More Recent Comments

Friday, October 16, 2015

Human mutation rates

I was excited when I saw the cover of the Sept. 25th (2015) issue of Science because I'm very interested in human mutation rates. I figured there would have to be an article that discussed current views on the number of new mutations per generation even though I was certain that the focus would be on the medical relevance of mutations. I was right. There was one article that discussed germline mutations and the overall mutation rate.

The article by Shendure and Akay (2015) is the only one that addresses human mutation rates in any meaningful way. They begin their review with ...
Despite the exquisite molecular mechanisms that have evolved to replicate and repair DNA with high fidelity, mutations happen. Each human is estimated to carry on average ~60 de novo point mutations (with considerable variability among individuals) that arose in the germline of their parents (1–4). Consequently, across all seven billion humans, about 1011 germline mutations—well in excess of the number of nucleotides in the human genome—occurred in just the last generation (5). Furthermore, the number of somatic mutations that arise during development and throughout the lifetime of each individual human is potentially staggering, with proliferative tissues such as the intestinal epithelium expected to harbor a mutation at nearly every genomic site in at least one cell by the time an individual reaches the age of 60 (6).
The main point in this paragraph is important. No matter what the exact value of the human mutation rate, every single possible point mutation will happen in just a few generations somewhere among the seven billion or so people on Earth. And each individual who lives to the ripe old age of 60 (i.e. youngsters) will have experienced a huge number of somatic mutations. Hundreds, possibly thousands, of these can cause cancer.

But let's think about the mutation rate. The value of ~60 new point mutations per generation in Shendure and Akay (2015) is quite a bit lower than the number I prefer (>100). My estimate is based on my understanding of the scientific literature. I've discussed this before in Human mutation rates - what's the right number?. The discussion in the comments section of my post is very enlightening; it reveals that the issue is complicated and all estimates, by any method, are open to criticism.

Shendure and Akay gave us four references to support their claim that the rate is ~60 mutations per generation.
  1. The first reference is paper I had not read. The authors (Campbell et al, 2012) report the genome sequences of five trios in the Hutterite population and conclude that the mutation rate is 76 mutations per generation. This is similar to the value obtained by other direct sequencing methods and subject to the same qualifications.1
  2. I also hadn't read the second paper. Francioli et al. (2015) sequenced the parents and offspring of 258 Dutch families and found a total of 11,020 de novo mutations. This corresponds to 43 mutations per generation. The paper frequently mentions "mutation rate" but they are referring to their own data and the number of mutations per generation that they detect. On first glance, I don't see how to correct for the amount of sequence they compared—was it about half of the reference genome sequence, in which case the real mutation rate would be 83 mutations per genome?
  3. The third paper is Roach et al. (2010). This is also a sequencing paper but one that I've discussed previously. The authors sequenced parents and two children and arrived at an error rate of 70 new mutations per generation.
  4. The fourth paper is Kong et al. (2012). It's the one that looked at genome sequences of Icelanders with fathers of various ages and concluded that the number of mutations went up as the age of the father increased. They estimate that the average mutation rate is 76 mutations per generation.
Theme

Mutation

-definition
-mutation types
-mutation rates
-phylogeny
-controversies
All of the rates are based on genomic sequencing. I'm not sure how you get from three estimates of 76, 76, and 70 to the Shendure and Akay estimate of ~60 point mutations per generation. Perhaps they misinterpreted the data of Francioli et al. (2015)?

Shendure and Akay (2015) discuss a number of different ways of calculating mutation rates. The first is based on the rates of mutation causing genetic diseases. These estimates date back to the 1920s and they are the reason why genetic load arguments were developed around 1950.

Typical values for genetic diseases were about 2 × 10-5 per gene per generation. It's almost impossible to relate this to an overall average mutation rate. There are more modern estimates that may be helpful according to Shendure and Akay ...
A number of distinct approaches have been used to estimate the germline mutation rate of base substitutions, which we focus on here unless otherwise noted. Historically (18), and even more recently (6), estimates of mutation rates have been derived from the incidence of highly penetrant Mendelian diseases. The largest such study aggregated data across ~60 loci, estimating an average germline mutation rate of 1.28 × 10−8 per base pair (bp) per generation (6) [82 mutations per generation].
Reference 18 is a 2004 reprint of a 1935 paper by J.B.S. Haldane. A much more appropriate reference would be Haldane's 1949 review (Haldane, 1949) where he discusses all of the problems associated with these studies. Haldane concludes that typical mutation rates are about 2 × 10-5 per gene per generation.

Reference 6 is Michael Lynch's 2010 PNAS paper that looks at a database of deleterious mutations causing genetic diseases in humans (Lynch, 2010). Lynch is mostly interested in the somatic cell mutations rate because that's what causes the most problems. He concludes that the somatic cell mutation rate is about 8 × 10-10 per site per cell division and this is quite a bit higher that the germline mutation rate (about 1 × 10-10).

The somatic cell mutation rate gives rise to thousands of mutations per cell by the time an individual reaches maturity. Lynch notes that a high somatic cell mutation rate imposes an enormous mutational load on humans.

With respect to germline mutations, Lynch calculates that the average newborn has about 76 de novo point mutations although he points out that there will also be a number of new small deletions and insertions.

Shendure and Akay know that calculating germline mutation rates from genetic diseases is very difficult. They discuss how comparisons of DNA sequences from different species can be helpful.
Phylogenetic methods have also been used to estimate mutation rates at putatively neutral loci on the basis of the amount of sequence divergence between humans and nonhuman primates, yielding a higher genome-wide average germline mutation rate of 2.2 × 10−8 per bp per generation (19). Phylogenetic methods also make assumptions such as the time to most recent common ancestor between humans and nonhuman primates, generation time, and that the loci studied do not have fitness consequences. In addition, phylogenetic estimates may be influenced by evolutionary processes other than mutation and selection, such as biased gene conversion, which influences substitution rates in mammals (20).
Reference 19 is the chimpanzee genome sequence paper from 2005. It's just one of many papers that use the phylogeny method to estimate a mutation rate [see Estimating the Human Mutation Rate: Phylogenetic Method]. Reference 20 is a 2008 paper on biased gene conversion. It's not likely that biased gene conversion will have a serious effect on calculating a germline mutation rate by the phylogenetic method. There are other, more important, sources of error.

A mutation rate of 2.2 × 10−8 per bp per generation correspond to 140 new mutations in every newborn baby (actually, in every zygote). This is about twice the rate determined by the direct sequencing method [see Estimating the Human Mutation Rate: Direct Method].

There's another way of estimating the mutations rate and that's the biochemical method [Estimating the Human Mutation Rate: Biochemical Method]. It's based on our understanding of the error rates of DNA replication and repair and it yields a value of 130 mutations per generation. This is closer to the phyogenetic method than to the direct sequencing method. Shendure and Akay don't discuss this method.

Shendure and Akey obviously prefer the estimates derived from direct sequencing of parents and children although they've picked a value (60 mutations per generation) that's on the low side of all those estimates. Nevertheless, they present a fair summary of the problems with all these estimates.
New sequencing technologies have enabled more direct estimates of mutation rates by identifying de novo mutations in pedigrees (i.e., those observed in a child but not their parents). Whole-genome sequencing studies (1–3, 9–11) of pedigrees estimate the germline mutation rate to be ~1.0 × 10−8 per bp per generation [extensively reviewed in (8)], which is less than half that of phylogenetic methods but in better agreement with disease-based estimates. An important caveat to pedigree-based sequencing is that heavy data filtering is necessary and analysis choices may influence both false-positive and false-negative rates (8). Nonetheless, complementary approaches for estimating mutation rates on the basis of the number of “missing mutations” that would be expected to have occurred in the time between when an archaic hominin individual (such as Neanderthal) died and the present (21) and the accumulation of heterozygous variants within autozygous regions of founder populations (1) are broadly consistent with pedigree-based approaches (~1.1 and 1.2 × 10−8 per bp per generation, respectively).
It is not correct to say that the direct sequence estimates are "broadly consistent" with the analysis of Neanderthal and Denisovan genomes. In fact, a low estimate (e.g. 60 mutations per generation) would place the common ancestor of humans and chimps at 10 million years ago and the divergence of Neanderthals and Denisovans from the modern human lineage would be pushed back to unreasonable times [see Human mutation rates - what's the right number?].

A twofold difference has implications so let's hope this debate is resolved soon. If we consider the upper estimates by direct sequencing then we approach 100 mutations per generation and that's not too much different than the values calculated by the phylogenetic method and the biochemical method.

I don't think that estimating human mutation rates from the frequency of human genetic diseases is anywhere near as accurate as the other methods.


1. There is a huge background of unique differences in every genome due to sequencing errors and artifacts. The only way to identify new mutations in offspring is to use sophisticated algorithms to eliminate false positives and false negatives. The number of true mutations is one or two orders of magnitude less that the noise due to errors.

Campbell, C.D., Chong, J.X., Malig, M., Ko, A., Dumont, B.L., Han, L., Vives, L., O'Roak, B.J., Sudmant, P.H., and Shendure, J. (2012) Estimating the human mutation rate using autozygosity in a founder population. Nature Genetics, 44:1277-1281. [doi: 10.1038/ng.2418]

Francioli, L.C., Polak, P.P., Koren, A., Menelaou, A., Chun, S., Renkens, I., van Duijn, C.M., Swertz, M., Wijmenga, C., van Ommen, G., Slagboom, D.I., Ye, K., Guryev, V., Arndt, P.F., Kloosterman, W.P., de Bakker, P.I.W., and Sunyaev, S.R. (2015) Genome-wide patterns and properties of de novo mutations in humans. Nature Genetics. 47:822-826. [doi: 10.1038/ng.3292]

Haldane, J. (1949) The rate of mutation of human genes. Hereditas, 35(S1):267-273. [doi: 10.1111/j.1601-5223.1949.tb03339.x]

Kong, A., Frigge, M.L., Masson, G., Besenbacher, S., Sulem, P., Magnusson, G., Gudjonsson, S.A., Sigurdsson, A., Jonasdottir, A., and Jonasdottir, A. (2012) Rate of de novo mutations and the importance of father/'s age to disease risk. Nature, 488:471-475. [doi: 10.1038/nature11396]

Lynch, M. (2010) Rate, molecular spectrum, and consequences of human mutation. Proceedings of the National Academy of Sciences, 107:961-968. [doi: 10.1073/pnas.0912629107

Roach, J.C., Glusman, G., Smit, A.F.A., Huff, C.D., Hubley, R., Shannon, P.T., Rowen, L., Pant, K.P., Goodman, N., and Bamshad, M. (2010) Analysis of genetic inheritance in a family quartet by whole-genome sequencing. Science, 328:636-639. [doi: 10.1126/science.1186802]

Shendure, J., and Akey, J.M. (2015) The origins, determinants, and consequences of human mutations. Science, 349:1478-1483. [doi: 10.1126/science.aaa9119]

The Chimpanzee Sequencing and Analysis Consortium (2005) Initial sequence of the chimpanzee genome and comparison with the human genome. Nature, 437:69-87. [doi: 10.1038/nature04072]

167 comments :

John Harshman said...

Is it possible that the methods used to correct sequence data for false positives are systematically overcorrecting? I no very little about those methods.

judmarc said...

In your numbered paragraph two above: Should the number in the last line be 86; should the text in the second-last line be "a little more than half;" or am I just misunderstanding?

Larry Moran said...

I don't know. I read the papers but I don't really understand how they filter their data.

There was one study that sequenced monzygotic twins plus the parents and that should have been more reliable but I don't think it was,

In theory you need to sequence the DNA from several different tissues in order to eliminate somatic cell mutations but that would lead to an overestimate of the number of mutations not an underestimate.

Razib Khan said...

some ppl at ASHG meeting were just using 1.5*10^-8.

Unknown said...

From the post-
“The main point in this paragraph is important. No matter what the exact value of the human mutation rate, every single possible point mutation will happen in just a few generations somewhere among the seven billion or so people on Earth.”

This conclusion doesn’t follow from the evidence presented.
The statement assumes mutations are random with respect to location on the sequence but that probably isn’t the case.

http://rsos.royalsocietypublishing.org/content/1/2/140172

Paul McBride said...

John, I'd say it's seems possible--or at least there some scope for it.

I remember hearing Reed Cartwright talk a few years ago about a study he was involved with (Conrad et al 2011). The number of false positives and somatic mutations in parent-offspring trios is enormous next to accepted germline mutations. Something like only 1.5% of the detected variation were germline DNMs. Many apparent mutations were assumed statistically to be errors based on sequencing error rates, and were filtered on this basis.

Larry Moran said...

Mutations are close enoigh to being random to make the statement correct.

Anonymous said...

The authors have found a mechanism that produces lots of mutations at particular sites in DNA, where repeats lead to slip-strand mispairing during DNA replication and then repair of mismatched base pairs. The authors studied an area of DNA where these changes made changes in an important proteins, and those changes are subject to natural selection. This is really cool.

Enthusiastic though the authors are about their findings, this doesn't mean that there aren't other mechanisms that produce point mutations elsewhere throughout the genome. If the rate of mutations by the mechanism the authors found is much much higher than that of randomly placed mutations, perhaps another generation or two should be added to "every single possible point mutation will happen in just a few generations somewhere among the seven billion or so people on Earth.”

Woody Benson said...

Larry writes, "No matter what the exact value of the human mutation rate, every single possible point mutation will happen in just a few generations." YES! The estimate defended in the first box suggests that each nucleotide in a genome will mutate about 30 times in a human generation and thus every one of the 3 types of mutation will be produced in our species about 10 times every 30 years or so. Similar mutation rates in absolute time may apply to small organisms, even if they are 'rare', such as a multivoltine insect species with a total population size of 70 million individuals with 3 generations a year. Consider, for example, a grasshopper in Texas having a mean abundance of only 100 individuals per square kilometer. [Neutral evolution will, of course, be much faster in the grasshopper.]
Really rare species probably become severely constrained in their adaptive evolution due to the slower rate at which random variation arises. But this has been known ever since 1859 when Darwin wrote about it.

Unknown said...

Perhaps I’m being overly picky but I’m not sure the mutations are random enough.

Actually we can model the mutations as random but then we would have to recognize the distribution is not uniform across the genome. In such a case the use of a genome wide description of the mutations could be misleading.
For example, we could model ‘being Mormon’ as a random variable of humans. We could then say the probability of being mormon is about 14 million divided by 7.3 billion or approx. .2%.
We could then make statements like- In a city of 100,000 people, one would expect to find about 200 mormons.
But if we looked more closely we would find the vast majority of mormons live in a fairly small area, partly because mormons make mormons and there are clusters.

Isn’t that a more apt description of what is happening with mutations?

Anonymous said...

It could be if the rate of these clustered mutations is high enough. However, the authors limit the mutations they evaluated so much that for all I can tell (which may not be enough) they're not in a position to say that the clustered mutations are all that common, compared to mutation rates elsewhere in the genome.

AllanMiller said...

Mutational hotspots occur due to meiosis as well. But there remain mechanisms which are largely position-independent - polymerase errors, incorrect proofreading, lesions (granted transition-transversion bias, so not fully 'random' either, depending on one's prerferred sense of That Word).

Mutation isn't random in the sense of being 3-way equiprobable at every position. Nonetheless, as we have a mechanism that is (approximately) equal, or at least 60-40 split, this at least will fully explore point-mutational neighbours in genetic space, granted that some changes get more hits.

peer said...

Lets assume repair mechanims evolved...

The initial mechanims surely wasn't as good as it is now. Currrently copying fidelity is almost 100% (60/3billion = 99.998%).

Copying fidelity of 90% (suboptimal evolutionary trials) are useless, as they will introduce 10% of errors to the offspring.

Copying fidelity had to be nearly perfect from the start to keep the information together.






peer said...

Why, I wonder, isn't there a Science megazine dedicated to mutations and human evolution?

peer said...

"In fact, a low estimate (e.g. 60 mutations per generation) would place the common ancestor of humans and chimps at 10 million years ago and the divergence of Neanderthals and Denisovans from the modern human lineage would be pushed back to unreasonable times..."

You only accept real science data when it confirms your preconceived ideas, right?

Confirmation driven pseudoscience.

Unknown said...

There is more literature about hotspots. I may be over estimating the skew of the distributions of mutations, but an analysis based on a uniform distribution of independent events is not an accurate one.
I don’t know what the actual distributions are or the degree of independence of each event but I’m pretty sure that the distributions are not uniform and the events are not independent.
It seems an inappropriate math is being applied to the situation.

If the different analysis of human mutation rates looked at different parts of the genome, then if the distribution is not uniform one would expect differing results from the analysis.

Anyway, I would agree that estimation from disease rate would be less reliable than the other methods discussed.

Unknown said...

Can you please go over for me how you arrived at 82 mutations above? As I see it 3.3 x 10^9 times 1.28 x 10^-8 yields 42.
The largest such study aggregated data across ~60 loci, estimating an average germline mutation rate of 1.28 × 10−8 per base pair (bp) per generation (6) [82 mutations per generation].

Larry Moran said...

Shendure and Akay multiplied the mutation rate per base pair times the amount of DNA in a human cell. Since human cells are diploid, and the genome size is 3.2 × 10^9 this works out to 6.4 × 10^9 bp.

(6.4 × 10^9) × 1.28 × 10^-8 = 82

Unknown said...

Thank you for your response. What puzzles me at this point is how 82 fixed mutations per generation is going to justify a common ancestor in 12 million years. If we take the 6.4 x 10^9 base pairs and calculate the number of differences between a man and a chimp at a very conservative 6% difference this will yield 384 million differences. Dividing that by 82 would requite ~4.6 million generations or ~90 million years. What am I doing wrong?

Unknown said...

Your estimate of difference includes differences caused by other types of genomic changes that are not caused by single nucleotide substitutions. If you exclude things like ERVs, TEs and indels (all of which happen at unrelated rates), you end up with ~35 Million differences.
Now these lead to ~420,000 generations, or 8Ma, which of course is the time total for both the lineage leading to humans and the lineage, so a divergence estimate based on this alone would be ~4Ma. But there are conserved regions for which the substitution rate is lower than the neutral rate, so you end up with the estimates that range between 5 and 7Ma.

Unknown said...

Why would we assume that evolution is dependent only upon SNS?

Unknown said...

Nobody assumes that. But we know that ERV insertions are pretty much point events, and there's not enough of them to give a reasonable rate for them. TEs tend to produce multiple copies during short bursts, but are not that active otherwise. And indels are less often neutral, which means you have to estimate mutation rates differently. The mutation rate estimates in Nature are estimates of the rate for single nucleotide substitutions. Hence you can't use them for anything but single nucleotide substitutitions.

Unknown said...

My question was how can we justify a 12 million year old common ancestor? If it takes 8 million years just to explain a 1% difference how much time is needed to justify the other 5%? If we don't have any data or the data is not verifiable then it seems presumptuous to make any kind of estimate.

Unknown said...

Another question comes to mind. If humans are 99.5% similar then this would mean that there are no changes due to SNS in ~4 million years. But I thought humans were only 200,000 years old. Additionally if there have been 16-20 changes in hominids over the last 8-12 million years it seems that SNS are far short of causing a change if in fact they can cause a change. It seems far more likely to me that SNS are not a cause of evolution. There must be something else at work.

Bill Cole said...

Hi Graham
Here is a paper that will add some additional questions. I have sent this paper to 3 Phd cellular biologists/biochemists. One thought the data was suspect the other two sent me confirming papers. DOI: 10.1126/science.1230612The Evolutionary Landscape of Alternative Splicing in Vertebrate Species

Unknown said...

I appreciate your help but I am a layman and that paper was far above my head. I failed to grasp where there was any explanation for how AS occurs. Have any tests been done to confirm that AS will alter an organism and allow it to produce fertile offspring?

Bill Cole said...

Hi Graham
I was watching you try to reconcile mutations and that is just one layer of the evolutionary story. There are differences in AS between vertebrates and if real it is an observable testable mechanism of evolution. It is new at this point so data should be taken w a grain of salt but it is seems to be an emerging part of the story. Here is an article that is in simple english. High speed sequencing should help provide a lot more confirming data going forward. news.mit.edu/2012/rna-splicing-species-difference-1220

John Harshman said...

Graham, your main problem is that you are trying to reconcile two different sorts of difference. Your figure of 6% difference includes insertions and deletions but assumes that a 1000-base deletion (a single mutation) is 1000 differences instead of 1 mutation. Also, modern humans shouldn't have 4 million years of accumulated differences between them but merely a couple hundred thousand years.

Unknown said...

Thanks Bill. Although easier to understand, I still have no clue how As occur and how this results in a new taxa. You mentioned that this could be an observable and testable means to verify evolution. I think that this is a big step because as far as I can tell macro evolution isn't observable or testable. Since these are criteria of the scientific method it is imperative that elution theory meet these criteria to be good science.

Unknown said...

John, I am going to have to disagree with you on your assessment. A difference is a difference no matter how it occurs. If a 1000 base mutation occurs that accounts for 1000 bases. If we are to establish we have a common ancestor 12 MYA then for myself I need to see how this is possible. I would want to know how I can explain that to others. The SNS mutation rate accounts for 1% of the difference. If we need to quantify other means of effecting a difference so be it. What I am hoping to learn is how we can validate a 12 MY common ancestor.

Unknown said...

John, I agree that we shouldn't have 4 million years of accumulated differences, but if there is a .5% difference between humans and by SNS alone this would take 4 million years. Don't we need to have something which can account for the .5% in 200,000 years. SNS would be expected to be 820,000 differences when we have 32 million differences (based upon a diploid number of 6.4B bases). SNS account for 2.5% of the difference. How do we quantify the other 97.5%?

Anonymous said...

Graham, you wrote, "John, I am going to have to disagree with you on your assessment. A difference is a difference no matter how it occurs."

No. You should be counting changes, not differences. A mutation that changes a single nucleotide = 1 change. A mutation that inserts 1000 base pairs = 1 change.

Also, if any of us disagree with John Harshman about population genetics, we should think really, really hard about what we think because we're probably wrong. He knows a LOT about the topic.

Faizal Ali said...

Graham, you'd be well advised to ignore what Bill Cole writes. He has no idea what he is talking about. I know from experience.

Larry has already posted calculations that show that the difference between the human and chimp genomes can have arisen entirely thru genetic drift alone (not that that is how it happened, of course):

http://sandwalk.blogspot.ca/2012/01/whats-difference-between-human-and.html

Diogenes said...

Graham is using the count of apples to estimate oranges, and justifies this by saying: "John, I am going to have to disagree with you on your assessment. A difference is a difference no matter how it occurs."

This is analogous to saying "I will use the count of apples to estimate the number of oranges, because a fruit is a fruit no matter what color it is."

Apples --> Single nucleotide changes at one base pair

Oranges --> Insertions/deletions of thousands of base pairs at a time.

Yeah, if you count each orange as being equal to thousands of "apples", then plug in the number of apples one tree can produce, you'll wind up needing many acres of apple trees to produce one gallon of orange juice.

Bill Cole said...

Hi Graham

I think that this is a big step because as far as I can tell macro evolution isn't observable or testable. Since these are criteria of the scientific method it is imperative that elution theory meet these criteria to be good science."

You are right at this point but I don't think reconciling mutation rates will fix this issue.

Diogenes said...

No Bill, the idea that hypothesized processes must be "observable and testable" is a creationist redefinition of the scientific method. In the scientific method, a theory must predict (logically entail) properties that are observable, and repeatably observable. It is not necessary that they processes (or other entities) hypothesized by the theory be observable. If the creationist redefinition of science were true, scientists could not say scientifically that a lava flow came from the nearby volcano. They could never have tested atomic theory until after atoms became visible (which they have been in the last few years.)

The creationist trick is to replace "predictions are observable, repeatable" with "hypothesized processes are observable, repeatable." This would prevent us from saying, scientifically, that plants grow from seeds, grafting, etc.

The point of science is to use invisible, hypothesized entities (which may be past events, but don' t need to be-- they could be atoms or particles or molecules or forces, etc.) to simply explain a large number of observables. Scientific explanation does not require that the atoms, molecules, forces, or past events hypothesized *necessarily* be observable themselves, only that they predict (logically entail) other things that are observable, and repeatably observable.

John Harshman said...

atoms became visible (which they have been in the last few years.)

...for a very relaxed definition of "visible". I don't actually see how inferring the existence of atoms from the patterns some electrons make when they impact (or are inferred to impact) whatever is being used to detect them is different from any other inference. You do't see the atoms; you don't see protons in a bubble chamber. And what does "see" mean anyway? Why privilege data from one narrow band of the electromagnetic spectrum over all other sorts of data? Inference is inference, and inference is how science works.

John Harshman said...

Graham, we know there are around 5 million indels separating us from chimps, which accounts for the 4% difference you claim not to be able to explain. And this is fully compatible with what we know about the rate at which indels occur. And the only way to get modern humans 0.5% different from each other is, again, to count a 1000-base indel as 1000 differences. It's a silly way to count, but even if you do, you can't complain that the rate of point mutations won't account for the difference, since the difference you count isn't made up of point mutations.

By the way, I don't actually know that much about population genetics, which I studiously avoided for 6 years in a school filled with population geneticists. I avoided Drosophila too, despite all the fly geneticists. But I know enough about both for some purposes.

Faizal Ali said...

I have already provided Bill Cole with examples, in other discussions, of how macroevolution can, and has been, tested. That he would continue to deny this indicates he is either too stupid to understand these examples or too dishonest to admit that they exist. Either way, he is showing himself unsuited to a serious discussion of the topic.

Joe Felsenstein said...

Larry, with regard to the 82 figure, I talked with Josh Akey a couple of weeks ago. He says that he regards the 82 figure as very rough and approximate.

Bill Cole said...

Hi Diogenes
Here is a paper that shows the methodology Darwin used " inference to the best explanation" This standard does not require a testable hypothesis. Interestingly enough it is the same standard that the ID guys use.www.rochester.edu/.../okasha_van_fraassen_ibe....
University of Rochester
If you research current papers by the NCSE you will see they are focusing on common ancestry vs random mutation because they can do genetic comparisons.

Larry Moran said...

Bill Cole says,

I think that this is a big step because as far as I can tell macro evolution isn't observable or testable. Since these are criteria of the scientific method it is imperative that elution theory meet these criteria to be good science."

You are confused about the meaning of "evolutionary theory." It refers to things like natural selection, random genetic drift, population genetics, speciation etc. etc.

Macroevolution has to do with the unique history of life. [see Macroevolution]. We try to explain this unique history based on what we know about evolution and evolutionary theory but evolutionary theory doesn't stand or fall based on whether dinosaurs went extinct because of an asteroid impact or a volcanic explosion.

Modern evolutionary theory would be intact even if we never heard of marsupials in Australia or the Cambrian Explosion.

Think of it this way. Physicists have pretty good theories but they cannot predict how many planets revolve around Epsilon Eridani or Tau Ceti. When we find out, we will never be able to repeat the history of formation of those planets but the theory of general relativity is still valid.

You may be thinking that the history of life is a "theory." It is not. It is a series of provisional facts, and/or hypotheses, that describe a unique series of events that occurred on a particular planet in a particular galaxy at a particular time. It's just like any history.

You may also be thinking that evolutionary theory is really good at explaining microevolution but fails at explaining macroevolution. This is partly true since there's more to macroevolution than just lots and lots of microevolution. However, to the extent that microevolution and evolutionary theory are involved there's a great deal of evidence that our understanding of these processes is correct. The best "proof" comes from molecular evolution; for example, the fact that humans and chimps evolved from a common ancestor is consistent with the differences in their genomes and evolutionary theory.

There's also no other theory or hypothesis that explain the data as well as evolution. If you have one, now is a good time to describe it.

Unknown said...

@John:"for a very relaxed definition of "visible"."
Nope. For a quite intuitive definition as well. In a magneto-optic trap you can see single atoms with the naked eye. Usually you can't see single atoms for two reasons:
a) there are a lot of other atoms around them and the resolution of our retinas is too low to distinguish them
b) The number of photons being absorbed and reemitted by a single atom is too low.

In a MOT you have single atoms in something very close to a vacuum so that you usually only get one in the trap. And the optical part of the MOT is of course pummeling that atom with loads of photons. As a result you have a glowing dot in the trap and that's a single atom, which you can see in the same way you see anything else.

John Harshman said...

That's a new one on me. I suppose that could be construed as seeing a single atom. But what that shows is that "seeing" isn't any more direct observation than any other sort of inference from data. You know you're seeing an atom only because you set up the conditions under which that spot of light represents an atom.

Bill Cole said...

Hi Larry
"The best "proof" comes from molecular evolution; for example, the fact that humans and chimps evolved from a common ancestor is consistent with the differences in their genomes and evolutionary theory."

While I agree that there is evidence of common genetics, I don't agree that there is a credible mechanism to tie it together.

You are confused about the meaning of "evolutionary theory." It refers to things like natural selection, random genetic drift, population genetics, speciation etc. etc.

I don't believe that any of these mechanisms describe how new sequences that code for proteins arrived or other regulatory sequences arrived. I think the answer is we don't know. The sequential space of the genome is too large for any of this to make sense. I know you have heard this argument many times but for me it is an absolute show stopper. Sequences are great for making variety but they are terrible for finding function through a random search. This is why computer passwords work so well...since they are sequences therefore they block random search. This is what makes the description of any evolutionary transition problematic.

Microevolutionary changes can happen through a few mutational changes but large scale evolutionary change requires new proteins driven by new sequences.

For a new theory I would start with the origin of new sequences being unknown and would try to establish testable biochemical mechanisms that are driving changes. I have proposed the spliceosome as a possible candidate yet I know you are skeptical of alternative splicing at this point. But in general, I think it is the architecture of the protein manufacturing mechanism inside the cell that enables diversity. The sequences that create the probability problem enable diversity, and is inherent in the architecture which is common with the way modern languages are structured. As an analogy the DNA creates the letters and words and alternative splicing crates the sentences. This is architected in two layers to form diversity. It will be interesting to see as new high sequencing data comes out if AS is truly playing a major role here.

Faizal Ali said...

Gee, Larry. Don't you feel fortunate that Bill Cole is here to teach you all the things he knows about molecular evolution?

Faizal Ali said...

Sequences are great for making variety but they are terrible for finding function through a random search. This is why computer passwords work so well...since they are sequences therefore they block random search. This is what makes the description of any evolutionary transition problematic.

Let's follow up on that analogy, Bill Cole. Suppose, rather than trying to find the security code for a specific account used on a specific website, you designed a computer program in which you could type any set of numbers, and then the program would scan the entire internet for some website in which a user employs that sequence as his security code and grants you access. True, this program would be useless if you wanted access to one particular person's account on a specific website. But if you just wanted access to some account,(any account) on some website (any website) how many tries do you think you would need before you accomplished this? Would this be easier than randomly finding one specific security code? Just a little easier, or a lot easier?

Now, which analogy do you think, more closely models the evolutionary process: Yours or mine?

(BTW, I see you still haven't figured out how to use HTML tags. Need some more help with that?)

Unknown said...

To bwilson295 How can we count changes if we don't know how many changes there are? How does one know if an indel is a single occurrence or multiple occurrences or a combination of an indel and a SNS? It seems to me that we can never know the answer. If we count differences then we can have a lower bound on how long it will take to accumulate the differences assuming no site changes more than once.

Unknown said...

Isn't that the purpose of dialogue? To share thoughts and learn from each other? Now I don't know if Bill is right or not but he doesn't seem to "not have a clue" as you imply. Your analogy shows virtually no function as Bill suggests sequences do not. So if you are saying that evolution shows virtually no function how do you explain the millions of functions attributed to evolution if your analogy is best?

Unknown said...

To John Harshman. I have tried to find information on how many indels there are and what is the average size of the indel. More importantly is any data on what the frequency of indels are. Let's say there are 5 million indels. If we don't know a rate at which they occur, then how can we extrapolate a time for indels to accumulate. Can you provide any links? Thanks in advance.

Unknown said...

To Diogenes: I am willing to discard SNS as it only accounts for 1% of the genome. Is there any evidence SNS can produce an evolutionary change? By this I mean has any good come out of a transcription error? If not then we can just focus on INDELs as the mechanism for evolutionary change. Presuming you agree with me then the next step for me is to quantify an average INDEL and find out how often they occur. If I can do that then I can explain to others how I know we had a common ancestor 12 MYA based upon the math.

Faizal Ali said...

@Graham Abbott

Your analogy shows virtually no function as Bill suggests sequences do not.

I don't know how you come to that conclusion. Bill was suggesting finding the correct sequence of numbers for a security code as analogous to a random set of mutations arriving at a functional protein. His analogy involves a random search attempting to find one specific security code. That's not how evolution works. My analogy is far from perfect, but it is a lot closer to the true state of affairs than his.

John Harshman said...

Graham,

The figure of 5 million indels comes from the chimp genome paper.

http://www.genome.gov/Pages/Research/DIR/Chimp_Analysis.pdf

Now, if you want to assume that a thousand contiguous gaps in the alignment between chimp and human genomes comes from a thousand different events, that's your prerogative, but it sure seems highly unparsimonious to me. Your notion that point mutations can't result in evolutionary change is quaint. Just remember that there are 7 times as many point mutations as indels between chimps and humans. The whole "percent difference" measure you want to use is meaningless.

Unknown said...

I come to that conclusion because your analogy (which is the subject) shows virtually no function. What use is it to have access to one of billions of accounts? So you asked which analogy was best, implying yours was the best. Now let's say that your analogy is best and that it represents evolution. How does this answer my question which was how do we justify as 12 MY common ancestor when the analogy of random change has virtually no function. To me this seems like we would have to push back the common ancestor as by chance virtually no function will take a long time to accumulate to millions of functions.

Faizal Ali said...

The "function" in my analogy is that the code provides access to the account. That is to say, a sequence of numbers that opens an account is functional. One that does not, has no function. It's the same "function" that Bill Cole uses in his argument. If you reject my argument, then you must reject his as well.

Anonymous said...

Graham, hypothetical example. Two DNA sequences are compared. One has a 1000-base-pair long sequence that is not present in the other, a 1000 bp indel. Let's say the real history of this indel is that there was an 800 bp insertion, a 200 bp deletion, and a second, 400 bp insertion, plus a couple single nucleotide changes, for a total of 5 changes. But we don't know that. From our observation of the 1000 bp indel, we can score this as 1 change, or 1000 individual indels.

Is 1 or 1000 a better approximation the actual 5 changes?

Anonymous said...

What is SNS the abbreviation for, in this context? My attempts to find it have led me to an interesting hypothesis on the early evolution of the genetic code, but I don't think that's it.

Dazz said...

The thing, Graham, is that it's not the first time it's been explained to Bill why he is wrong. He just keeps ignoring that and keeps using these wrong analogies. Evidence that the known mechanisms are enough to explain macroevolutionary changes has been presented. He just turns a deaf ear everytime

John Harshman said...

SNS= typo for SNP?

Unknown said...

Thanks for you help. I haven't had time to read the whole paper but I did find what I was looking for. What I gleaned from that is that the average insertion would be 6 bp and that if happening randomly the average time would be 2.4 years. What I am interested in is knowing if we are still seeing this pattern? Is there any reason we wouldn't expect to see insertions still occurring?
.
Remember my objective is to show how we can have a common ancestor in 12 million years. For me I would expect that there would be some kind of consistency (within reason) to the insertions happening. If we can't establish that they are happening now how do we really know that they did in the past?

Unknown said...

I didn't say it had no function. I said it has virtually no function. That a code can gain access to presumably only one of millions of accounts is functional but how relevant is it? Since I haven't heard from Bill on this I don't know what is answer will be. It seems rather useless to continue, if I agree that his analogy is virtually non functional also (which I have no problem with) I am no closer to being able to explain to people how we can have a common ancestor in 12 MY.

Larry Moran said...

Graham Abbott says,

I have tried to find information on how many indels there are and what is the average size of the indel.

The 1000 Genomes Project gives us some idea of the different kinds of mutations that we can expect [A global reference for human genetic variation]. The average human genome (yours and mine) differs from the standard reference genome at about 5 million sites.

The breakdown is ...

single nucleotide changes (SNPs): about 4 million
short indels: about 600,000 (15% of SNPs)

Together, these account for 99.9% of the variants.

Larger structural variants occur at about 2,300 sites and they can be classified as follows ...

about 1,000 large deletions/insertions
about 160 copy-number variants
about 915 Alu insertions
about 128 L1 insertions
about 51 SVA insertions
about 4 NUMTs
and about 10 inversions

The total number of base pairs involved in these structural variants is about 20 million bp. This is mostly due to very large deletions covering >10,000 bp. Many of these deletions are "unique" meaning they are only found in a single individual out of 2,504 people screened.

It would be very strange to argue that such a deletion is due to anything other than a single event and almost as strange to argue that most other deletions/insertions were anything but single mutations that have no effect on fitness.

HTH HAND

Faizal Ali said...

If an enzyme can only catalyze one reaction, does it serve no function?

Have a look thru this article on inherited metabolic disorders. Then tell me if you still believe that an enzyme that serves only one function is of no relevance.

http://www.webmd.com/a-to-z-guides/inherited-metabolic-disorder-types-and-treatments

Unknown said...

John Harshman: Your notion that point mutations can't result in evolutionary change is quaint.
.
As I have admitted, I am a layman. Is there any evidence a SNS can produce a change in taxa or even a benefit? By all estimates there are 16+ million differences in the human genome and we are all human. If we look at the Lenski experiment, he estimated millions of mutations occurred but <100 became fixed within the population. It wasn't until a frame shift mutation occurred that anything of significance occurred and after 60,000 generations we still have an e.coli. So if I run into a creationist, what do I have to show him except opinion? He is going to tell me that all we are seeing is a degeneration of the genome and what will I have to answer him?

Unknown said...

I don't know if Bill is right or wrong. I have expressed to him that I can't see how AS occurs and if it can lead to a macro evolutionary change. He hasn't answered me. I don't see anything on this thread which disproves AS as a mechanism for change. I don't even see him stating it is the only means of change. Anything said in the past or on other blogs I know nothing about. I prefer to see evidence Bill is incorrect. I am not very partial to opinions unless they are testable.

Anonymous said...

Graham, yesterday I spent some time debating whether to get a poinsettia with solid red bracts or one with irregular cream-colored patches on the red bracts. The latter pattern is often caused by sequences called transposons jumping in and out a gene within the time it takes for a bract to grow. In other words, these are self-splicing indels active right now.

Transposons are an extreme case, but comparing human sequences we do see evidence of indels happen now or in the very recent past.

Also, biochemists know enough about the way DNA functions to understand how indels happen and why they are inevitable.

Faizal Ali said...

It might help you to know, Graham, that this is the article that Bill Cole relies upon to support his claim that it is not possible for evolutionary processes to navigate "sequence space" to arrive at functional sequences:

https://aghunt.wordpress.com/2008/12/26/axe-2004-and-the-evolution-of-enzyme-function/

As you can see, the article demonstrates just the opposite of what Bill claims. That does not seem to bother him at all.

If you insist on making up your own mind, you can feel free to read thru his comments on the threads below. However, I would advise simply ignoring him and concentrating on the posts from Larry Moran, John Harshman and the other members here who are not ignorant, stupid and dishonest.

http://sandwalk.blogspot.ca/2015/10/intelligent-design-needs-to-clean-up.html

http://sandwalk.blogspot.ca/2015/11/the-birth-and-death-of-salmon-genes.html

Anonymous said...

I'm afraid I have to ask again. What's AS the abbreviation for in this context?

Dazz said...

Alternative Splicing I guess

Unknown said...

I would agree that a deletion of 10Kbp would be highly unlikely to have occurred without a single or multiple large events. But 20 million differences out of 32 million (or less) differences is a high percentage and yet this isn't enough to effect a macro change in humans. Furthermore 32 million (or less) differences over 12 million years is 2.6 base changes per year. Aren't we seeing more than this in humans?
.
As to SNS and fitness. I don't see people becoming more fit. I see an increase of disease due to mutations and by the time the diseases affect us a new generation is born with more mutations so NS is ineffective except in the minority of cases. So I agree with you. SNS can affect fitness. My statement was I don't see them leading to a macro change as there aren't enough of them to evolve into another organism in the time allowed.

Faizal Ali said...

So, Graham, you don't think humans and chimps are different organisms from one anothetr? Or are you saying that the differences in their morphology are due to something other than genetics? If so, what might that be?

Faizal Ali said...

BTW, Graham, what "increase of disease due to mutations" are you seeing?

John Harshman said...

Graham,

What's this about 12 million years? Common ancestor of what? The common ancestor of humans and chimps was somewhere around 5-8 million years ago.

As for point mutations, most do not become fixed. Bacteria and humans are quite different because bacteria have very little junk DNA, while the human genome is 90% junk. Most of those 40 million fixations (35 million point mutations, 5 million indels) between humans and chimps are neutral. We don't know how many in the human lineage are subject to selection, but I'd imagine a few thousand at most. Of those few thousand, we don't know what most of them are or what they did. Nevertheless, we do know that many point mutations can result in changes, though frequently they come to our attention because the changes are deleterious. Then again, what's bad in one environment can often be good in another. A change of one amino acid in a protein can alter its function. A change of one nucleotide in a promoter region can strengthen or weaken a transcription factor binding site, or cause a new one to appear, with effects on morphology. This is elementary genetics.

As for whether indels happen now, there are plenty of studies of the human population showing indel polymorphisms.

Piotr Gąsiorowski said...

I don't see people becoming more fit.

Fitter than whom? How can you determine fitness by casual observation? Fitness is relative, not absolute.

Larry Moran said...

Graham Abbott says,

As I have admitted, I am a layman.

You claim that you don't understand anything about evolution. That's fine. There's nothing wrong with that. Many people here will be happy to help you.

But then you say ...

If we look at the Lenski experiment, he estimated millions of mutations occurred but <100 became fixed within the population. It wasn't until a frame shift mutation occurred that anything of significance occurred and after 60,000 generations we still have an e.coli.

That doesn't sound like someone with an open mind who's willing to learn. It sound's like you've already studied Lenski's papers and you're convinced that he was trying to show the evolution of a new species.

You can't have it both ways. Either you don't understand and you're willing to learn or you do understand and you reject evolution.

Please make up your mind whether you want to play dumb or play the fool.

Faizal Ali said...

@Graham Abbott

Is there any evidence a SNS can produce a change in taxa or even a benefit? By all estimates there are 16+ million differences in the human genome and we are all human.

Is your understanding of evolutionary theory that a SNP (presuming that's what what you mean by "SNS") can, by itself, produce a new taxonomic group? That's really weird.

Also, how do you think we would know if we had witnessed the emergence of a new taxon as it happened? Do you think the children of the last common ancestor of chimps and humans were of different taxa from each other?

Piotr Gąsiorowski said...

P.S. I mean, the kind of fitness I think you are talking about is relative. But then it's hard to discuss "SNS and fitness" if nobody knows what you mean by "SNS".

Unknown said...

Lutesuite: So, Graham, you don't think humans and chimps are different organisms from one another?
.
Do you really have to ask that question? All I said was I can't see how SNS can lead to a change of taxa. I am asking for help to see how we can have a common ancestor 12 MYA when the rate of mutation is so low. It also doesn't seem to me the math adds up. If we have 82 SNS mutations per generation and 52 insert bps per generation over 200,000 years we should have 1.34 million differences accumulate. But a .5% difference among humans is 32 million differences. So what accounts for the other 30+ million differences?

Unknown said...

lutesuiteWednesday, November 25, 2015 3:59:00 PM
BTW, Graham, what "increase of disease due to mutations" are you seeing?
.
I see a rise in cancers, heart disease, eye disease, allergies, autism. I believe I read that mutations lead to an increase in diseases. If you want me to find out where I read that I will have to google it.

Unknown said...

Piotr GąsiorowskiWednesday, November 25, 2015 9:09:00 PM
P.S. I mean, the kind of fitness I think you are talking about is relative. But then it's hard to discuss "SNS and fitness" if nobody knows what you mean by "SNS".
.
SNS is an abbreviation for single nucleotide substitution.

Piotr Gąsiorowski said...

If people live longer, away from war zones, in decent material conditions, and with good medical care, they'll die of cancer rather than a trivial infection, famine, or a spear in their guts. You don't expect them to be immortal, do you? If by "fitness" you don't mean the relative fitness of one genotype versus another used as a reference, but absolute fitness as a function of viability and fertility, it has to be larger than 1 if the total human population keeps increasing.

Faizal Ali said...

SNS is an abbreviation for single nucleotide substitution.

Great, thanks. So the answer to this:

All I said was I can't see how SNS can lead to a change of taxa.

Is that they don't. Not over the timescales that the fossil record and molecular genetics demonstrate it to have happened.

If we have 82 SNS mutations per generation and 52 insert bps per generation over 200,000 years we should have 1.34 million differences accumulate. But a .5% difference among humans is 32 million differences. So what accounts for the other 30+ million differences?

Where does 200,000 years come in?

Faizal Ali said...

I see a rise in cancers, heart disease, eye disease, allergies, autism.

What leads you to conclude that those increases are due to random mutations? e.g. You think the increase in heart disease is just due to genetics? Diet and lifestyle plays no role?

Faizal Ali said...

Anyway, if that's what you're using to define "fitness", just compare the average human life expectancy today to what it was 100 years ago. Then tell us whether we are not becoming more "fit". (That's not to obscure the fact that you are using the term "fitness" improperly as it pertains to evolutionary theory. Just to show that, even taken on your own terms, your argument is nonsense.)

You've indicated you're aware of Lenski's long-term evolution experiment. So why are you still questioning whether random mutations can lead to adaptive traits?

John Harshman said...

Graham,

You have two central difficulties in calculation: 1) You are confusing mutation with fixation. There are billions of mutations in each human generation, but the number of fixations (according to neutral theory) is equal only to the number of mutations in a single individual; 2) You persist, even when informed of it, in equating number of differences with number of mutations, i.e. assuming that the 1000-base indel we've been talking about is 1000 mutations, which is how you get that .5% difference.

Unknown said...

I think your analogy was bad. But if you wish to press it, I will give you an example. If humans were by mutation to acquire a random enzyme that could metabolize nylon, it would have a function but it would serve virtually no function in humans. The difference as I see it is Bill searched for a specific result and you searched for any result.

Unknown said...

lutesuiteThursday, November 26, 2015 11:28:00 AM
I see a rise in cancers, heart disease, eye disease, allergies, autism.

What leads you to conclude that those increases are due to random mutations? e.g. You think the increase in heart disease is just due to genetics? Diet and lifestyle plays no role?
.
Of course diet and lifestyle play a role. What makes you think that genetics play NO role?
.
Remember the question was what increases due to mutations are you seeing. You didn't limit the reason to only mutation. So yes it can be a combination.

Unknown said...

lutesuiteThursday, November 26, 2015 11:25:00 AM
Where does 200,000 years come in?
.
The timeline for humans as we know them.

Unknown said...

Piotr GąsiorowskiThursday, November 26, 2015 11:23:00 AM
If people live longer, away from war zones, in decent material conditions, and with good medical care, they'll die of cancer rather than a trivial infection, famine, or a spear in their guts.
.
Yes longevity is a factor in cancer rates but only 2/3 of the increase in cancer is due to longevity. This in spite of heightened cancer awareness and measures taken to prevent cancers like asbestos banning and reduced smoking.
.
What is your point? I didn't come here to argue over what the detrimental effects of mutations are. I came here to be able to show how macro evolution is working.

Unknown said...

lutesuiteThursday, November 26, 2015 11:32:00 AM
You've indicated you're aware of Lenski's long-term evolution experiment. So why are you still questioning whether random mutations can lead to adaptive traits?
.
Because long term adaptive traits are not going to cause a single celled organism to become multi-celled nor did it change the taxa of the e. coli.
.
Secondly there is a time issue. Given enough time anything may be possible. We are working within a 12 MY window.
.
60,000 generations in human time is 1.2 million years with only one significant change. I don't see how that is comparable to multiple changes of taxa in hominids.

Unknown said...

John Harshman Thursday, November 26, 2015 1:09:00 PM
Graham,

You have two central difficulties in calculation: 1) You are confusing mutation with fixation. There are billions of mutations in each human generation, but the number of fixations (according to neutral theory) is equal only to the number of mutations in a single individual; 2) You persist, even when informed of it, in equating number of differences with number of mutations, i.e. assuming that the 1000-base indel we've been talking about is 1000 mutations, which is how you get that .5% difference.
.
I don't think I am confusing mutation with fixation because the figures quoted by Dr. Moran referred to fixed mutations. As you have pointed out and so have I, billions of mutations are occurring. So using a number like 82 per generation cannot mean any mutation. As far as I can see only fixed mutations are critical to an evolutionary change. If I am wrong I am willing to see it, but you are asserting an error instead of demonstrating an error. I will give you an example. Earlier I made a mistake in not accounting that both men and chimps are mutating and this would lessen the requirement on each line to ~1/2 of the needed changes. But no one caught this blunder and pointed it out. That to me is far better than making an assertion. So if Dr. Moran wasn't talking about fixed mutations in base pairs, I didn't catch that.
.
I don't believe I was the one who started the mutation/difference argument. I stated that if there is a 6% difference then this would require (if we use a diploid number) 384 million differences between a man and a chimp. I never made any claim as to how many mutations this would require. It is possible that one mutation changed the entire genome. The question is how can it be demonstrated that the number of mutations required to explain the difference is occurring. That is why I asked for any data on inserts especially data showing what rate it is occurring.

Piotr Gąsiorowski said...

What is your point? I didn't come here to argue over what the detrimental effects of mutations are.

Well, it was you who started talking about the deleterious effect of mutations.

I came here to be able to show how macro evolution is working.

To show or to learn? You seem to ignore what more knowledgeable people are telling you.

Unknown said...

bwilson295Wednesday, November 25, 2015 1:32:00 PM
Graham, hypothetical example. Two DNA sequences are compared. One has a 1000-base-pair long sequence that is not present in the other, a 1000 bp indel. Let's say the real history of this indel is that there was an 800 bp insertion, a 200 bp deletion, and a second, 400 bp insertion, plus a couple single nucleotide changes, for a total of 5 changes. But we don't know that. From our observation of the 1000 bp indel, we can score this as 1 change, or 1000 individual indels.

Is 1 or 1000 a better approximation the actual 5 changes?

Graham: I would prefer to use 1000 differences because as you have pointed out there is no way to tell how many events were required to effect the difference. So why not use what we know instead of what we can't know?
.
To me the heart of the matter is not what we don't know. The heart of the matter is that there should be evidence that indels are occurring so we can get a better feel for how long it takes to effect a change and what will effect a change. I would think with all the genetic engineering going on, we would have better data on what can lead to a change of taxa and how long it takes for that to occur naturally.

Dazz said...

Graham: I would prefer to use 1000 differences because as you have pointed out there is no way to tell how many events were required to effect the difference. So why not use what we know instead of what we can't know?

Yeah, because having 1000 insertions of a single bp right in the same exact place is so much more probable, right?

Creationist logic FTW

Unknown said...

lutesuiteWednesday, November 25, 2015 8:45:00 PM
Is your understanding of evolutionary theory that a SNP (presuming that's what what you mean by "SNS") can, by itself, produce a new taxonomic group? That's really weird.

Graham: If you are going to continue to misrepresent what I have said (when in fact I have said the opposite) then I will decline to interact with you. I have come here to ask for the evidence. What I get instead are attacks on everything I say including the opposite of what I have said. So far you haven't helped me understand anything.

Unknown said...

lutesuiteWednesday, November 25, 2015 8:45:00 PM
Also, how do you think we would know if we had witnessed the emergence of a new taxon as it happened? Do you think the children of the last common ancestor of chimps and humans were of different taxa from each other?
.
Graham: Firstltly this is not my claim. This is what you find on the internet, 20 or so changes of taxa over the last 12 million years. If there has been no change of taxa then why do we classify fossils as different taxa?
.
If there are no change of taxa due to mutation then what is the explanation for the billions of taxa there have been?

Piotr Gąsiorowski said...

Graham: Firstltly this is not my claim. This is what you find on the internet, 20 or so changes of taxa over the last 12 million years. If there has been no change of taxa then why do we classify fossils as different taxa?

Whe the heck is it supposed to mean? What is "a change of taxa"? Speciation? Are you trying to say that there have been 20 or so speciation events in the last 12 million years? If it's something you found on the Internet, how about providing a link?

Dazz said...

If there has been no change of taxa then why do we classify fossils as different taxa?

What he's saying is that the change is gradual and wouldn't be obvious from one generation to the next... obviously. It's you who's either misunderstanding or misrepresenting others. There's plenty evidence for speciation

Faizal Ali said...

Your ignorance is confusing you, Graham. You're trying to perform calculations based on factors you do not understand and so are coming up with meaningless results.

I thought I'd already linked this for you, but I guess I hadn't. Larry has already done the relevant calculations, and shown that the observied mutation and fixation rates are easily consistent with humans and chimps arising from a common ancestor, just thru random drift alone. Natural selection is not even needed (though, of course, it played a major role):

http://sandwalk.blogspot.ca/2012/01/whats-difference-between-human-and.html

If there's anything about that you don't understand, I'm sure Larry will help you out.

ealloc said...

Graham is confusing two different questions: 1. how to get the chimp-human divergence time 2. Given the divergence time, how to understand their overall genetic diffence. The point he is not hearing is that you can calculate the divergence time using SNPs even while ignoring other types of genetic change.

Analogy for Graham: Say you photocopy a sheet of text, then photocopy the photocopy, and so on. Specs of dust randomly cause letters to be photocopied incorrectly (1 in 1000 letters on average, each time). Also, sometimes you misalign the page so that some lines of text gets chopped off the top or bottom. Some time later, of the original 10,000 letters, 1000 were lost at the top and bottom (9000 remain, a 10% difference). However, you can easily line up those 9000 letters with the middle region of the original text, and see that 180 letters there are corrupted (2% of the 9000 letters are different).

Based on this info, estimate how many photocopies separate the old text from the new text. (Hint: solve 0.98 = 0.999^n). Note that this calculation only depends on the 2% difference due to "point" changes, not the overall 10% difference due to deletions.

Faizal Ali said...

Because long term adaptive traits are not going to cause a single celled organism to become multi-celled

Why not? Show your work.

...nor did it change the taxa of the e. coli.

You expected a taxonomic change to occur in a miniscule sample of organisms in a single lab, in a miniscule period of a few years? You really don't understand much about evolution, do you?

60,000 generations in human time is 1.2 million years with only one significant change.

There was not only one change. Sorry, I thought you had understood the experiment. My mistake. Here, read this:

https://en.wikipedia.org/wiki/E._coli_long-term_evolution_experiment.

Unknown said...

Laurence A. Moran Wednesday, November 25, 2015 8:38:00 PM
That doesn't sound like someone with an open mind who's willing to learn. It sound's like you've already studied Lenski's papers and you're convinced that he was trying to show the evolution of a new species.
>
Dr. Moran even a layman can understand that if no change of taxa occurred after 60,000 generations in not a natural environment but a man made environment to induce a change that the time in generations produced no change in taxa. That is my conclusion, not any conclusion Dr. Lenski made to my knowledge. I never said Dr. Lenski was attempting to produce a change in taxa. I presume he was trying to effect a change in the ability to metabolise citrate in an aerobic environment. If he was trying to change a taxa he didn't admit any success or failure.
.
Have you written a paper on how we know that changes of taxa (assuming you believe this happened) were the result of mutation? I will be happy to read it. So far the help you say is available has not been forthcoming. Ironically enough the best help I have gotten has come from Bill Cole who referred to Alternate Splicing as a cause, but the paper he referred me to I couldn't understand how alternate splicing causes a change in taxa.
.
Lastly I claimed I was a layman. I never stated I understood nothing about evolution. What I don't understand is why there doesn't seem to be any evidence of macro evolution in any of the 8 + million species on the planet. Assuming it takes 1 million years for enough mutations to accumulate to effect a change in taxa wouldn't we expect 8 changes of taxa per year if this is happening randomly?
.
Lastly, I don't see how asking for some help in increasing my understanding of evolution theory makes me dumb or a fool.

Larry Moran said...

Graham Abbott says,

Lastly I claimed I was a layman. I never stated I understood nothing about evolution.

Maybe you never said that but it's clear from you comments that you know very little about evolution.

What I don't understand is why there doesn't seem to be any evidence of macro evolution in any of the 8 + million species on the planet. Assuming it takes 1 million years for enough mutations to accumulate to effect a change in taxa wouldn't we expect 8 changes of taxa per year if this is happening randomly?

Do you honestly believe that every species on the planet is gradually undergoing speciation so that in one million years there will be 16 million species?

Does that mean that one million years ago there were only 4 million species?

That means that 16 million years ago there were only 100 species according to your understanding of how evolution works. (That's a rough calculation ... you get the idea.)

Now do you see why we think you don't understand evolution?

Faizal Ali said...

Why would that not serve a function? It would mean we could eat nylon and, moreover, solve the problem of disposing of nylon products.

Faizal Ali said...

The timeline for humans as we know them.

Why are you using that? That makes no sense.

Jmac said...

Looks like your blog is dying Larry.... I think you have just banned your only hope for the revival of your sad-walk blog-Robert Byers. If you both took your meds and at the same you could possibly have a meaningful conversation about your delusions, though so far apart and, so different.

BTW: I've always wondered what or who drives some people to deny all logic and lie to themselves....It was going to be my thesis but I was warned that it would not fly very well for no particular reason whatsoever....

Faizal Ali said...

Of course diet and lifestyle play a role. What makes you think that genetics play NO role?

Don't misrepresent me. I didn't say genetics play no role. You are claiming that, e.g., the rise in incidence of heart disease is due to deleterious mutations arising in the human genome over the recent few years. Show the evidence to support this. (Hint: You won't find it, because your claim is false.)

Faizal Ali said...

lutesuiteWednesday, November 25, 2015 8:45:00 PM
Is your understanding of evolutionary theory that a SNP (presuming that's what what you mean by "SNS") can, by itself, produce a new taxonomic group? That's really weird.

Graham: If you are going to continue to misrepresent what I have said (when in fact I have said the opposite) then I will decline to interact with you.


Sorry, I misworded that. What I should have asked is "Is your understanding of evolutionary theory such that it would predict that a SNP, by itself, would produce a new taxonomic group?" Because if you do, then you don't understand evolutionary theory, and are trying to disprove something that no one believes to be true. Strawmanning, IOW.

I'm not just attacking you. I'm also genuinely trying to find out what you're trying to say, because it's not really at all clear.

Let's try this. You keep asking about the evidence to show a new taxon arising thru mutations. Let's assume the standard evolutionary understanding of how speciation occurs is true. What kind of evidence of a new taxon arising would expect to have been observed in the 150 years that the theory of evolution has existed?

Jmac said...

What's the proof that diet and lifestyle cause or prevent autism, schizophrenia, bipolar disorder, skin cancer, mania, adjustment disorder, learning disability, ADHD?

Faizal Ali said...

Well, some forms of skin cancer are clearly related to UV ray exposure, and adjustment disorder is, by definition, related to external stressors. As for the others, I`m not aware of diet being demonstrated to be a causal factor, even though stressors clearly can influence the course of those disorders.

Why do you ask?

Unknown said...

lutesuiteThursday, November 26, 2015 5:56:00 PM

Sorry, I misworded that. What I should have asked is "Is your understanding of evolutionary theory such that it would predict that a SNP, by itself, would produce a new taxonomic group?" Because if you do, then you don't understand evolutionary theory, and are trying to disprove something that no one believes to be true. Strawmanning, IOW.
.
I entered this thread after being directed to this site for data on mutation rates. I didn't understand how Dr. Moran arrived at the figure of 82 based upon the rate listed. He explained that I was only using a haploid number so then I understood how he arrived at 82. I then asked how 82 bp SNPs can be all there is as this is not enough. I stated that this CANNOT POSSIBLY be enough to explain evolution. Here's evidence of that:
Graham Abbott Monday, November 23, 2015 3:15:00 PM
Why would we assume that evolution is dependent only upon SNS?
Graham Abbott Monday, November 23, 2015 5:33:00 PM
Additionally if there have been 16-20 changes in hominids over the last 8-12 million years it seems that SNS are far short of causing a change if in fact they can cause a change. It seems far more likely to me that SNS are not a cause of evolution. There must be something else at work.
.
So it baffles me why you think that I believe SNPs are the cause of evolution.
.
Additionally I would doubt that one SNP could cause a change of taxa. I would take considerably more as humans probably have millions (as I have also pointed out) and no change has occurred.

lutesuite Thursday, November 26, 2015 5:56:00 PM
Let's try this. You keep asking about the evidence to show a new taxon arising thru mutations. Let's assume the standard evolutionary understanding of how speciation occurs is true. What kind of evidence of a new taxon arising would expect to have been observed in the 150 years that the theory of evolution has existed?
.
As I pointed out if it takes 1 million years to change a taxa (which is less than what is told us in hominid evolution) then we should expect 8 new species per year based upon 8 million species existing. Now it is a given that many of these new species might not be observed, but in 150 years I think the chances of none being observed is not reasonable.
.
Furthermore speciation is really not enough to explain diversity. We have many taxon greater than species which can't be explained by speciation. So really we should be talking of higher level changes. If I go to https://www.google.com/search?q=hominid+evolution&safe=strict&tbm=isch&imgil=TGn3YIGA3aTmLM%253A%253BmIhZIFZvbV77mM%253Bhttp%25253A%25252F%25252Fvisual.ly%25252Ftimeline-hominid-evolution&source=iu&pf=m&fir=TGn3YIGA3aTmLM%253A%252CmIhZIFZvbV77mM%252C_&biw=1360&bih=600&usg=__kIGBA1G1MtHsatopLf-lL4BRf5A%3D&ved=0ahUKEwjquIuF97DJAhUD6SYKHcAkD8EQyjcIRg&ei=PHhYVuqqBYPSmwHAybyIDA#imgrc=TGn3YIGA3aTmLM%3A&usg=__kIGBA1G1MtHsatopLf-lL4BRf5A%3D
we will see 6 changes in genera in 7 million years. Since humans reproduce slower than most species then I think it reasonable to conclude that other species will evolve quicker than humans as evolution should be measured in generations not years. Since the average is slightly over 1 million years per hominid genus, I think it reasonable to use 1 million years as an expectation for all genera.

Unknown said...

Laurence A. MoranThursday, November 26, 2015 5:21:00 PM

Do you honestly believe that every species on the planet is gradually undergoing speciation so that in one million years there will be 16 million species?
.
No, because the extinction rate is about 800 species per year if not more. In a million years I expect there will be very few species unless something changes.
(That's a rough calculation ... you get the idea.)

Unknown said...

Piotr GąsiorowskiThursday, November 26, 2015 3:24:00 PM
Are you trying to say that there have been 20 or so speciation events in the last 12 million years? If it's something you found on the Internet, how about providing a link?
.
Graham: I'm not trying to say anything but this is what I saw on the internet. The speciation events for hominids are listed at:
https://www.google.com/search?q=hominid+evolution&safe=strict&tbm=isch&imgil=TGn3YIGA3aTmLM%253A%253BmIhZIFZvbV77mM%253Bhttp%25253A%25252F%25252Fvisual.ly%25252Ftimeline-hominid-evolution&source=iu&pf=m&fir=TGn3YIGA3aTmLM%253A%252CmIhZIFZvbV77mM%252C_&biw=1360&bih=600&usg=__kIGBA1G1MtHsatopLf-lL4BRf5A%3D&ved=0ahUKEwjquIuF97DJAhUD6SYKHcAkD8EQyjcIRg&ei=PHhYVuqqBYPSmwHAybyIDA#imgrc=TGn3YIGA3aTmLM%3A&usg=__kIGBA1G1MtHsatopLf-lL4BRf5A%3D

Unknown said...

Do you really think if people could metabolize nylon that they would? I can think of many things people can metabolize and won't.

Unknown said...

Piotr GąsiorowskiThursday, November 26, 2015 2:35:00 PM
To show or to learn? You seem to ignore what more knowledgeable people are telling you.
.
I will be happy to listen to you answer the question on how I can show the evolutionary history of humans from the data. Please be prepared to show the math as I am going to want to see it. So as an example if I bid on a construction contract and I was asked to show how I was going to complete the task and I couldn't show that I had the manpower and equipment to do the job, I am not going to win the contract. So just as saying I am going to excavate and build is not enough to satisfy a client, it will not satisfy me. If you can't provide what I ask, you can just say so and say that is unreasonable.

Dazz said...

You've already been provided with the math, why are you ignoring it? Lutesuite linked you to the relevant thread, let me repost it:

What's the Difference Between a Human and Chimpanzee?

In an evolving population the rate of fixation of neutral alleles is equal to the mutation rate [Random Genetic Drift and Population Size]. How many mutations would we expect in the human lineage since it diverged from a common ancestor with chimpanzees if all of the fixed alleles were neutral? The two species diverged about 5 million years ago. The average generation time in the human lineage is about ten years, so that means 500,000 generations. If the rate of mutation is about 100 new mutations per generation, then we would expect to see about 50 million new mutations in the human lineage. The actual number is about 22.5 million (half of 45 million). We're certainly in the right ballpark.

Unknown said...

The reason I am not hearing that we can calculate the divergence time with SNPs while ignoring other types of genetic change is that I didn't see anyone make the claim or show how we establish the time needed. To do that we would need to know how far we need to go and how fast we are going there. If we use the 82 bp change per generation as the speed, what is the distance?

Faizal Ali said...

So it baffles me why you think that I believe SNPs are the cause of evolution.

Because you keep writing things like this:

Is there any evidence a SNS (sic) can produce a change in taxa or even a benefit? By all estimates there are 16+ million differences in the human genome and we are all human.

If you already understand that evolutionary theory does not state that a single SNP can produce a speciation event, then why do you keep asking for evidence that it can?

Anyway, you sure can spew a lot of words without addressing the question you were asked. Here, again, is my question

What kind of evidence of a new taxon arising would expect to have been observed in the 150 years that the theory of evolution has existed?

You continue to insist that we should be seeing 8 new species arising a year, and that's so stupid it's almost cute. As Larry says, it shows how very little you understand about evolution. But, even so, I'm interested in what you think 8 new species arising in a year would look like. Do you expect to see fish giving birth to monkeys, or something?

Faizal Ali said...

Way to miss the point, Graham.

The difference as I see it is Bill searched for a specific result and you searched for any result.

Precisely. And that's how evolution works: Not by searching for a specific result, but searching (or, more, accurately, by accidentally arriving at) any result. That's why Bill's analogy is a false one.

Faizal Ali said...

Those numbers don't give Graham the answer he wants. So he'd rather make up his own, then ask us inane questions based on them.

Faizal Ali said...

I'm not trying to say anything but this is what I saw on the internet. The speciation events for hominids are listed at (link snipped)

Do you think those are all speciation events illustrated there?

Piotr Gąsiorowski said...

Since speciation does not happen overnight (or in a year) and is a process rather than a sudden event, there are quite certainly tens of thousands of ongoing speciations happening right now, some just starting, some advancing in their mid course, and some almost complete. Examples are easy to find.

Unknown said...

lutesuite Thursday, November 26, 2015 3:55:00 PM
Graham writes:
Because long term adaptive traits are not going to cause a single celled organism to become multi-celled

Lutesuite writes: Why not? Show your work.
.
I will admit I cannot show the work to support my position. Now I will ask you to show the work that it can.
.
Lutesuite writes: You expected a taxonomic change to occur in a miniscule sample of organisms in a single lab, in a miniscule period of a few years? You really don't understand much about evolution, do you?
.
Graham: I never said I expected anything. I merely noted that thru 60,000 generations we haven't seen one. As I noted this is 1.2 million years in human terms. Now I realize that we can't directly compare the two but there should be some similarity. How many changes of taxa occurred in the last 1.2 million years in hominids?
.
So what happens if after 600,000 generations of e.coli we have no change in taxa? Is that going to change anything you believe about mutations causing a change in taxa?
.
Lastly I am going to request you be less condescending. It is poisoning the well.

Faizal Ali said...

I will admit I cannot show the work to support my position.

Ah. So you're just making shit up. I'm glad we're finally clear on that.

Now I will ask you to show the work that it can.

It's actually been observed in the laboratory:

http://www.wired.com/2012/01/evolution-of-multicellularity/

I never said I expected anything. I merely noted that thru 60,000 generations we haven't seen one.

Are you sure about that? How do you know?


Piotr Gąsiorowski said...

I see candidates for the position of village idiot are already queuing to be inteviewed.

Unknown said...

" If you exclude things like ERVs, TEs and indels (all of which happen at unrelated rates), you end up with ~35 Million differences."
That's what I posted a week ago. We can identify which changes are changes of one base into another one and which are not. That's something we get when we have aligned the sequences. We use these because the rates for other types of change are nowhere near as constant. We know TEs tend to have short periods where they are very active and then they are not (how do we know this? Each copy of a TE in a genome of course is also subject to point mutations. So we can align the copies of that TE from a particular genome with each other and work out when they diverged from one another).

Unknown said...

DazzFriday, November 27, 2015 11:39:00 AM
You've already been provided with the math, why are you ignoring it? Lutesuite linked you to the relevant thread, let me repost it:

What's the Difference Between a Human and Chimpanzee?
.
I have read the article and these are my impressions.
1. The figure of 45 million differences is a haploid number. It seems to me that the number is 96 million differences. This is off by a factor of over 2X.
2. The average generational time of ten years is <1/2 of chimp or human generations today. This is off by a factor of over 2X.
3. The mutation rate was given as 100 but since Roach could only confirm 28 and even if we double this as he doesn't indicate if he was testing haploid or diploid, this is off by a factor of a little less than 2X.
4. We are assuming that mutation rates are constant. I think there are reasons to say it is increasing and we need more long term data, but I am willing to use what we have and I don't want to argue over what we don't know.

Conclusion: I can see that I disagree at a factor of 8 with the article. I don't think it is proper to use a haploid number as a difference and then use a diploid number as the speed of mutation when calculating a time for divergence.
So as I see it the number of SNPs required in 5 million years is 192 per generation. So assuming that 96 are occurring per generation would allow the potential for a 10 million year divergence which is within the timeline I asked for.

Now I would like to know why Roach could only validate 28 mutations, if this a haploid or diploid number and were these other estimates subject to the same "correction" as Roach did?
.
Lastly since I was accused of saying that SNPs can cause a change of taxa, I don't see how this answers my question on how do we know we had a common ancestor whatever number of years ago if SNPs don't lead to a change in taxa?

John Harshman said...

The figure of 45 million differences is a haploid number. It seems to me that the number is 96 million differences. This is off by a factor of over 2X.

Where do you get 96 million? And again you seem to be confusing differences with fixed mutations. Why count a single mutation twice just because it's present in two copies in a diploid? It's still the result of one event.

Once again. 35 million SNPS in 10 million years (remembering that some have happened in the chimp lineage and some in the human) comes to 3.5 per year. If a generation is 20 years, that's 70 per generation. How is that a problem?

how do we know we had a common ancestor whatever number of years ago if SNPs don't lead to a change in taxa?

One problem is that nobody is clear what you mean by "a change in taxa". Are you talking about speciation? Significant morphological change? Dogs giving birth to cats?

Notice, by the way, how you can mark quotes with italics. This would make your posts clearer. Open italics with less-than-sign i greater-than-sign. Close them with less-than-sign /i greater-than-sign.

Dazz said...

Listen, I'm a layman too, but I don't think this is rocket science...

1. The figure of 45 million differences is a haploid number. It seems to me that the number is 96 million differences. This is off by a factor of over 2X.

No, those 45M account for the entire diploid genome. Do you seriously believe scientists could make such a basic mistake and get away with it for decades? Are you really that kind of special stupid? Besides, you can run BLAST comparisons of genes and find that there's an almost constant 98.5% match between the equivalent human and chimp genes. Unsurprisingly, 45Mbp/3200Mbp = 98.5%. Magic?

2. The average generational time of ten years is <1/2 of chimp or human generations today. This is off by a factor of over 2X.

Human females are fertile at 12-13 years old. You may find this disturbing this but primitive humans probably didn't find it morally wrong to reproduce at the first opportunity. And of course, you chose to do your math based on the highest estimate for some reason, applied to all of the ancestors.

3. The mutation rate was given as 100 but since Roach could only confirm 28 and even if we double this as he doesn't indicate if he was testing haploid or diploid, this is off by a factor of a little less than 2X.

I'm in no position to question that, but I'm aware of certain molecular clock calculations that push the latest common ancestor of humans and chimps back to 12My. So what?

4. We are assuming that mutation rates are constant. I think there are reasons to say it is increasing and we need more long term data, but I am willing to use what we have and I don't want to argue over what we don't know.

What are your reasons to believe that mutation rate is increasing? And are you on board with the creationist trope that junk DNA is bunk? And also that mutations are always deleterious? If you put that together, and you calculations put our latest common ancestor 40MY back, how come we haven't gone extinct with an increased mutation rate if those are so harmful?

Unknown said...

John Harshman Friday, November 27, 2015 2:17:00 PM
Graham writes: The figure of 45 million differences is a haploid number. It seems to me that the number is 96 million differences. This is off by a factor of over 2X.
.
John writes:
Where do you get 96 million? And again you seem to be confusing differences with fixed mutations. Why count a single mutation twice just because it's present in two copies in a diploid? It's still the result of one event.
.
If someone states the difference is 1.5% and uses a diploid number to calculate the difference, I don't see this as any difference than Dr. Moran using "Shendure and Akay multiplied the mutation rate per base pair times the amount of DNA in a human cell. Since human cells are diploid, and the genome size is 3.2 × 10^9 this works out to 6.4 × 10^9 bp."
.
It seems to me that if we are going to make any calculations then the numbers should be consistent. If you are going to use haploid numbers then I will ask why would we double the number of mutations? Aren't those mutations in both copies?
.
John writes: Once again. 35 million SNPS in 10 million years (remembering that some have happened in the chimp lineage and some in the human) comes to 3.5 per year. If a generation is 20 years, that's 70 per generation. How is that a problem?
.
I have a problem starting with the 35 million. 1.5% of 3.2 billion is 48 million. But I agreed that if we can substantiate 96 mutations per generation we have the time needed to accumulate the SNPs.
.
The most important disagreement I have is that throughout this blog I have been accused of thinking (even though I don't) that SNPs can effect a change of taxa. Everyone is saying don't think that, so why are we using SNPs to determine a common ancestor?
.
John writes: One problem is that nobody is clear what you mean by "a change in taxa". Are you talking about speciation? Significant morphological change? Dogs giving birth to cats?
.
I can appreciate that. I have been using the term "change of taxa" but in reality speciation will not explain the differences in higher taxa. That is why I would like a family change but even a genus change would be a start.

Unknown said...

Thanks for the tip. I will use those when I quote others.

Unknown said...

Dazz Friday, November 27, 2015 2:35:00 PM
No, those 45M account for the entire diploid genome. Do you seriously believe scientists could make such a basic mistake and get away with it for decades? Are you really that kind of special stupid? Besides, you can run BLAST comparisons of genes and find that there's an almost constant 98.5% match between the equivalent human and chimp genes. Unsurprisingly, 45Mbp/3200Mbp = 98.5%. Magic?
.
I agree it isn't rocket science. If Dr. Moran stated that the 82 mutations were calculated on a diploid number of 6400Mb then you are using a haploid number if you are using 3200Mb.
.
Now I don't really care which number you use as long as you aren't mixing the types. So if you want to calculate the differences in a haploid then why use diploid to measure the speed of change. This doesn't seem right to me.

Unknown said...

Dazz Friday, November 27, 2015 2:35:00 PM
< i > Human females are fertile at 12-13 years old. You may find this disturbing this but primitive humans probably didn't find it morally wrong to reproduce at the first opportunity. And of course, you chose to do your math based on the highest estimate for some reason, applied to all of the ancestors.
.
Human females may have the ability to reproduce at 12-13 but historically they don't. Furthermore even though chimps have the ability to reproduce in their early teens the average is >24. So I am not using the highest possible but the < the average. It seems to me you are using the lowest possible and even at that you are off 30-40% of the claim. < i/ >
.

Dazz said...

You know full well the numbers work. It's 100 diploid in two lineages, so we're diverging at a 200 mutation rate. You have no excuse because you mentioned this overlook yourself. When you compare two genes, there's no haploid/diploid issue.

Human females may have the ability to reproduce at 12-13 but historically they don't. Furthermore even though chimps have the ability to reproduce in their early teens the average is >24. So I am not using the highest possible but the < the average. It seems to me you are using the lowest possible and even at that you are off 30-40% of the claim

Ridiculous. You're ignoring the fact that females can procreate every year after reaching their fertile age. A 40 year old woman with 4 kids gives an average of one generation per 10 years.

Unknown said...

DazzFriday, November 27, 2015 2:35:00 PM

What are your reasons to believe that mutation rate is increasing? And are you on board with the creationist trope that junk DNA is bunk? And also that mutations are always deleterious? If you put that together, and you calculations put our latest common ancestor 40MY back, how come we haven't gone extinct with an increased mutation rate if those are so harmful?

Pushing back the common ancestor does something else. It pushes back every evolution in the lineage. Sooner or later we are going to exceed the 3.5 billion year window to accomplish everything. If you allow us to go beyond that point, at what point does it become unreasonable? 16 billion years?
.
The reasons I can see for increased mutation are:
1. Increased cosmic radiation due to a reducing magnetic field.
2. Modifications to our diet due to processed foods and possibly genetic engineering and the use of pesticides, steroids and penicillin.
3. Additional stress of the modern culture.
4. Accumulating mutations, especially deleterious mutations exceeding beneficial ones by a huge margin.
.
This directly effects your last question. If mutations are increasing and the effects are generally bad, I would expect to see a decline in the fitness of a species but not necessarily an extinction. As long as the species can compete then it will not become extinct until the accumulation of mutations reaches a point where it is fatal and even then only a decline in population is ensured.

Dazz said...

Pushing back the common ancestor does something else. It pushes back every evolution in the lineage

So you're positing an ever degrading genotype? Why would the successors survive then if the ancestors were fittest?

The reasons I can see for increased mutation are:

1. Increased cosmic radiation due to a reducing magnetic field.
2. Modifications to our diet due to processed foods and possibly genetic engineering and the use of pesticides, steroids and penicillin.
3. Additional stress of the modern culture.
4. Accumulating mutations, especially deleterious mutations exceeding beneficial ones by a huge margin.


LOL, good to know it's not because of the original sin... pheew! I'm sure you have studies to confirm those expectations of yours, and a reasonable explanation of why other species (or even other humans) that don't eat fast food or have to cope with the same level of stress we do in the first world seem to have pretty much the same mutation rates.

If mutations are increasing and the effects are generally bad, I would expect to see a decline in the fitness of a species but not necessarily an extinction

What's your model then? The bacteria flagellum being the perfect, special creation and everything else spiraling down ever since? You seem to suggest that deleterious mutations are generally fixed somehow

Unknown said...

DazzFriday, November 27, 2015 4:02:00 PM
You know full well the numbers work. It's 100 diploid in two lineages, so we're diverging at a 200 mutation rate. You have no excuse because you mentioned this overlook yourself. When you compare two genes, there's no haploid/diploid issue.
.
Firstly I agreed that a 10 million year timeline was possible with 96 SNPs per generation assuming that we are using a diploid number.
.
Secondly I don't know that the number of 100 is correct. I asked why Roach admitted he could only verify 28.

Thirdly I accounted for both lineages in the 10 MY timeline assuming that we are using a diploid number.
.
Lastly no one here is saying that SNPs can cause a change in taxa so what does it matter? All we can say is that 5-10 million years is enough to accumulate 48M SNPs. This doesn't get us any closer to showing how a change of taxa can occur.

Unknown said...

Dazz Friday, November 27, 2015 4:02:00 PM

Ridiculous. You're ignoring the fact that females can procreate every year after reaching their fertile age. A 40 year old woman with 4 kids gives an average of one generation per 10 years.

.
I will listen to what others have to say about this explanation of a generation. If I see no rebuttals I will assume they agree. Then I will answer you.

Unknown said...

Dazz Friday, November 27, 2015 4:31:00 PM
Pushing back the common ancestor does something else. It pushes back every evolution in the lineage

So you're positing an ever degrading genotype? Why would the successors survive then if the ancestors were fittest?

.
I fail to see how these two statements connect?
.
But to answer you. It is not necessary for offspring to be as fit or more fit than their ancestors. All they have to do is survive to reproduce fertile offspring.

Unknown said...

Dazz Friday, November 27, 2015 4:31:00 PM
LOL, good to know it's not because of the original sin... pheew! I'm sure you have studies to confirm those expectations of yours, and a reasonable explanation of why other species (or even other humans) that don't eat fast food or have to cope with the same level of stress we do in the first world seem to have pretty much the same mutation rates.
.
We have been irradiating fruit flies for years producing all kinds of mutations which far exceed the norm. This leads me to the conclusion there is a link between radiation exposure and mutations. So I see this as a reasonable model for mutation accumulation.
.
I am not aware there are any studies linking diet to mutation. Maybe someone here knows of something. We do have studies linking diet to fitness ie cancer rates.
.
I am not aware of studies done of people taking any account for stress. So I don't think you have made a statement which you can confirm. Stress is related to fitness also, so I conclude that this may be a factor in mutation.
.
Lastly it is reasonable to conclude that if we tinker with a computer program a random change may not have any noticeable effect on the performance. As random changes accumulate the program will exhibit greater and greater problems and sooner or later the program will crash. I can not see how random changes will lead to a program with a different function.
.
For example, if we make random changes to a checker game program will it ever turn into a chess program no matter how many random changes occur?
.
So you asked me for reasons which I gave, and now you are asking for studies. If you had asked for studies in the first place I would have told you I only have reasons.

Faizal Ali said...

Jesus Christ on a stick. Graham Abbott doesn't even understand what "diploid" means, and yet presumes to lecture experts like John Harshman and Larry Moran on molecular evolution.

You couldn't make it up.

Faizal Ali said...

Lastly it is reasonable to conclude that if we tinker with a computer program a random change may not have any noticeable effect on the performance. As random changes accumulate the program will exhibit greater and greater problems and sooner or later the program will crash. I can not see how random changes will lead to a program with a different function.

You're wrong. (Surprise!) The best way to improve computer program would be to have millions of copies of the programs running and make random changes to them, trying out as many different combinations of changes as possible, keeping the best ones, and repeating the process on those over millions of years. You'd end up with programs that would perform tasks you would not even imagined was possible. It's just impractical to do this, so we have to use the more time efficient, but ultimately less effective, method that could be called "intelligent design."

ealloc said...

Is SNP actually a better term than SNS here? SNP = single nucleotide polymophism, which measures polymorphism within a population, rather than differences between two species, right? SNS for "single nucleotide substitution" isn't standard, but it does seem a little more accurate. "Single nucleotide difference" seems even better.

Unknown said...

Piotr Gąsiorowski Friday, November 27, 2015 12:05:00 PM
Since speciation does not happen overnight (or in a year) and is a process rather than a sudden event, there are quite certainly tens of thousands of ongoing speciations happening right now, some just starting, some advancing in their mid course, and some almost complete. Examples are easy to find.
Since there are millions of species, why would you say tens of thousands are in the process. Is there anything which will stop the process?
.
Since the claim is 6 changes of genera in 7 million years why don't we have many (if any) examples of this level change?

Piotr Gąsiorowski said...

All living species are evolving, but only some of them are undergoing population divergence which may result in speciation.

"Genera" and other taxonomic grades above the level of species are not natural units. It's just a convention to call a group of related species a "genus". Linnaeus originally placed chimps and orangutans (he didn't know gorillas) in the genus Homo.

Do you realise that if you have a family tree, speciations (representing as branchings in the tree) are not linearly ordered but may happen in parallel? If you have a common ancestor giving rise to two descendants, and each of them splits into two descendants again after two million years, and the same thing happens again in another two million years, and again in another two, 7 million years later you have eight new taxa, but the average time from speciation to speciation is two million years.

Faizal Ali said...

Since the claim is 6 changes of genera in 7 million years why don't we have many (if any) examples of this level change?

For at least the 3rd time: What do you expect to see if this happens?

And do you really expect to witness something that happens less then once in a million years?

John Harshman said...

Ridiculous. You're ignoring the fact that females can procreate every year after reaching their fertile age. A 40 year old woman with 4 kids gives an average of one generation per 10 years.

No, that isn't what a generation is. You don't find the generation length by dividing the parent's age by the number of kids. Generation time is the age of a parent at the birth of a child. If your 40-year-old mother had one child at 20, one at 25, one at 30, and one at 35, the generation time calculated from that would be 27.5 years.

John Harshman said...

the number of mutations is reasonably measured using a diploid genome, as that's where mutations happen. But the number of fixations is reasonably measured using a haploid genome, as in order to become fixed the same allele must be present in both chromosomes, and counting it twice is silly. It's 35 million point mutations fixed in the human and chimp lineages combined. Use the right numbers, at least.

Still don't know what you mean by a "change in taxa", and I don't think you do either.

Dazz said...

Thanks for the correction John

Faizal Ali said...

Still don't know what you mean by a "change in taxa", and I don't think you do either.

That would explain why he refuses to answer my question re: what a "change of taxon" would look like when we observed it.

Piotr Gąsiorowski said...

I asked it too and got no straightforward answer either. I suspect Graham more or less identifies genera with "created kinds" and thinks speciation is a microevolutionary process which can give rise to a new species but not a new genus, family, order or phylum. It's always a problem for people with an essentialist frame of mind.

Unknown said...

@John:
The number of mutations can be reasonably given for either haploid and diploid genomes. It's worth noting that the rate for either is the same. But it is worth noting that since the rate of fixations is roughly the rate of mutations (allowing for slight divergences from strict neutrality) the number of fixations per generation corresponds to the number of mutations per haploid genome.

My guess at what Graham means by "change of taxa" is. En example where post-speciation one of the daughter species does not belong to a clade the ancestral species belonged to. It's obviously a non-starter when you understand phylogenetics, but it's not as if that was something laypeople usually had a great understanding of (nor do all professional biologists - I just got to read a paper in which for a species tree ((A,B),C) gene trees with ((A,C),B) and ((B,C),A) were discussed and the only possible explanations given were convergence and HGT. Retention of the plesiomorphic state wasn't even in contention!).

Unknown said...

DazzThursday, November 26, 2015 3:24:00 PM
What he's saying is that the change is gradual and wouldn't be obvious from one generation to the next... obviously. It's you who's either misunderstanding or misrepresenting others. There's plenty evidence for speciation
.
But there is no evidence that a change at the family level can occur. So even though I would expect slow gradual changes sooner or later a unique feature must occur to explain why we have unique features which classify animals into higher taxa.

Faizal Ali said...

So give us an example, Graham? What would you expect to see if a new family arose in the next year?

Faizal Ali said...

Let's get specific, Graham. As an example, the order Carnivora breaks down into families that include Canidae, Ursidae, Mustelidae, Mephitidae, and Pinnipedia (all within the suborder Caniformia).

So are you telling us that, eight times a year, you expect to see a dog give birth to a bear or a weasel or a walrus, or something like that? Are you really that stupid, or are you just trying to be funny?

Piotr Gąsiorowski said...

Graham,

We all (including you, I'm sure) agree that dogs (Canis familiaris) and cats (Felis silvestris catus) belong to different families (called, respectively, Canidae and Felidae). The last common ancestor of those two families is at the same time the last common ancestor of all modern Carnivora (including bears, seals and walruses, raccoons, hyaenas, civets, etc.) -- about two dozen families, extant or extinct. Some carnivorans (e.g. bears, seals, raccoons and weasels) are more closely related to dogs than to cats, whereas others (e.g. civets, fossas, mongooses and hyaenas) are more closely related to cats than to dogs. So Carnivora can be divided into two sister groups (you can assign them the rank of suborders if you think ranks are important), Caniformia (doglike) and Feliformia (catlike).

This proto-carnivoran lived an estimated 42 million years ago, in the middle Eocene. By definition, it underwent speciation into two descendants, one of them ancestral to dogs (as well as bears, etc.), and the other to cats (as well as civets, etc.). That doesn't mean that the former already looked like a dog and the latter like a cat. As a matter of fact, they were closely related and closely similar. A hypothetical biologist living at the time would have placed them in the same genus. But once they parted ways, there was no coming back. Differences between them gradually accumulated.

For example, vaery early in the "catlike" line the auditory bulla enclosing the middle and inner ear developed into a double-chambered structure composed of two bones, while "doglike" carnivorans have a single chamber. This is not even something you can see looking at a live animal -- it's a hidden part of their anatomy. But (together with a few other diagnostic features) it was inherited by the descendant of the "catlike" ancestor and it can be used by palaeontologists to tell apart members of the two sister groups, no matter how similar they look to each other, or how different they are from modern cats and dogs.

The most recent common ancestor of the modern canids lived close to the Eocene/Oligocene boundaryu (some 34 million years ago). The most recent ancestor of modern felids lived much later, about 11 million years ago. It means that inherited features shared by canids (dogs, foxes, raccoon dogs, etc.) but not shared with felids (cats, lions, cheetahs, etc.) had some 9 million years to develop, while typical felid features (not shared with canids) accumulated over 31 million years or so. It's plenty of time to grow different.

Hope it helps.

Unknown said...

Piotr GąsiorowskiFriday, November 27, 2015 6:32:00 PM
Do you realise that if you have a family tree, speciations (representing as branchings in the tree) are not linearly ordered but may happen in parallel? If you have a common ancestor giving rise to two descendants, and each of them splits into two descendants again after two million years, and the same thing happens again in another two million years, and again in another two, 7 million years later you have eight new taxa, but the average time from speciation to speciation is two million years. <./>
.
This is a good point Piotr which I hadn't considered. So lets say you are right and it takes 2 million years for a change at the genus level (even though most animals reproduce far faster than mankind). We have no evidence this has happened in any of the 8 million species on the planet in the last 150 years. Doesn't this seem out of the bounds of statistical probability?

Faizal Ali said...

Graham, you still seem to find it surprising that no dog has given birth to a bear in the last 150 years. Why is this? Do you really think that should have happened?

Faizal Ali said...

If that's too difficult a question, Graham, try these: What are the odds that 2 million years could elapse over the course of 150 years? Are those odds increased if we have 8 million periods of 150 years running concurrently?

Unknown said...

Simon are you saying that Shendure and Akay shouldn't have doubled the number of haploid mutations? Or are you agreeing with me that if we are going to measure the mutations number in a diploid context that we have to measure the differences in a diploid context?
.
As to the change of taxa, yes you stated what I think is reasonable to expect. If we only end up with new species then we won't have new families. Maybe you could expound on that to demonstrate how it could happen? So for instance Piotr was explaining about the differences between dog and cat families. This is more than pupil changes and a retractable claw. It involves skeletal and inner ear changes and probably a lot more. What kind of mutation is required to make such changes?

Faizal Ali said...

As to the change of taxa, yes you stated what I think is reasonable to expect.

Wait, are you referring to this:

My guess at what Graham means by "change of taxa" is. En example where post-speciation one of the daughter species does not belong to a clade the ancestral species belonged to.

Seriously? And:

If we only end up with new species then we won't have new families.

Are you expecting that "new families" can possibly arise? Again: Seriously?



Piotr Gąsiorowski said...

This is a good point Piotr which I hadn't considered. So lets say you are right and it takes 2 million years for a change at the genus level (even though most animals reproduce far faster than mankind). We have no evidence this has happened in any of the 8 million species on the planet in the last 150 years. Doesn't this seem out of the bounds of statistical probability?

Once again, Graham. There is no "change at the genus level", "at the family level", or "at the order level". There are only speciations: one ancestral species splits into two, still closely related but separated by a reproductive barrier. Over millions of years, speciation happens again and again, producing more descendant species. Some of them are more closely related to one another (e.g. lions and tigers) because they share a relatively recent common ancestor. Others are more distantly related, i.e. derived from a much older common ancestor (e.g. lions and bobcats). We group them into genera depending on how different they look today, so both lions and tigers are placed in the genus Panthera but bobcats in a fifferent genus, Lynx, and even in a different subfamily. This does not mean that the separation between lions and tigers was "at the species level" but the separation between lions and bobcats was "at the genus/subfamily level". Genera, families, orders, classes, etc. are defined from the modern point of view, with hindsight. Every new lineage starts off as a new species, and given the limited time we spend on earth, you can only hope to observe speciation in progress directly (complete speciation normally takes much, much longer than 150 years). Do you want to see a new genus? Come back in a few million years.

Anonymous said...

Graham Abbott wrote, "We have no evidence this [change at the genus level] has happened in any of the 8 million species on the planet in the last 150 years." This statement fails to make sense at so many levels that it's hard to respond to. I see basic problems with understanding both evolution and taxonomy, as well as a relatively trivial error of fact.

First, the trivial. Humans have produced a new genus at least once, by applying colchicine to hybrids of Brassica and Rapa (different genera of mustards). Importance: proof of concept, that many of the species that seem to have originated as intergeneric, allopolyploid hybrids (e.g. Bluebunch Wheatgrass) really could have done so.

Note: Speciation within a genus by hybridization and polyploidy has been observed in the wild in Tragopogon and several other genera of plants.

More basic. Speciation is usually a slow process. Species A becomes divided into two populations, which gradually become different. We can see that this happened after the fact, when we find that species A1 and A2 are so different that they can't interbreed. The accumulation of changes that we would recognize at the genus level is even slower and we can only see that it happened long after the first splits in the lineage happened. Therefore, we would not expect to be able to see this amount of change in any period of a few centuries.

Here's an even more basic point, which I wouldn't mention to most people arguing about kind of thing, but you seem smart enough to hold two ideas in your head at the same time.

1. The pattern of variation that we name as species, genera, families, etc. is real.

2. The choice to group species into genera or genera into families, etc., is arbitrary. It is done by humans to allow us to talk about the variation (the species) in larger named units.

For example, the stonecrops (certain succulent plants) can be treated as one genus, Sedum, or divided up into different genera, Sedum, Gormania, Amerosedum, etc. The patterns of variation and the relationships are real and the same no matter what we call them. You aren't going to see some beginning point labeled "genus level change" out there in the world.

Minor side note: The ranks of taxonomists are thinning as older ones die and new ones aren't hired. Therefore, we have no real idea of what's going on with most of the far more than 8 million species are out there, especially the smaller organisms that reproduce fastest.

Anonymous said...

Graham Abbot writes, with regards to the difference between cat and dog lineages, "This is more than pupil changes and a retractable claw. It involves skeletal and inner ear changes and probably a lot more. What kind of mutation is required to make such changes?"

Mutations can produce an enormous diversity of changes in bones, ears, claws, and all the other traits that organisms have. The same kind of mutations we always see: SNP's, indels, etc. The thing is, in the long time the cat and dog lineages have been separate, several different mutations have become fixed, causing multiple differences between these groups.

Unknown said...

First, the trivial. Humans have produced a new genus at least once, by applying colchicine to hybrids of Brassica and Rapa (different genera of mustards). Importance: proof of concept, that many of the species that seem to have originated as intergeneric, allopolyploid hybrids (e.g. Bluebunch Wheatgrass) really could have done so.
.
This is not a natural event. Someone might think to claim this is intelligent design. Secondly I found no articles claiming a new genera was created or named, only a new species. Please provide a link so I can see what you do.
.
That speciation occurs is not in dispute. What I would like to see is evidence new information can be introduced into animals naturally. Now Pietr made mention of a different inner ear in the lineage of cats and dogs. Now this is what I am talking about. A structural change occurred. So when cats developed a different iris or a retractable claw, this is a significant change which is more than a speciation where no one can tell that something has happened. I would imagine that a significant difference in the DNA would accompany the change. If this could be shown to occur naturally then I would consider this an observation.
.
Even though the division of genera is arbitrary sooner or later a change has to happen which defines the organism. So for instance the retractable claw. I don't see any evidence of this happening gradually and there are multiple structures involved making the chances of it evolving concurrently low.
.
So when millions of organisms are evolving albeit at different rates and they are at different stages, it would be expected that a few of them are at the point of a change each year. If only speciation occurs then how did plants or animals develop? At some point an organism had to develop the means to manufacture chlorophyll and utilize it to process solar energy.
.
When we compare this to what we observe in bacteria (because they multiply faster increasing the chances of a change in a short time) we are not seeing a change which involves any structure. This is a way for us to observe millions of years of evolution because in human terms this is 60 million years worth of reproduction. If 3 million generations have passed when bacteria have been cultured in labs and no significant change has occurred, then why would we expect changes of far greater significance occurred in hominids in a far less amount of reproductions?

Unknown said...

Mutations can produce an enormous diversity of changes in bones, ears, claws, and all the other traits that organisms have.
.
Can you give an example we have observed? I am not talking about wings duplicating or taking the place of antenes, I am talking about something new like a foot replacing a hand.

Pete Dunkelberg said...

Hi Larry, how are you? I've been out of contact with evo-debates for some years I'm afraid. I would like to know how many mutations other than SNP's I am likely to have, counting an addition or loss of a block of N base pairs as just one mutation. And can you also estimate the number of gene duplications? What other kinds of mutations are there?

Thanks,
Dunk