Mutations Confirm Common Descent

It isn't just the pattern of similarities that evidence common descent and evolution. The differences also provide us with big pieces of evidence.The inspiration for this post comes from EvoGrad and Stephen Schaffner, from whom I will be borrowing various figures. Ebersberger et al. (2002) published on this subject, and could have served as direct or indirect inspiration for EvoGrad and Schaffner.With credit given, let's move on to the evidence.Mutations are heritable changes, specifically heritable changes in the DNA sequence of a genome. The type of mutation I will be discussing is a substitution mutation where one base is swapped out for a different base. For example:AGGCTAATCG --original
AGGGTAATCG --mutatedThere are two main types of substitutions: transitions and transversions. They are called this because if the mutation is between two similar bases it is a transition, and a transversion if it is between two dissimilar bases. The two classes of bases are purines and pyrimidines. For clarity's sake, I like to refer to them as one ring and two ring bases, as shown in the picture below:
credit: EvoGradDue to the biochemistry of genetics, transitions tend to happen more often than transversions. That is, substitutions occur more often between bases that have the same number of rings. Even though there are two possible transversion mutations per base compared to just one possible transition, we still see more transitions than transversions.Evograd compiled a total of 220,000 de novo (i.e. new mutations detected in experiments in green) human mutations from various papers and compared them to 78.6 million substitutions found in the existing human population (i.e. the standing variation in the human population in blue) from public databases. This is what that comparison looks like:
credit: EvoGradLike I stated earlier, transitions outnumber transversions in this figure. The first set of bars are the transitions, and the other three sets of bars are the transversions. Also, the rate at which these mutations occur in real time matches the standing variation in the human population. In other words, this is smoking gun evidence that the process we observe creating mutations in real time is responsible for the variation we see in the human population. The fingerprint produced by the natural process of mutation is measurable and present in the human population.But what if we do the same thing for a comparison of the human and chimp genome? The model for common descent and evolution states that humans and chimps share a common ancestor. Therefore, this model predicts that our lineages started from the same ancestral genome and population. As our lineages diverged, the same process of mutation should have created differences between those lineages. Therefore, if this model is correct then we should see the same fingerprint when we compare the human and chimp genomes.
credit: EvoGradWouldn't you know it, there's that fingerprint. In fact, let's extend it out to other primates:
credit: SchaffnerThere's that same fingerprint, just as we would expect from common ancestry and evolutionary mechanisms.This is smoking gun evidence for common ancestry. This evidence is exactly what we would expect to see if our models are true.

Bump for administrative review.If there are changes needed for approval please let me know. Thanks.

In science we like to use statistics to measure the fit between data and model, so we should do the same for spectrum of mutations. Luckily, Francioli et al. (2015) have already done this for us. They compared mutations in the context of three base pair motifs, as well as in the context of CpG and non-CpG mutations (a subject I will probably touch on later).
Figure 6 | Correlation between observed de novo mutation rates and human/chimp substitution rates for mutation types in different trinucleotide contexts. De novo mutation rate spectrum (Y-axis) is plotted against substitution rate spectrum inferred from human vs chimp comparison (X-axis). Each dot represents a type of mutation in a specific trinucleotide context. The Pearson’s correlation coefficient r2 = 0.993. Figure from Francioli et al. (2015) (Supplemental Figure 6).An r-squared of 0.993 means that the processes we observe producing mutations in genomes explains 99.3% of the differences observed between the human and chimp genomes. You don't often see this tight of a regression in biology or genetics.

Transitions and transversions are the two main categories of substitution mutations, but there is yet another subset below that. These are CpG mutations. CpG is short for Cytosine-phosphate-Guanine. In other words, CpG is a CG in basic DNA sequence notation. For example, this randomly generated DNA sequence:AAACTGCTATTCGTCGCCGATGGTGGTAACAre there any CpG's? Yes, there are 3 CpG's.AAACTGCTATTCGTCGCCGATGGTGGTAACSo what is so different about CpG's? In eukaryotes (like humans and other primates) the C in a CpG can be methylated. That is, a methyl group is covalently attached to the base. This can lead to a process called deamination which produces a T.So this is a case of a mutation occurring in place instead of a base mismatch occurring during replication. CpG mutations occur at a much higher rate on a per base basis than do other types of mutations. The important thing to note is the use of rates in this example. There are far fewer CG in a sequence than there are single A, T, C, and G. If we look at the rate we are asking how often these types of mutations occur compared to where they could occur. So we aren't looking at the simple sum of all types of mutations, but their rates. In this context CpG transition mutations occur at a much, much higher rate than do other types mutations, and we see that in the multiple papers that directly measured mutation rates in human parent-offspring trios.So let's go back to the Francioli paper and figure in message 4. They looked at the rate of mutations in all possible three base sequence. I created a list of all three base combinations in python and then highlighted all of the CpG's.['ATC', 'ATG', 'ACT', 'ACG', 'AGT', 'AGC', 'TAC', 'TAG', 'TCA', 'TCG', 'TGA', 'TGC', 'CAT', 'CAG', 'CTA', 'CTG', 'CGA', 'CGT', 'GAT', 'GAC', 'GTA', 'GTC', 'GCA', 'GCT']As you can see, there are 4 triplets that have CpG's. Therefore, there are four triplets that should see a much higher rate of C to T transitions than any other three base combination. So what do we see in the Francioli paper? Take a look at the figure in message 4. There are a group of 4 triangles all by themselves towards the upper end of the regression line. They are transitions given the color and the key in the figure (C:G>T:A) and they are CpG mutations indicated by the triangle shape of the data points. Those 4 data points off by themselves are the CpG transitions, and the rate of CpG mutations observed in humans is the same as the rate of CpG transitions seen between the comparison of the human and chimp genomes.It can be seen even more clearly in a bar plot created by Stephen Schaffner:Those sets of bar graphs did not include the de novo rates which is why I discussed the figure from the Francioli paper.In combination, we can see that CpG are observed to occur at the highest rates in humans, CpG mutations occur at the highest rate when looking at standing human genetic variation, and (most importantly) CpG mutations occur at the highest rate when comparing the human and chimp genomes.This is yet another massive and obvious fingerprint of mutation and common descent in the genetic data demonstrating that humans and chimps do share a common ancestor and that the differences between our genomes was produced by the very same processes we see operating in nature today.

There is scientific evidence that all primates share a common ancestor, and that evidence was presented in this thread. Given your inability to actually address that evidence I can only assume it is really, really good evidence.

A few off the top of my head: somatic mutations, DNA methylation, T and B cell recombinants from V(D)J recombination, and telomere deletions in somatic cells. It's the DNA sequence that mutates which produces new alleles. The frequency of an allele is really just the distribution of mutations within a population in a specific gene.

You complete inability to address the evidence is noted.

Why don't you address the evidence?

You still can't address the evidence in the opening posts?

I think the average is somewhere around 99.5%. Depends on the primate. We share around 96.5% of our DNA with chimps if you factor in indels, and they are our closest relatives. Out of all primate species chimps share the most DNA with us. Conversely, chimps share more DNA with humans than they do any other primate species. Are you going to address the evidence in the beginning of the thread, or is this just a gotcha post?

Can you address the evidence presented at the beginning of this thread?

With as many living humans as there are right now (~7 billion) it is almost guaranteed that every non-lethal mutation exists somewhere in the human population, even if it is just in a few individuals. Therefore, nearly all DNA is not shared amongst all humans. Only a tiny fraction of all bases in the human genome are going to be shared by all humans. I don't think it is that easy because you would need human populations that are not interbreeding with one another. However, the math does exist:

quote:

---

Each of the 2N genes in the original population has an equal chance of being the lucky one

​

Any one gene in the population, therefore, has a 1/( 2N ) chance of eventual fixation by random drift (and a ( 2N -1)/( 2N ) chance of being lost by it). Because the same argument applies to any gene in the population, it also applies to a new, unique, neutral mutation.

​

When the mutation arises, it will be one gene in a population of 2N genes at its locus (i.e., its frequency will be 1/(2/N )), and it has the same 1/(2N ) chance of eventual fixation as does every other gene in the population. It is therefore most likely that the mutation will be lost but it does have a small chance of success. The probability that a neutral mutation will eventually be fixed is 1/(2N ).

​

The rate of evolution equals the probability that a mutation is fixed, multiplied by the rate at which mutations appear. We define the rate at which neutral mutations arise as µ per gene per generation. (µ is the rate at which new selectively neutral mutations arise, not the total mutation rate. We are here considering only those mutations that are neutral.) At each locus, there are 2N genes in the population: the total number of neutral mutations arising in the population will be 2N µ per generation.

​

The rate of neutral evolution is 1/(2N ) x 2Nµ = µ. It is equal to the neutral mutation rate.

​

Evolution - Random events in population genetics

---

One of the difficulties could be the mutation rate which won't necessarily stay the same over a given period of time. Some papers suggest that the mutation rate in chimps is 50% higher than that in humans, as one example:

quote:

---

Mutations generate genetic variation and are a major driving force of evolution. Therefore, examining mutation rates and modes are essential for understanding the genetic basis of the physiology and evolution of organisms. Here, we aim to identify germline de novo mutations through the whole-genome surveyance of Mendelian inheritance error sites (MIEs), those not inherited through the Mendelian inheritance manner from either of the parents, using ultra-deep whole genome sequences (>150-fold) from a chimpanzee parent-offspring trio. We identified such 889 MIEs and classified them into four categories based on the pattern of inheritance and the sequence read depth: de novo single nucleotide variants (SNVs), [ii] copy number neutral inherited variants, [iii] hemizygous deletion inherited variants, and [iv] de novo copy number variants (CNVs). From de novo SNV candidates, we estimated a germline de novo SNV mutation rate as 1.48 × 10−8 per site per generation or 0.62 × 10−9 per site per year. In summary, this study demonstrates the significance of ultra-deep whole genome sequencing not only for the direct estimation of mutation rates but also for discerning various mutation modes including de novo allelic conversion and de novo CNVs by identifying MIEs through the transmission of genomes from parents to offspring.
Direct estimation of de novo mutation rates in a chimpanzee parent-offspring trio by ultra-deep whole genome sequencing | Scientific Reports

---

I was going from memory, so if there are published papers with 99.9% then go with that. Yes. If you pick a specific base from a specific person, there will be at least a handful of people with a different base at that position as long as it isn't a lethal mutation. This is because 7 billion births is enough to create every possible mutation in the human genome several times over. For the idealized neutral mutation rate the size of the population doesn't matter. All that matters is the mutation rate. If you look at the equation the population size cancels out in the equation.

It is an an average between any two human beings. Some will share more (such as close relatives) and some will share less. This article is probably right up your alley:What Genetics Says About Adam and Eve - Article - BioLogosIt discusses the distribution of rare and common mutations and how they play out in a population over time. That is a simplification that many use because it is difficult to calculate a meaningful indel or recombination rate. Substitutions are much more common and occur at a rate that is more clock like.

In the equation for neutral fixation the probability of a mutation reaching fixation is just the number of genomes with the mutation compared to the total population of genomes. Since humans are diploid, you would take the population size and multiple by 2 to cover 2 genomes per person. When a mutation first occurs in a population with 1 million individuals the probability of that mutation reaching fixation is 1 in 2 million (1 million people = 2 million genomes).