5%, might have originated de novo from mammalian noncoding genomic regions.
All of the DNA for these de-novo genes is accounted for in other mammalian species. They were either highly divergent copies of existing genes or they contained DNA from existing transposable elements or they originated from non-coding genomic regions that exist in other mammals. These genes weren’t built up from nothing.
Regarding that paragraph you quoted, do you know what a gene homologue is?
Homologous genes are genes that share an arbitrary threshold level of similarity determined by alignment of matching bases
All the author is saying here is that equivalent genes didn’t appear in other eukaryotes. The author isn’t saying that the sequence making up this gene appeared out of nowhere.
Genetics
Genetic evidence gives us better resolution and is more relaible. This is especially true for species that only diverged from each other recently (a mere 15 million years in this case)
by the way- the de novo genes in human may be indeed non-functional proteins or they can be the result of nonsense mutations in both gorila in chimp. the uox pseudogene is a good example:
UOX isn’t a denovo gene. It’s a pseudogene. It was once a functional gene but it was broken in the common ancestor to the apes (including Gibbons, Orangutans, Gorillas, Chimpanzees, Bonobos and Humans). Now it is simply a relic - it is a fossil of a gene that once existed and this same fossil exists in all apes as a testament to common descent.
In his first paragraph he was talking about sequence identity overall (how close is this genome to that genome). In his third paragraph he was talking about sequence identity between non-coding sequences.
There is no good explanation why a non-coding sequence should be similar between related species.
Ace, I only checked one of your references, re. Orrorin tugenensis. 20 bones were found, called specimens. Not entire skeletons, not even partial skeletons, but pieces of bone or piece of jaw, etc. A 6 my projection means no dna. From this to link humans and chimps? I don’t think so. Speculation, yes. But science, no.
Ace, I guess 60 non-functional genes compared to 30,000 (or 17,000 genes as more recently proposed) is not so many. But my understanding is that most non-coding sectiojns still perform important functions even though they do not code for protein synthesis, ie. they are not junk dna. Please correct me if I am wrong about this. So when I said “these” non-coding genes are still functional, I was referring to non-coding genes(or should say sequences) in general, not necessarily to these specific 60, although yes I did infer that these 60 could also still be functional in some way.
I understand what you are saying about the “remarkable” coincidence of the ape types all having the non-functional gene due to the deletion of GT. However, if they all derived from a similar ancestor, then the deletion could have happened before they diverged. It is also a remarkable coincidence that they all had the same sequence, which performed no function, and then suddenly voila, remove a stop codon, and a beautiful function emerges out of nothing, as it were. I don’t mean nothing genetically(since there was a sequence), but I mean isn’t it amazing that the sequence(s) were valuable rather than useless or terrible or pathetic once they became activated?
While knowledge (not science) grows with new evidence, I disagree that makes it trustworthy. In fact, because the growth is based on revision and correction, not on simple addition of knowledge, it makes the field of knowledge somewhat untrustworthy. Not totally untrustworthy. But the new knowledge must then also be taken cautiously, because based on previous experience, we are likely to find necessary revisions in the future, perhaps for the way alignments are done, or for the understanding of non-coding sequences, or for our understanding of the multiple functions of various pieces of dna. I agree that the revisions are necessary, but on the other hand it seems likely that there will be revisions to the revisions.
I find your statement totally unfair, “a reason to dismiss science”. It’s like saying that if you are a christian you must obviously be in favor of stoning your son if he dissis you. Or if you like salad, you must be a vegetarian. To suggest that one field of study, or one branch of science, or one theory in science is more uncertain, is certainly not equivalent to a reason to dismiss science. That’s kind of a “shut him up” type of statement. That’s an unscientific statement made without evidence… a broad generalization made on inductive reasoning, on incorrect inductive reasoning. That statement disappoints me.
I must have a mental block in terms of understanding what you are saying, Ace. You say recent estimates place a divergence of human and chimp dna to 10 my ago. Then you indicate generation times and number of mutations indicate a correspondence to 10my. But, you have no dna from 10 my ago, since it does not survive more than 100,000 years. So where do the estimates come from, without dna evidence? Is this a circular thing… taking generation times mutation rate, and then finding the 10 my, and reversing the calculations to prove itself? What were the estimates based on? The fossil record with 20 bones (in one case)? Here is the first thing I read from one of your references:
Our lineage’s fossil record is especially sparse between 4 – 8 million years ago. From that time period we have roughly 50 fragmented specimens making up only 6 or so individuals. It isn’t much to go on, which is particularly disappointing since genetic data indicates that humans and chimps diverged during this period. Indeed, with no archaeology from this time period (the archaeological record doesn’t start until ~2.6 million years ago) genetic data is pretty much the only source of information on the aforementioned split.
Yet genetics doesn’t provide that much information either, or at least what it does tell is fairly uncertain.
I feel you are also trying to minimize differences by saying that differences should be measured by mutation events, and not by base pairs differences. However, this is misleading and inaccurate. If one mutation (theoretically) were to result in a 10% change in the genome, would you say that based on mutations, the genomes were 99.99% similar, while the actual bp difference is 10%? Especially if that genome change had dramatic impacts on the morphology or physiology, then that would not make sense. I’m not sure it is even provable that what looks like a single mutation could not be the result of several or many mutation events giving the same result. Mutations are events. Base pairs are the result. Mutations also could have happened which are simply not obvious anymore. For example, a bp change (mutation) could only exist for one or two generations, and then get removed again by another mutation, so that we never see it, or it might get replaced by a different insertion in such a way that we would not distinguish it, we would not know whether a previous mutation had existed in that location in the sequence before.