Testing YEC Hyper-Evolution from Common Ancestors: Comparisons of mtDNA Genome Diversity in Mammals

Ken Ham’s Ark Encounter will vividly display what many young earth creationists believe the common ancestors of today’s species of animals may have looked like as they rode out a global flood 4350 years ago.   After disembarking from the ark, young earth creationists have proposed that these common ancestors experienced a burst of adaptive radiation into new  environments leading to the creation of thousands of new species.   If this prediction is accurate then I would expect there to be a very close genetic relationship between those organisms that have shared a common ancestor less than 5000 years ago.  On the other hand while Ken Ham and others believe in common descent for some species they deny common descent between somewhat ill-defined “kinds.”  Kinds, therefore do not share a common ancestor and therefore should not show the same sort of genetic relationships as do those that do share common ancestors.   I have argued that creationists are ill-advised to believe that the genetic data support their hyper-evolution/sparate kinds hypothesis.

To test my observations about YE creationist’ genetics, I made some predictions about the genetic similarities of some species of mammals (See: How Similar is Similar: Barmins, species and common ancestors).  I didn’t make my predictions in a vacuum but rather personal experience with comparing DNA sequences of numerous individual genes in many of these species.  However, I had never compared complete mitochondrial genomes before undertaking this exercise. Below I present some background about what I did and then I show you actual data that I collected to test the predictions that I have made.   I make no claim that the data I present are convincing evidence of common ancestry or a lack of common ancestry. My point is not to prove or disprove common ancestry but to provide some provocative data that all who wish to understand the relationships of organisms and their origins must account for.

What is the Mitochondrial Genome and Why Compare these Sequences?

The nuclear genome is very complex, very large, and not all species have been completely sequenced and so careful apples-to-apples comparisons would be difficult.  I have chosen a simpler system to compare genetic differences among species: The mitochondrial genome.

The human mitochondria (power plant of the cell) has its own genome of 16,569 base pairs.  Each one of us has multiple copies of this genome in every cell and you inherited this small genome from your mother.   What makes this genome very useful for comparing individuals and species is that it is passed down generation to generation and when there is a mutation that mutation will be passed to the next generation.  There is virtually no opportunity for lineages of mitochondrial genomes to get mixed or have genetic exchange.  Hence you can trace your mtDNA lineage back down your material lineage theoretically to the first human female.  This is were the idea of “mitochondrial eve” comes from.

Eve would have had one particular version of a mtDNA genome and then passed it to all her offspring.   If any of her daughters experienced a mutation in the mtDNA genome of their eggs, they would pass those variants to their children and so on. Thus if you and I differ at 100 locations among our 16,569 base pair then I would know that at least 100 mutations had occurred in mine and your lineages combined since we shared a common mother  – our common ancestor.  In the present day human populations one recent study (http://genome.cshlp.org/content/early/2016/02/16/gr.198754.115.full.pdf) reported that from comparisons of just 320 complete human mtDNA genomes there were 1156 locations in the mtDNA genome which vary in the population.  This  tells us that there are a minimum of 1156 unique mutations in the human mtDNA genome since mitochondrial Eve. Another analysis has identified more than 10,000 separate mutational events in the mtDNA genome in human history (see reference below for an explanation).

So these properties of the mtDNA genome of animals are useful for comparing time of divergence for species from an ancestor because if you have one species that splits into two species the differences in their mitochondrial genomes will generally represent the number of changes since there was an ancestral population (see footnote for caveats to this).  For the young earth creaitonists we could call this the mitochondrial eve for each “kind.”  For example, the ancestor of wolves (and domestic dogs) and coyotes had a mitochondrial genome sequence or some set of variant genomes in the population but when the population diverged and then became coyotes and wolves as separate species the time that they have been breeding separately will result in an accumulation of unique changes to each of their mitochondrial genomes.

How does all this relate to the question of genetic similarity between species that I have talked about previous (see: How Similar is Similar)?   The 70% genetic similarity of the human and chimpanzee genome according to Tomkins (update: he now has recalculated and say is more like 88%) is a difficult number to understand because it has no context.  I asked, what if we were to compare other genomes, how much difference would we see using the same criteria for estimating similarity?  Because Tomkins examined many types of changes to the nuclear genome it is hard to assess his values. I have chosen to look at much simpler system. The mtDNA genome is nearly the same size in all land animals. It has the same complement of genes and those genes are always in the same order.  This means that anyone can take the DNA strand from one species and just line it up with another species and ask – how many differences are their in their DNA code – The As,Ts,C,s and Gs?  For example we can line up a human mtDNA and chimpanzee code and see that each of us has on average has about 1450 differences between our genomes of nearly identical size.

With this simple system we can then ask, how different are the mtDNA genomes of some representative species of various mammals.  Or, put another way, how much variation is actually observed between species that young earth creationists believe are in the same “kind” (ie. have the same common ancestor).

I’ve made some comparisons of those species and here is how I did it.   It is very simple and you can do it yourself.  Go to this site: http://blast.ncbi.nlm.nih.gov/Blast.cgi?PROGRAM=blastn&BLAST_PROGRAMS=megaBlast&PAGE_TYPE=BlastSearch&SHOW_DEFAULTS=on&LINK_LOC=blasthome  and type in this code EF667005.1 for the Black bear mtDNA genome in the Query search window and then go to the bottom of the page and click on “BLAST”.  In a few seconds to minutes you will have comparisons and alignments of that genome with other bears and organisms that have similar sequences.   Scrolling down through the results you will be able to see the total number of differences between the “query” sequence and other species or individuals of the same species.   What I am not reporting in the table below are the number of small insertions or gaps that in the genome. They are not a significant number in most cases.

Finally – The Results!

Ok, lets get to the results.  Below is an image of the table of my results. Click for a larger and more legible version.  You can also download a PDF version by clicking here: mtDNA-genome-mammals-comparison-naturalis-historia.

Let’s make some observations: 

While mtDNA differences are not always going to reflect a 1-to-1 relationship with the differences we would see in the nuclear genome or the morphology (ie. what it looks like) of an organism what we see here is a good proxy for the relative differences we would see.   If two species have more mtDNA differences they probably also will have more total genetic differences than two species that have fewer mitochondrial differences.*  What can we infer from the data then?

1)  All members of the human race are very similar with respect to their mtDNA genomes.  There is remarkably little variation among humans (probably not much more than 100 differences among any two individuals).  Compare this to populations of chimpanzees which show over 300 variants among populations.   Chimpanzee individuals are therefore more variable than humans (BTW, sequences of nuclear DNA genome also confirm that chimpanzees have more total genetic variation than humans).  This might not seem intuitive since most people think that chimpanzees all look the same and humans seem to have many differences in appearance but appearances can be deceiving when assessing overall genetic diversity.   We will see this with dog breeds as well which are more similar to each other and to wolves than humans are to each other.

2) The Neanderthal and Denisovan sequences are more different from any living human than any living human is from one another.  However, even this amount of difference is less than is found in other closely related species or even among chimpanzees of the same species.  This is why Neanderthals are considered to be a separate branch and yet still very closely related to modern man (sometimes considered a subspecies or variety).  See the graphic below to get some sense of the differences of these hominid and modern human sequences.  This shows that Neanderthals and any person alive today will have about 200 differences in their mtDNA genomes whereas any two humans alive today only differ by a maximum of 100 differences.

3) The average human has 1462 differences with that of a typical chimpanzee.  That sounds like a lot of differences. This is 8.8%.  Notice that for Tomkins  (see previous article) he claims only 70% similarity  between humans and chimps but this was only for the nuclear genomes. Had he applied his algorithm to the mtDNA genome I am quite confident he would come up with nearly the same 8.8% difference or 91.2% similarity because this genome is simple and it is just a matter of looking at simple sequence differences. Everything else about the chimp and human mtDNA genome is the same.   But what does this 8.8% mean. We see above that human are less than 1% different so let me be clear that there is a very real difference between these genomes.   But lets look at some other genomes and see if this 8.8% is as much as it sounds.

4) I didn’t show the numbers for chimps compared to great apes and orangutans but the differences (<2000 differences) are much greater than the differences between humans and chimps. However, creationists who study “kinds” have placed all the great apes in a single “holobaramin” implying that all apes came from a common ancestor just 4320 years ago.   By putting apes, orangutans and chimps together they are saying that they believe that the great genetic distances between these groups could have formed by evolutionary processes since God created the original “ape” ancestor.  At the same time they are claiming that humans could not have evolved from apes because they so genetically distinct!?  Do you see why I have been saying that creationists should be careful about proclaiming genetic differences as proof of no common ancestor.  There is a severe double standard in interpretation going on here.

5) Let look at the canine sequences next.  Domesticated dogs and wolves are more similar than humans are to each other. Despite the obvious morphological differences in dog breeds they are genetically almost identical both in the mtDNA and nuclear genomes and they are barely different from wolves.   Look at the coyote and dog comparison.  There is an average of 700 differences.  Coyotes and dogs can interbreed and so this amount of difference is no big deal.    But look at the red fox and dog/wolf comparison. Here we see a whopping 2269 differences.  This is close to double the differences that we find between chimps and human mtDNA genomes.   Remember that many creationists wish to claim that foxes and wolves are part of the same holobaramin which means they all share a common ancestor.  They posit this likely ancestor to be a pair of canines that were on Noah’s ark.  If this is the case that pair would have contained  single (in the female remember) mtDNA genome which then would have had to have been copied into all of the descendant species of dogs including coyotes, wolves and the foxes.  Just between these two we would have to hypothesize that there have been 2269 mutations between these lineages.  Does this reflect how different the nuclear genomes are?  I would bet that it does.  I would be quite shocked if the red fox and wolf genomes were not more different from each other than a chimpanzee and humans  genomes are using Tomkins methods because there are differences in chromosome number and genome size in addition to high rates of sequence divergence and probably even novel gene differences between the two.

6) Now look at the “cat” family members.  The lion and tiger, which are poster species for creationists who say that evolution has happened within kinds, have 1373 differences.  This is almost the same as the number of differences between humans and chimpanzees and I think reflect some very real differences between lions and tigers. Once again, if their entire nuclear genomes were compared I am sure we would find very similar differences (ie. 70% overall similarity by Tomkins calculation method) between these two “closely” related big cats. But what of the other cats species.  Look at the domesticated cat and cheetah – 1765 differences.  Can creationists really claim that all these cats are actually related to a single common ancestor and probably had a single mtDNA sequence just 4000 years ago but then claim that the genetic difference between chimps and humans is so vast that it is ridiculous to think they had a common ancestor?

7) The rest of the chart starts to be a bit redundant if not even more extreme.  Look at the bears.  Brown bears and polar bears are clearly quite similar and have a common ancestor by everyone’s (creationists or not) estimation.  They differ by only 295 base pairs which is similar to the difference among chimpanzees or difference between humans and Neanderthals.  But look at the other bear species, especially, the Panda comparisons.  Some are far more different from one another than the chimp and human sequences.    Lastly, I threw in some ungulate samples here including sheep and goats.  Sheep and goats are almost universally proclaimed by creationists to represent the same “kind” and yet again we see again that these two taxa have as many differences as humans and chimps.  Again, I would say that I am fairly sure that when the entire genomes of these organisms are finally sequenced we will find that they will have as many significant differences as humans and chimps.

Summary

I expect that these data will be very surprising to some people. However, there really isn’t anything new here. These data have been around for quite a while even while hundreds of new genomes are becoming available every month.  I was not at all surprised by these numbers which is why I’ve been telling my creationists friends that they shouldn’t get excited about these percentage similarity numbers that they hear.   Of course there are many aspects of genomes that are much harder to compare.  Simple sequence similarity doesn’t tell the whole story (see footnote below for more details).  I do think that chimps are significantly different from humans but neither do I believe they are any more different, with respect to overall genetic similarity, than foxes and wolves or even sheep and goats.  It isn’t our code that makes us special!  If am am right it makes no sense for creationists to strenuously argue that humans and chimps are not related by common descent but then promote a view of rapid radical evolution of kinds into thousands of species many of which are as distinct as we are from primates.

Extra notes:

Comparing genomes is very complex because there are so many different way to look at the genome.   There is total genome size, number of chromosomes, amount of repetitive DNA, overall sequence similarity, coding vs non-coding (ie. genes vs non-genic regions) sequence similarity, DNA hybridization, etc..   Even genetic codes that are very similar can be read (translated) into profoundly different phenotypes (physical manifestations of the code). For example, single differences in important genes can result in huge differences even among individual humans but some genes could have many sequences differences and yet result in little difference to the organisms.  For example,the hormone leptin is very similar in structure among mammals despite many individual differences.  A human and a whale have many differences at the level of the DNA sequence  but you could effectively swap the whale gene for your gene and probably be fine.

Because of these complexities the mtDNA genome is a more elegant system to look at differences because it is much easier to compare. However, I am not unaware of some real and potential problems with such comparisons. First, mtDNA genomes don’t all “evolve” at the same rate. The mutation rate in some lineages is known to be a bit faster than others. I could have shown rat/mouse differences but it is known that they have a faster rate of mutation accumulation and so their differences are exaggerated   I do believe that the comparison of fairly similar species that I have done above reflects lineages that have similar mutation rates and thus the numbers really have meaning in relationship to one another.  Second, you have to remember that the actual number of differences is probably greater than you are seeing.  Some mutations that happened in the past could have back mutated to appear exactly like another lineage even though they way they both arrived at a T at a particular position was different and thus that two Ts are not the same.

Another complication is lineage sorting as a result of having a starting population with multiple versions of mtDNA.  Over time if a population gives rise to new species the individuals of that new species will have varying amounts of differences between them because they both started with multiple copies of the mtDNA genomes.  However, for creationists the assumption set is much simpler and lineage sorting doesn’t present as much of a problem.  If all “kinds” were reduced to a single pair in Noah’s ark, then no matter how much variation there was before the flood that variation will go through an extreme bottleneck probably being reduced to a single or several very closely related mtDNA genomes.  Thus all new species formed after the flood in the creation model of fast evolution start from a single version of mtDNA and any variation we see today in populations or between species must be the result of mutation to that initial genome.

I didn’t want to get into functional constraints and directional selection because these are not important factors in these comparisons (another reason that comparing mtDNA genomes is nice).  Someone might say that God might have started with a mtDNA genome from a primate model because it best fit the needs of a human physiology and this account for the similarity.  However, that person would also have to believe that large differences in sheep and goats was created naturally.  Why are these genomes SO different if these species are so similar and obviously share a huge amount of their physiology?   Any argument for the similarity of one set of genomes can’t be divorced from application to another set of genomes.

Below is an example of what a sequence search result looks like in BLAST.  In this case this is a pair of sequences of a gene from pacific oysters.

*This does not hold true when one starts to look at differences between more distantly related animals. For example, the differences between a fish and reptile are about the same as a fish and a human and a butterfly and a human. This is because the genome is so small that many changes are redundant and thus the genome become saturated with changes and for every new change another site in the genome will have converged on the same sequence as something it is not related to.

References:

Soares, P; Ermini, L; Thomson, N; Mormina, M; Rito, T; Röhl, A; Salas, A; Oppenheimer, S; Macaulay, V; Richards, MB (June 2009), “Correcting for purifying selection: an improved human mitochondrial molecular clock”, American Journal of Human Genetics 84 (6): 740–59,doi:10.1016/j.ajhg.2009.05.001, PMC 2694979, PMID 19500773   Comment: in this paper Soares et. al. compared just 2196 complete human mtDNA genomes and from those identify 10,683 separate mutation events.  This includes mutations that have occurred multiple time at the same location in the genome. For example, I might have a mutation of an A to a T at position #1000 in my mtDNA genome.  Another person for whom I have no common ancestor for at least 1000 years might also have the exact same mutation.  Because our common ancestor did not have this mutation we both gained this mutation independently.