In February with. g. Texas forensic scientist Melba Kechum and her colleagues published the results of the analysis of the yeti genome in the specially established Internet magazine De Novo. From the very beginning, it was clear that this was naive amateurism and an attempt to pass off the wishful thinking.
A healthy dose of folklore and a YouTube video (see below), which supposedly depicts a sleeping Bigfoot, are attached to the description of standard methods for working with genetic material. The conclusion from this is the following: Bigfoot exists, and it is not a monkey, but a hybrid of a man with an unknown hominin.
The editors of the Ars Tehnica website were not too lazy to analyze that article in detail and ask Ms. Kechum some unpleasant questions - mainly to find out why this is all. How is it that some see a publication as a carelessly performed study, while others see a major scientific discovery?
The article describes two genomes isolated from samples allegedly related to the Yeti. Mitochondrial DNA is clearly human. And a small part of the nuclear genome is a mixture of human and other sequences, some of which are related to each other, others are not.
The biologists who were consulted by the journalists confirmed the obvious: this is the result of contamination and degradation of samples, as well as careless assembly of DNA. Ms Kechum defends herself: “We did our best to make the article absolutely honest and scientific. I don't know what else you need. All we wanted was to prove their existence, and we succeeded."
Let's first understand why she thinks the job was done flawlessly and why she is wrong.
First of all, we emphasize that the processing and preparation of samples (and this is the main thing in such a case) was carried out by forensic experts. There is no doubt that these people can be called scientists, because forensic medical examination is based on the principle of reproducibility of experimental results, the essence of the cornerstone of the scientific method. But unlike a geneticist, a forensic scientist is focused on results. Apparently, this was the cause of the error.
DNA analysis has been used in forensics for many years, standard procedures have been developed that have proven their worth. But these methods never aim to completely decipher the genome - it is enough to obtain data that can be presented to the court as evidence. And the authors of the "study" really wanted to prove to an imaginary jury that the Yeti exists.
Promotional video:
Pieces of hair of different sizes were used as samples. Hair is a frequent visitor in the laboratories of forensic scientists, who usually need to determine whether it belongs to a person, more precisely, a specific suspect, etc. In this case, the researchers concluded that the hair is not human. Ok, let's move on.
If the hair bears follicles, DNA can be extracted from the cells. This was done - through a standard forensic examination procedure. Equally standard measures were taken to weed out all extraneous DNA.
According to Ms. Kechum, it succeeded: pure DNA of the owner of the hair was obtained, and therefore the conclusions from her analysis are correct, and Bigfoot exists. Just think - the hair is inhuman, and the mitochondrial DNA, which is inherited only through the maternal line, is human!
Yes, sometimes (or rather, almost always - any biologist will tell you) no precautions can insure against error. But what right do we have to say that in this case, the researchers really gave a blunder and analyzed the contaminated samples? It's very simple: the analysis brought up internally conflicting information. It was necessary to see in it an indication of an error and to double-check the results.
The polymerase chain reaction (PCR) method is used to amplify human-specific DNA sequences. By using short sequences that combine with parts of the human genome, multiple copies of a single DNA molecule in a sample can be made and thus easier to detect. In this case, such PCRs were used that allow identifying DNA regions whose lengths vary in different human populations, which is very useful for forensic science.
If human DNA is not very much destroyed, then the reactions should unequivocally testify to this. A similar result is obtained from the analysis of samples containing DNA from close human relatives (recall, the chimpanzee's genome coincides with ours by more than 95%). The farther an animal is from a person, the less often the reactions are triggered, and PCR increasingly often give sequences of the wrong length, because the composition of DNA changes during evolution.
But don't wait for DNA to sit quietly and wait for you to get to it. It easily breaks down into fragments, which become smaller over time, and this also affects the result of reactions.
Having done whatever they saw fit, Ms. Kechum and her colleagues saw chaos. Some reactions gave PCR products of the expected human size, others non-human. It would be logical to interpret the latter as a complete absence of reactions or even genetic material. The picture repeated itself over and over again, and it had to be concluded that either the sample belonged to an animal far removed from humans on the evolutionary tree, or the DNA was severely degraded.
Scientists used electron microscopy and saw short fragments of DNA, some of which had a single (rather than double) helix. The strands bind to some of the pieces and then diverge into single-stranded sections, which re-attach to individual molecules. A similar pattern is observed in the presence of the genetic material of a mammal far from us - the sequences coding for proteins coincide quite well, and the intermediate DNA regions are very different.
In a word, everything indicated that the DNA was poorly preserved and, probably, contaminated. This also suggests that the methods used to obtain pure DNA were insufficient. But the authors decided that in front of them was just a very unusual sample.
But, alas, there are no “just very unusual” samples. Human beings cannot interbreed with other primates. Yes, our ancestors mated with Neanderthals and Denisovans, but they can be considered only half sapiens, because our DNA is very similar. However, instead of getting off the obviously wrong path, the researchers went all out. After finding out that at least part of the DNA is human, they concluded that they had found a hybrid of a human and some other primate.
It is often overlooked in the non-specialized literature that human cells actually contain two genomes. One lives in the chromosomes and is stored in the nucleus - this is what is usually talked about when it comes to the human genome. The second is found in mitochondria, small organelles that produce most of the cellular ATP. These are the descendants of the once free bacteria that billions of years ago firmly linked their lives with cells, but retained a relic of their genome.
Mitochondrial DNA is a valuable tool for tracking populations of humans and other species. Since this genome does not have a complete set of DNA repair tools and is not able to undergo the process of recombination, it mutates much faster than the nuclear one. This leads to the fact that even closely related populations have differences in mitochondrial DNA. In addition, each cell contains hundreds of mitochondria, and each has dozens of copies of the genome. Therefore, it is always relatively easy to obtain samples of mitochondrial DNA, even if the sample is severely degraded or contaminated.
Consequently, by sequencing the mitochondrial genome of their samples, the authors obtained the sequences of a human, and not a primate, only distantly related to humans.
All indications are that successful interbreeding between humans and closely related species (such as Neanderthals and Denisovans) was relatively rare. It makes sense to expect that the creature that resembles a walking rug has less relationship to a person than the aforementioned ones, that is, the chances of interbreeding are further reduced. However, the samples gave different sequences of mitochondrial DNA, which means that the crossing happened a lot of times. In addition, it turns out that the hybrids never crossed with the females of those hypothetical primates. Finally, those primates appear to have become extinct since no example of their mitochondrial DNA has been found.
What kind of women are they who agreed to mate with some unknown monkeys? If you want to keep up with modern scientific ideas, then you should rely on populations that once lived in Asia and whose separate branches later penetrated into America. No other people lived in America (until very recently). But, alas, it was not possible to find an Asian trace in mitochondrial DNA. Most of the sequences are of European origin, and there are also a couple of African examples.
Ms Kechum described one of the samples in detail. According to her, it belongs to a haplotype that arose in Spain about 13 thousand years ago. Hence, hybridization could not have occurred before the appearance of this haplotype.
At first glance, it is impossible to build a coherent hypothesis based on this mess. Obviously, the researchers ended up with human samples, and heavily contaminated ones, which well explains the diversity and age of the sequences.
But let's not forget that for Ms. Kechum, the possibility of an obvious interpretation is out of the question. The authors suggest that during the last ice age, groups of Europeans and Africans (sic!), Wandering over the endless glaciers of the North Atlantic, wandered into North America. Indeed, there is a hypothesis that the hunters of the Solutrean culture crossed the Atlantic on the ice and founded several settlements on the eastern shores of North America, after which they either died out or were assimilated by immigrants from Asia. But Mrs. Kechum, for some reason, did not like this assumption. She does not exclude the possibility that crossbreeding could have occurred in Europe, after which Bigfoot somehow ended up in America - most likely, over a land bridge at the site of the Bering Strait. “They could cross the whole world on foot,” says the researcher. "They are so fast!"
In any case, according to Ms. Kechum, these are insignificant details: “We do not know how they got here. We just know they did it."
So, until now, the oddities of the study have not been related to the method, but to the interpretation of the results. But as soon as the authors began to study specific genome sequences, really serious problems began. Some samples carried enough DNA to be sequenced on one of the high-throughput platforms. The quality score indicated that there were enough sequences to assemble them into the genome. (Oddly enough, scientists have mistakenly interpreted this as indicating that they are in front of the sequences of the same individual.)
High-throughput machines typically produce short sequences of only a hundred bases, whereas even the smallest human chromosome has more than 40 million bases. There are programs that can recognize when a duet of such fragments of 100 bases partially overlaps each other and it becomes possible to assemble a new fragment from them - already from 150, for example, bases. Searching for such overlays allows you to gradually build up fragments of several million base pairs. This method is imperfect (it leaves "holes" in the genome), but it is convenient and widely used.
For some incomprehensible reason, our heroes did not use it. Instead, they took one human chromosome and, using computer programs, tried to assemble something similar from the material at their disposal.
But most of the mammalian DNA coding regions are conservative, and therefore they line up beautifully on the human chromosome, while the inappropriate part of the genome is ignored. In other words, this approach is almost guaranteed to give something similar to the human genome.
Another problem with this method is that the software usually considers the chromosome sequence of a person as a goal that must be achieved. If the program doesn't find a perfect match, it will look for the best available one.
Even so, it was not possible to collect the entire chromosome. The computer produced three sections of the chromosome of several hundred thousand base pairs each, and the human genome, recall, contains more than three billion of them. Given that the DNA quality indicator was high, we can talk about two things: either the software was chosen poorly, or there was little human DNA.
Let's stop for a while and try to imagine that the conclusions of the authors are correct, that is, Bigfoot exists and is the fruit of love of people with some unidentified hominid. It is known that our ancestors interbred with Neanderthals and Denisovans, and the result of those marriages, naturally, should have been something like a man, because Neanderthals and Denisovans were very similar to humans. Consequently, to produce Bigfoot, humans had to interbreed with some more distant relative, but not as distant as the chimpanzee.
What would the genome of such a hominin look like? The genomes of Neanderthals and Denisovans are very similar to humans. The genome of "hominin X" should be more different from ours, but not more than the genome of chimpanzees. In terms of the large-scale structure, the human and chimpanzee genomes are almost identical, with only six locations with a large structural difference, that is, a total of 11 breakpoints. And none of these points are on the 11th chromosome, which the authors were trying to reconstruct, so that's okay.
Smaller insertions and deletions are more widespread, but not much. If we focus on those regions of the genome where the above-mentioned large regions dividing humans and chimpanzees do not exist, then our genomes coincide with them by 99%. It can be assumed that the genome of the hominin, with which our ancestor could interbreed, should coincide with ours by 97–98%.
Hybrids of the first generation will have the genomes of their parents in a ratio of 50 to 50. Of course, natural selection will have its say, but in general, about 90% of the human genome is not under selective pressure, and the bulk of the rest, too, does not fall under it simply because that it is identical in both parents. As a result, 98–99% of the genes of both species will be randomly inherited.
Of course, after the first generation, the recombination of two genomes will begin, the unit of which is centimorgan. 1 cM corresponds to the distance between genes, the recombination between which occurs with a frequency of 1%. If you have 50 million DNA base pairs, then there is an equal chance of recombination and no recombination with each generation. In humans, one generation is an average of about 29 years, in chimpanzees - 25. It can be assumed that the Bigfoot is around 27 years old.
If the Yeti arose about 13 thousand years ago, then about 481 generations have changed since then. This means 241 recombinations. On average, we'll see one sign of recombination for every 200k base pairs or something.
So, we know what the hybrid genome should look like: sections of human DNA over 100 thousand bases in length alternate with segments of the same size that are similar to human ones, but still different from them. The similarities between the one and the other must be very strong, so it would be difficult to tell where the human sequence ends and the non-human begins. To deal with this, you would have to use equipment with a resolution in the region of a thousand base pairs. And since mitochondrial DNA hints at multiple episodes of interbreeding, no Yeti would share the same combination of human and non-human regions.
The genome that Ms. Kechum and her colleagues tried to foist on us is completely different. Human patches are only a few hundred base pairs long. They are mixed with regions that are completely unlike human ones. Such a genome does not support the hybrid hypothesis in any way. But Mrs. Kechum stands her ground: after all, hybridization could have occurred much more than 13 thousand years ago.
Employees of Ars Technica decided to independently find out what this genome is on the ENSEMBL website. The BLAST software showed that chromosome 11 corresponds best to this set of sequences, which is to be expected. But, as we remember, along with human there were some other areas. If we are really a hybrid, then they should at least resemble human ones, but nothing of the kind - for some, no matches were found in the database, while others, as it turned out, belong to bears, mice and rats, that is, we have ordinary human samples in front of us, very heavily contaminated with the DNA of animals common in the forests of North America.
Reanalysis of one of the samples in another laboratory led skeptical geneticists to a similar conclusion.
When the computer was asked to assemble the 11th human chromosome from this hodgepodge, it found the most suitable fragments, and filled in the gaps between them with everything - sometimes with human material, sometimes not.
As you can see, the authors of the study did not aim to find out what was in front of them. They proceeded from the firm belief that these were pure yeti DNA samples, and all the oddities were attributed to this. And the reason is that Mrs. Kechum saw Bigfoot with her own eyes and was eager to tell the world about it. According to her, she met with enthusiasts who live in some secret place where they can "communicate" with the yeti. The latter are accustomed to people and approach them at such a distance from which they can be viewed.
She claims that such "meeting places" are kept secret so that journalists, hunters and zoo owners are not interested in the snowman. In short, they want to protect the Bigfoot from excessive curiosity that will not do them good. But one day, human mitochondrial DNA was found in a hair that was identified as non-human, and Ms. Kechum wanted to put the skeptics to shame.
And she considers all the absurdities of her research a mystery that deserves further research. “The results are as they are, and I'm not going to fit them into conventional scientific models if they don't fit,” she says.