Friday, January 3, 2014

Chinese Researchers Demolish Evolutionary Pseudo-Science

Science Versus Religion

In recent decades the genomes of several species have been mapped out and evolutionists are using these genome data to refine their theory. They are also making some high claims. The genome data sets, say evolutionists, are adding striking new confirmations for their theory. One piece of evidence evolutionists point to is the high similarity between the human and chimpanzee genomes. The two genomes are about 95% the same and evolutionists say this shows how easily the human could have evolved from a chimp-human common ancestor. Evolution professor Dennis Venema explains:

For example, humans and our closest relatives, chimpanzees, have genomes that are around 95% identical, and most of the DNA differences are not differences that actually affect our forms. So, small changes accruing over time since we last shared a common ancestor was enough to shape our species since we parted ways – there is no evidence that evolution requires radical changes at the DNA level.

No evidence? This is an example of evolutionists seeing what they want to see in the data. Evolutionists are driven by their metaphysics and so want to believe that we are descended from a primitive ape creature. They want to believe that humans and apes “are one” and that the wall between human and animal “has been breached,” as the Smithsonian Institute put it.

But as I pointed out in my book Darwin’s Proof, if the DNA comparisons between human and chimp don’t reveal much significant difference, then we probably need to look elsewhere. Humans are vastly different than chimps and if our DNA comparisons aren’t revealing much difference, then those segments probably aren’t what is driving the difference between the two species.

In fact there are much more significant differences between the human and chimp genomes. Differences that may “actually affect our forms.” A 2011 paper out of China and Canada, for example, found 60 protein-coding genes in humans that are not in the chimp. And that was an extremely conservative estimate. They actually found evidence for far more such genes, but used conservative filters to arrive at 60 unique genes. Not surprisingly, the research also found evidence of function, for these genes, that may be unique to humans.

If the proteins encoded by these genes are anything like most proteins, then this finding would be another major problem for evolutionary theory. Aside from rebuking the evolutionist’s view that the human-chimp genome differences must be minor, 6 million years simply would not be enough time to evolve these genes.

In fact, 6 billion years would not be enough time. The evolution of a single new protein, even by evolutionists’ incredibly optimistic assumptions, is astronomically unlikely, even given the entire age of the universe to work on the problem.

Unfortunately none of this will influence the evolutionist because for evolutionists this never was about science. As Venema explains:

It’s one thing to explain away biogeographical patterns or claim that anatomical similarities reflect a non-evolutionary “design” pattern – but another thing altogether to attempt to explain away why humans (and other placental mammals) have a defective gene for making egg yolk in the exact spot in our genomes where chickens have the functional version of this gene, and that humans and chimpanzees share a large number of mutations in common in our two inactivated copies.

The argument from dysteleology says that these faulty genetics would not have been designed or created and that therefore they must have evolved. This argument is not new. It did not arise when the genomic data became available but has been influential for centuries. Earlier evolutionists found faults with all manner of biological, geological and cosmological aspects of nature.

This reasoning is not new and it is not science. It is based on personal religious beliefs that are not open to debate. Imagine if you believed these things. Imagine that you believed, with Venema, that common mutations, for example, rules out any possibility of the species having been created in any sort of direct sense.

Then of course you would be an evolutionist. Even though evolutionary theory fails every test. With evolution, the religion drives the science.

In the meantime, while the evolutionists make up rules for science to follow and insist the world spontaneously arose in spite of the evidence, these researchers in China and Canada are doing real science.

67 comments:

  1. In fact, 6 billion years would not be enough time. The evolution of a single new protein, even by evolutionists’ incredibly optimistic assumptions, is astronomically unlikely, even given the entire age of the universe to work on the problem.

    Since the new protein nylonase evolved in less than 60 years, you are out by 8 orders of magnitude.

    That is impressive, but not a record, since Lee Strobel's claims regarding the known precision of the cosmological constant are out by thirty orders of magnitude. Try harder.

    Roy

    ReplyDelete
    Replies
    1. Roy:

      If proteins can evolve so quickly then why are researchers having so much difficulty in their attempts to demonstrate protein evolution?

      Delete
    2. Are they? Anyway, it's irrelevant. Nylon-degrading proteins appeared mere decades after nylon did, not billions of years later.

      Roy

      Delete
    3. Ch,
      If proteins can evolve so quickly then why are researchers having so much difficulty in their attempts to demonstrate protein evolution?

      Your claim was it took a very long time to evolve a protein, not why Axe and Gauger had difficulties.

      Delete
    4. R: Nylon-degrading proteins appeared mere decades after nylon did, not billions of years later.

      J: Even if this is true, how do you know they weren't generated by a cellular program pre-designed to produce them under those conditions? That's what you'd have to know. HOW, in other words, did cells blindly evolve immune sytsems, error-correction systems, etc? We already know designed programs routinely do things that AREN'T seemingly done routinely apart from design. That's the whole basis of the distinction, for crying out loud.

      Delete
    5. V: I'm going by the science, which shows that it takes a "very long time to evolve a protein." Maybe we'll learn more in the future, but that is what the current science shows.

      Delete
    6. Jeff: You have correctly pointed out the equivocation.

      Delete
    7. Jeff: Even if this is true, how do you know they weren't generated by a cellular program pre-designed to produce them under those conditions?

      The answer to this is the same as in Russell's teapot problem. You go and find the program.

      Delete
    8. oleg:

      The answer to this is the same as in Russell's teapot problem.

      No, not quite. The evolutionary claim doesn’t make sense, but in this case it is not unfalsifiable, as is the claim in Russell's teapot problem. Nylonase doesn’t solve the problem of protein evolution any more than adaptive mutations or epigenetics solves the problem of macroevolution. Evolution insisted on natural selection acting on random variation (random with respect to need) which requires long time periods, but what we find over and over is rapid, directed response that addresses the need. Very much like physiological responses.

      Delete
    9. No one says that nylonase solves "the problem of protein evolution." It is, however, an example of a protein that evolved rather quickly. Something you said couldn't happen.

      Jeff suggests that perhaps they did not evolve through the usual means assumed in evolutionary biology (a well understood mechanism of copying errors and natural feedback). The burden is on him to establish that there was a program that allegedly guided the evolution of this protein.

      Delete
    10. oleg:

      No one says that nylonase solves "the problem of protein evolution." It is, however, an example of a protein that evolved rather quickly.

      Why not? If it is an example of a protein that evolved "rather quickly," then why is that not a solution to the problem of protein evolution?

      Jeff suggests that perhaps they did not evolve through the usual means assumed in evolutionary biology (a well understood mechanism of copying errors and natural feedback). The burden is on him to establish that there was a program that allegedly guided the evolution of this protein.

      So to recap, you claim the biological world arose spontaneously via chance events, requiring billions of years, a new protein arises in a matter of a few years in direct response to an environmental change, and yet the burden of proof is--get this--on me, to *prove* it wasn't a one-in-a-gazillion, astronomically lucky shot. Otherwise evolution, as always, stands firm. Too much.

      Delete
    11. It's your assertion, Cornelius, that protein evolution is extremely unlikely. Here is a protein that has evolved quickly. There is a paper describing the evolutionary pathway. If you wish to argue, as Jeff did, that the evolution of this protein was guided by some hidden program, the burden is on you to demonstrate the program's existence.

      Delete
    12. There is a paper describing the evolutionary pathway.

      Key word: "evolutionary". There is no scientific evidence that this pathway was at all probable under evolution's unguided search rules. Thus we cannot safely label the pathway as "evolutionary."

      Delete
    13. Can you estimate how improbable it is? Show your work.

      Delete
    14. Can you estimate how improbable it is? Show your work.

      Great question. It would be good to estimate the probability so the problem becomes more clear to people. I don't know of any such estimate, and will try to do this.

      Delete
    15. It's great that you will try to do that, Cornelius. But it naturally raises the question: if you don't know of any estimates, on what basis do you say that this is improbable?

      Delete
    16. You don't necessarily need a numerical estimate to know something is improbable.

      Delete
    17. Indeed, why would a UCA'ist want to work on such an estimate? That would work contrary to the research agenda.

      Delete
    18. oleg:

      So it's worse than I thought. The ability to chop up nylon, in this case, is achieved with merely two amino acid swaps to a preexisting gene. Not quite the evolution of a de novo gene! So this example is not relevant to the new human genes, as Roy had suggested.

      Delete
    19. If it takes just a couple of AA swaps to generate a novel protein, a new, functional, not previously in existence protein, this is bad news for you, not for me.

      It was you, Cornelius, who declared that evolution of novel proteins is "astronomically unlikely." Turns out it can be a piece of cake.

      Delete
    20. oleg:

      No, this is *not* a novel protein. Perhaps I didn't make that clear enough.

      Delete
    21. Why is it not a novel protein? Was there previously an enzyme that could help bacteria digest nylon? I don't think so.

      Delete
    22. A novel protein is a protein which has no known homologous proteins, which would have similar fold and would have, say, 20% sequence identity or more. The paper in the OP used conservative filters to search for novel proteins in humans, and even then found 60. They nylon protein, OTH, has 88% sequence identity with a pre existing protein! That is definitely not novel. And to make matters worse, all that actually is needed is 2 amino acids to be changed.

      Delete
    23. I see what you mean by "novel." It is a protein made by a newly originated (de novo) gene. Nylonase is coded by a gene that mutated from a previously expressed gene, so its gene isn't de novo. So by that definition, nylonase is not a novel protein.

      That said, I think you overestimate the difficulty of genes originating de novo. They aren't made from scratch. They can come from previously unused, non-coding regions of the DNA. The paper you cite refers to previous workers showing that this seems to have happened in Drosophila. The current paper identified non-coding genes in chimpanzee and orangutan orthologous to the human genes coding for the new proteins. It didn't require a complicated sequence of mutations, so this mode of creating new proteins seems to be fairly frequent.

      Delete
    24. That said, I think you overestimate the difficulty of genes originating de novo. They aren't made from scratch. They can come from previously unused, non-coding regions of the DNA.

      This is tricky because some “findings” are based on the assumption evolution occurred, such as:


      The paper you cite refers to previous workers showing that this seems to have happened in Drosophila.

      So the Drosophila paper “shows” that de novo protein-coding genes evolved from non-coding regions by comparing to similar genomes and eliminating other possibilities (such as duplication and mutation). Since they assume common descent, then therefore the protein in question must have evolved de novo.


      The current paper identified non-coding genes in chimpanzee and orangutan orthologous to the human genes coding for the new proteins. It didn't require a complicated sequence of mutations, so this mode of creating new proteins seems to be fairly frequent.

      Well it is not clear just how complicated the mutations would have to be. We don’t know the details on these 60 proteins. They could be fairly simple, as the OP caveated. So yes, it is tempting to see these de novo genes as not that difficult to evolve. That could be.

      We just don’t know for sure. That is why protein studies and experiments are more informative than comparative genomics that assume evolution. What we do know from these experiments is that evolving a mature protein from a random sequence starting point is not feasible. Perhaps that will change with future studies, but for now we have this as a solid empirical finding, not dependent on assumptions that evolution occurred.

      Delete
    25. If I understand correctly it is no big deal to create newish proteins which preform a new function. So what happens if the child protein has two more substitutions and creates a new protein that is now four from the original. Novel with respect to the original protein, yet? In other words is it the distance from where you started or the distance from the last step?

      Delete
    26. V:

      Well let me rephrase that. Creating a new protein from scratch (ie, a random DNA sequence) is not feasible, according to today's science. Modifying an existing protein is far more believable. For example, genes can duplicate and then undergo mutations, potentially leading to a new protein. Even here, though, there are a couple of items to note. First, from an evolutionary perspective, you need to have things happen, such as gene duplication. That is not a trivial process. So evolution would have had to create this process, which then became a crucial mechanism of evolution. Secondly, modifying an existing protein is not so simple if you need multiple mutations to occur in order to produce the new function. In the nylon protein example, two mutations were required. Either one alone provided very little function. And getting two (or more) mutations to occur more or less together is less likely than getting a single mutation to occur.

      Delete
    27. CH: Well let me rephrase that. Creating a new protein from scratch (ie, a random DNA sequence) is not feasible, according to today's science. Modifying an existing protein is far more believable.

      Evolution doesn't begin from scratch every time, Cornelius. De novo genes described in that paper do not start from a random DNA sequence. Their predecessors are non-coding regions of the DNA. It's junk, in other words, that was previously working.

      CH: In the nylon protein example, two mutations were required. Either one alone provided very little function. And getting two (or more) mutations to occur more or less together is less likely than getting a single mutation to occur.

      Indeed, two mutations are less likely than one. The process, however, is not astronomically unlikely. It has been observed in a lab, see Lenski's E. coli experiment.

      Delete
    28. Sounds reasonable. That is the significance of Lenski's ongoing experiment, right? Both the abilty to turn back the clock on evolution and the actual observation of how such a two mutation jump works?

      Delete
    29. O: There is a paper describing the evolutionary pathway.

      J: The pathway REQUIRES as conditions all the machinery that is ALSO dependent on DNA sequences. You've explained nothing in terms of rendering your hypothesis probable in the posited TIME-FRAME.

      I assure you, the burden of proof can't be on the person who can at least account for a distinction between warranted/plausible and unwarranted/plausible beliefs. Benevolent/competent teleology is required to account for that distinction until someone can do it otherwise. And intuitively recognizing this, most people realize the "appearance of design" is just analogical explanation that works fine, for now, in the ABSENCE of any other explanation, however far back that last libertarian cause would have to be for such analogical teleological explanation. Induction says we push it back as far as possible. Right now, that's not very far back since we have no naturalistic model that explains the relevant events. And yet most people want to use some criteria for inferring natural causality, like analogous degrees of observed variation from artificial selection etc. But none of that warrants an inference to UCA.

      Delete
  2. I think the "common dysteleology" argument, if the assumptions involved are true, is a decent argument against a big "D" designer of those sequences. But there are at least three problems with using that argument at all:

    1) As CH mentioned, we have precisely zero reason for believing that the accumulation of extant functional sequences in the posited time-frame is even possible, much less probable.

    2) In some cases, these arguments have been shown to be based on ignorance of extant function and are therefore just guesses to some extent. They're not known facts.

    3) Over-all viability-diminishing mutations are known to result in a lesser of evils for certain environments.

    In the meanwhile, benevolent/competent design of the universe is the only explanation for the validity of inductive criteria in the first place. UCA'ists (and radical/speedy transformationists, in general) depend on the argument from dysteleology. Unfortunately, there's no way they can know their assumptions are true. And in some cases, they've already been proved wrong. The argument ALWAYS amounts to deductions from metaphysical assumptions which are, per that very metaphysics, utterly a-plausible.

    ReplyDelete
  3. "The argument from dysteleology says that these faulty genetics would not have been designed or created and that therefore they must have evolved. This argument is not new. It did not arise when the genomic data became available but has been influential for centuries. Earlier evolutionists found faults with all manner of biological, geological and cosmological aspects of nature."

    It may well be my own fault but I can only admit I do not understand what you write. Many questions come to my mind but I'll be brief and only ask if you can name one or more of the "earlier evolutionists" referred to?

    ReplyDelete
    Replies
    1. Rolf:

      Thank you for the comment and please feel free to ask more questions about this. Here are some earlier examples:

      "Odd arrangements and funny solutions are the proof of evolution—paths that a sensible God would never tread but that a natural process, constrained by history, follows perforce. No one understood this better than Darwin. Ernst Mayr has shown how Darwin, in defending evolution, consistently turned to organic parts and geographic distributions that make the least sense." -Stephen J. Gould

      "Embryonic stages don't look like the adult forms of their ancestors, as Haeckel claimed, but like the embryonic forms of ancestors. Human fetuses, for example, never resemble adult fish or reptiles, but in certain ways they do resemble embryonic fish and reptiles. Also, the recapitulation is neither strict nor inevitable: not every feature of an ancestor’s embryo appears in its descendants, nor do all stages of development unfold in a strict evolutionary order. Further, some species, like plants, have dispensed with nearly all traces of their ancestry during development. … Yet we shouldn’t throw out the baby with the bathwater. Embryos still show a form of recapitulation: features that arose earlier in evolution often appear earlier in development. And this makes sense only if species have an evolutionary history. Now, we’re not absolutely sure why some species retain much of their evolutionary history during development. The “adding new stuff onto old” principle is just a hypothesis—and explanation for the facts of embryology. It’s hard to prove that it was easier for a developmental program to evolve one way rather than another. But the facts of embryology remain, and make sense only in light of evolution." -Jerry Coyne

      "This designer has been busy! And what a stickler for repetitive work! Although no fossil of the Indian elephant has been found that is older than 1 million years, in just the last 4 million years no fewer than nine members of its genus, Elephas, have come and gone. We are asked to believe that each one of these species bears no relation to the next, except in the mind of that unnamed designer whose motivation and imagination are beyond our ability to fathom. Nonetheless, the first time he designed an organism sufficiently similar to the Indian elephant to be placed in the same genus was just 4 million years ago—Elephas ekorensis. Then, in rapid succession, he designed ten (count’em!) different Elephas species, giving up work only when he had completed Elephas Maximus, the sole surviving species." -Ken Miller

      "The recurrent laryngeal nerve is a remarkable piece of unintelligent design. The nerve starts in the head, with the brain, and the end organ is the larynx, the voice box. But instead of going straight there it goes looping past the voice box. In the case of the giraffe, it goes down the full length of the giraffe’s neck, loops down one of the main arteries in the chest and then comes straight back up again to the voice box, having gone within a couple of inches of the voice box on its way down. No intelligent designer would ever have done that." -Richard Dawkins

      Delete
  4. Cornelius, I hear you. By saying that the position and defects of human egg yolk genes is evidence consistent with a common ancestor, evolutionists are just trying to manipulate logic and facts to make us change our minds. But like you, I'm having none of that.

    ReplyDelete
    Replies
    1. No fair not reading.

      It’s one thing to explain away biogeographical patterns or claim that anatomical similarities reflect a non-evolutionary “design” pattern – but another thing altogether to attempt to explain away why humans (and other placental mammals) have a defective gene for making egg yolk in the exact spot in our genomes where chickens have the functional version of this gene, and that humans and chimpanzees share a large number of mutations in common in our two inactivated copies.

      Delete
    2. I thought you were quoting that pesky evolutionist Venema on egg yolk genes, because you disagreed. Please tell me that you don't agree that the egg yolk gene evidence is consistent with a common ancestor. I want to believe, no, I NEED to believe, that evolution is not true.

      Delete
    3. hi myron. actually a share mistakes on pseudogenes can be evolve independently without commondescent:

      for exmple: the oux pseudogene in both gorila and orang utan share stop codon at codon 107 and also a GGGATGCC duplication at codon 911 ,but both human and chimp dont have them . so this is evidence of "hot spots" in the genome according to the evolution scientist because its contradict the phylogenetic tree . but again- if we can get share mistake without commondescent, then share mistake cant prove commondescent. simple logic

      Delete
    4. dcscccc: Small steps. You (and Cornelius) are not giving me the simple reassuring comfort I requested. Is the egg yolk gene evidence consistent with a common ancestor? Yes or no?

      Delete
    5. sorry for my english. but what you mean by that? if i understand you correctly so my answer is not.

      Delete
    6. Chickens and us share a gene related to making an egg yolk. Ours is broken. Is that evidence consistent with the idea that chickens and us have a common ancestor? (I'm _not_ asking if that evidence proves a common ancestor, or if you have other evidence that disproves common ancestry.)

      Delete
    7. sorry but what you mean by "consistent"? that this evidence for commondescent of human and chiken? by the way- how can we tel if this gene its realy a gene for egg yolk? and 2- why its not possible that mammal have in the past a little yolk?

      have a nice day

      Delete
    8. Use the dictionary definition. "Consistent": in agreement. "Seeing a black crow is consistent with the claim that all crows are black."

      Please save your other questions until you understand my much simpler question.

      Delete
    9. if i understand right. so yes.

      Delete
    10. dcscccc: Thank you for taking the time to understand my question and responding that at least one piece of evidence supports evolution, but just some tips for dealing with evolutionists:

      1. Never answer an evolutionist's direct simple question.
      Remember: Evade, Equivocate, Obfuscate, then Iterate.

      2. Never agree with anything an evolutionist says, no matter how obvious.

      3. When trapped or all else fails, feign righteous indignation.

      You're welcome.

      Delete
    11. hi myron. where i said that this kind of claim support evolution?

      Delete
    12. dcscccc: Because of your response, "if i understand right. so yes."
      BTW, good job of taking my previous post to heart.

      Delete
    13. so i understand you wrong. see below me coment in the ends.

      Delete
  5. Dr. Hunter, although the 99% genetically similar figure is still bandied about on the internet, even the 95% figure you quote is far too generous. Supposed genetic similarity between chimps and humans has probably been the most manipulated and abused piece of evidence that Darwinists put forth (which is really saying something considering their abuse of all the other evidence). ,,, First let’s focus on what Darwinian processes can actually explain in terms of genetic differences:

    Thou Shalt Not Put Evolutionary Theory to a Test – Douglas Axe – July 18, 2012
    Excerpt: “For example, McBride criticizes me for not mentioning genetic drift in my discussion of human origins, apparently without realizing that the result of Durrett and Schmidt rules drift out. Each and every specific genetic change needed to produce humans from apes would have to have conferred a significant selective advantage in order for humans to have appeared in the available time (i.e. the mutations cannot be ‘neutral’). Any aspect of the transition that requires two or more mutations to act in combination in order to increase fitness would take way too long (>100 million years).
    My challenge to McBride, and everyone else who believes the evolutionary story of human origins, is not to provide the list of mutations that did the trick, but rather a list of mutations that can do it. Otherwise they’re in the position of insisting that something is a scientific fact without having the faintest idea how it even could be.” Doug Axe PhD.
    http://www.evolutionnews.org/2012/07/thou_shalt_not062351.html

    More from Ann Gauger on why humans didn’t happen the way Darwin said – July 2012
    Excerpt: Each of these new features probably required multiple mutations. Getting a feature that requires six neutral mutations is the limit of what bacteria can produce. For primates (e.g., monkeys, apes and humans) the limit is much more severe. Because of much smaller effective population sizes (an estimated ten thousand for humans instead of a billion for bacteria) and longer generation times (fifteen to twenty years per generation for humans vs. a thousand generations per year for bacteria), it would take a very long time for even a single beneficial mutation to appear and become fixed in a human population.
    You don’t have to take my word for it. In 2007, Durrett and Schmidt estimated in the journal Genetics that for a single mutation to occur in a nucleotide-binding site and be fixed in a primate lineage would require a waiting time of six million years. The same authors later estimated it would take 216 million years for the binding site to acquire two mutations, if the first mutation was neutral in its effect.
    Facing Facts
    But six million years is the entire time allotted for the transition from our last common ancestor with chimps to us according to the standard evolutionary timescale. Two hundred and sixteen million years takes us back to the Triassic, when the very first mammals appeared. One or two mutations simply aren’t sufficient to produce the necessary changes— sixteen anatomical features—in the time available. At most, a new binding site might affect the regulation of one or two genes.
    http://www.uncommondescent.com/intelligent-design/more-from-ann-gauger-on-why-humans-didnt-happen-the-way-darwin-said/

    ReplyDelete
  6. Now to actual genetic differences. Dr. Sternberg highlights the potential for abuse here:

    Guy Walks Into a Bar and Thinks He’s a Chimpanzee: The Unbearable Lightness of Chimp-Human Genome Similarity – 2009
    Excerpt: One can seriously call into question the statement that human and chimp genomes are 99% identical. For one thing, it has been noted in the literature that the exact degree of identity between the two genomes is as yet unknown (Cohen, J., 2007. Relative differences: The myth of 1% Science 316: 1836.). ,,, In short, the figure of identity that one wants to use is dependent on various methodological factors.
    http://www.evolutionnews.org/2009/05/guy_walks_into_a_bar_and_think.html

    In late 2011 Jeffrey P. Tomkins, using an extremely conservative approach, (and not nearly as much bias as Darwinists have) reached the figure of 87% similarity:

    Genome-Wide DNA Alignment Similarity (Identity) for 40,000 Chimpanzee DNA Sequences Queried against the Human Genome is 86–89% – Jeffrey P. Tomkins – December 28, 2011
    Excerpt: A common claim that is propagated through obfuscated research publications and popular evolutionary science authors is that the DNA of chimpanzees or chimps (Pan troglodytes) and humans (Homo sapiens) is about 98–99% similar. A major problem with nearly all past human-chimp comparative DNA studies is that data often goes through several levels of pre-screening, filtering and selection before being aligned, summarized, and discussed. Non-alignable regions are typically omitted and gaps in alignments are often discarded or obfuscated.
    In an upcoming paper, Tomkins and Bergman (2012) discuss most of the key human-chimp DNA similarity research papers on a case-by-case basis and show that the inclusion of discarded data (when provided) actually suggests a DNA similarity for humans and chimps not greater than 80–87% and quite possibly even less.
    http://www.answersingenesis.org/articles/arj/v4/n1/blastin

    Then earlier this year, 2013, with better resolution of data, and still using an extremely conservative approach (I don’t even think he included ORFan genes in his comparison), Tomkins reached the figure of 70% genetic similarity between chimps and humans:

    Comprehensive Analysis of Chimpanzee and Human Chromosomes Reveals Average DNA Similarity of 70% – by Jeffrey P. Tomkins – February 20, 2013
    Excerpt: For the chimp autosomes, the amount of optimally aligned DNA sequence provided similarities between 66 and 76%, depending on the chromosome. In general, the smaller and more gene-dense the chromosomes, the higher the DNA similarity—although there were several notable exceptions defying this trend. Only 69% of the chimpanzee X chromosome was similar to human and only 43% of the Y chromosome. Genome-wide, only 70% of the chimpanzee DNA was similar to human under the most optimal sequence-slice conditions. While, chimpanzees and humans share many localized protein-coding regions of high similarity, the overall extreme discontinuity between the two genomes defies evolutionary timescales and dogmatic presuppositions about a common ancestor.
    http://www.answersingenesis.org/articles/arj/v6/n1/human-chimp-chromosome

    ReplyDelete
    Replies
    1. Tomkins' result of 70% is erroneous. In short, there was a bug in the version of BLAST+ that he used (v2.2.27), and it had the effect of severely understating his results.

      It's likely that the same bug affected his 86%-89% result as well.

      Delete
    2. The problem with this is "On the other hand mutations can also modify large sections of DNA, physically moving, reversing, adding, and deleting them. When DNA has been removed or added in one lineage but not the other, like in the above, we would get a section that does not “align,” or align poorly. These sections are important when considering the overall picture of differences between two species’ genomes, but are rightly discarded when percentages and splitting times are calculated.

      To Jeffrey Tomkins and his supporters this is apparently a conspiracy – the selective omission of contrary evidence. His own research derives the 70% figure in part by leaving some of these sections in. The trouble is that sections that don’t align aren’t comparable: the deletion of a continuous segment of 1000 bases are not equivalent to 1000 individual nucleotide deletions, because the former can be removed a single stroke – they amount to a single mutation, and happen by a different mechanism at a different rate to the latter variety. Tomkins is entitled to think that his figure is a more honest reflection of the true differences between humans and chimps at the genetic level if he wants to – but more on that some other time, maybe – but it is not useful when we want to calculate how long ago humans and chimps split from each other."
      He makes more mistakes as explained here: http://www.reddit.com/r/NaturalTheology/comments/2625uu/my_first_reply_to_jeffrey_tomkins/
      I've compared chimp and human sequences, they match pretty well.

      Delete
    3. Hi Jacob, please see my comment above. Tomkins ran into a bug in the BLAST software - and that gave him his 70%. Correcting for the effects of the bug gives a result around 96.90% similarity.

      I've submitted a paper to Answers In Genesis' Journal (way back in September 2014). It is still sitting in peer-review. Tomkins is aware of the problem, and has not responded since December 2014.

      See my paper here:

      https://www.dropbox.com/sh/dm2lgg0l93sjayv/AAATnWSJdER53EYEYZvcgiwma?dl=0

      Delete
  7. And though you quote the 60 ORFan gene paper DR. Hunter, the actual number of ORFan genes in humans is far higher than that figure: This following article, which has a direct bearing on the 98.8% genetic similarity myth, shows that over 1000 ‘ORFan’ genes, that are completely unique to humans and not found in any other species, and that very well may directly code for proteins, were stripped from the 20,500 gene count of humans simply because the evolutionary scientists could not find corresponding genes in primates. In other words evolution, of humans from primates, was assumed to be true in the first place and then the genetic evidence was directly molded to fit in accord with their unproven assumption. It would be hard to find a more biased and unfair example of practicing science!

    Human Gene Count Tumbles Again – 2008
    Excerpt: Scientists on the hunt for typical genes — that is, the ones that encode proteins — have traditionally set their sights on so-called open reading frames, which are long stretches of 300 or more nucleotides, or “letters” of DNA, bookended by genetic start and stop signals.,,,, The researchers considered genes to be valid if and only if similar sequences could be found in other mammals – namely, mouse and dog. Applying this technique to nearly 22,000 genes in the Ensembl gene catalog, the analysis revealed 1,177 “orphan” DNA sequences.,,, the researchers compared the orphan sequences to the DNA of two primate cousins, chimpanzees and macaques. After careful genomic comparisons, the orphan genes were found to be true to their name — they were absent from both primate genomes.
    http://www.sciencedaily.com/releases/2008/01/080113161406.htm

    Jerry Coyne, the grand inquisitor of Darwinism himself, states that 6% of genes are ORFans:

    From Jerry Coyne, More Table-Pounding, Hand-Waving – May 2012
    Excerpt: “More than 6 percent of genes found in humans simply aren’t found in any form in chimpanzees. There are over fourteen hundred novel genes expressed in humans but not in chimps.”
    Jerry Coyne – ardent and ‘angry’ neo-Darwinist – professor at the University of Chicago in the department of ecology and evolution for twenty years. He specializes in evolutionary genetics.
    - per ENV

    Helen Pilcher, also no friend of ID, states that up to a third of genes in each new genome sequenced are ORFan genes

    Genes from nowhere: Orphans with a surprising story – 16 January 2013 – Helen Pilcher
    Excerpt: When biologists began sequencing genomes they discovered up to a third of genes in each species seemed to have no parents or family of any kind. Nevertheless, some of these “orphan genes” are high achievers (are just as essential as ‘old’ genes),,,
    But where do they come from? With no obvious ancestry, it was as if these genes appeared out of nowhere, but that couldn’t be true. Everyone assumed that as we learned more, we would discover what had happened to their families. But we haven’t-quite the opposite, in fact.,,,
    The upshot is that the chances of random mutations turning a bit of junk DNA into a new gene seem infinitesmally small. As the French biologist Francois Jacob wrote 35 years ago, “the probability that a functional protein would appear de novo by random association of amino acids is practically zero”.,,,
    Orphan genes have since been found in every genome sequenced to date, from mosquito to man, roundworm to rat, and their numbers are still growing.
    http://ccsb.dfci.harvard.edu/web/export/sites/default/ccsb/publications/papers/2013/All_alone_-_Helen_Pilcher_New_Scientist_Jan_2013.pdf

    ReplyDelete
    Replies
    1. Wow - did you actually read that ORFan gene paper? [Human Gene Count Tumbles Again]

      The conclusion was that these were simply random open reading frames, NOT genes:

      "Here, we provide strong evidence to show that the vast majority of the nonconserved ORFs are spurious. [...] By studying their properties in primates, we show that the vast majority are neither (i) ancestral genes lost in mouse and dog nor (ii) novel genes that arose after divergence from mouse or dog."

      Delete
    2. I assume here that Jerry Coyne is referring to Demuth's 2006 paper about gene families, when he talks about the figure of 6%.

      Demuth's paper had some serious methodological errors, and really should be withdrawn:

      "In conclusion, the rate of gene family extinction in the human lineage (Table 2) appears to be overestimated ... by a factor of 100%."

      http://www.plosone.org/annotation/listThread.action?root=8729

      Delete
  8. lifespy, in this following video, conservatively guessed that 20% of genes in humans are probably ORFan:

    Orphan Genes (And the peer reviewed ‘non-answer’ from Darwinists) – video
    http://www.youtube.com/watch?v=1Zz6vio_LhY

    Branko Kozuli PhD points out that the problem of ORFan genes is getting worse and worse for evolutionists with every new genome sequenced, with no resolution in sight:

    Proteins and Genes, Singletons and Species – Branko Kozuli PhD. Biochemistry
    Excerpt: Horizontal gene transfer is common in prokaryotes but rare in eukaryotes [89-94], so HGT cannot account for (ORFan) singletons in eukaryotic genomes, including the human genome and the genomes of other mammals.,,,
    The trend towards higher numbers of (ORFan) singletons per genome seems to coincide with a higher proportion of the eukaryotic genomes sequenced. In other words, eukaryotes generally contain a larger number of singletons than eubacteria and archaea.,,,
    That hypothesis – that evolution strives to preserve a protein domain once it stumbles upon it contradicts the power law distribution of domains. The distribution graphs clearly show that unique domains are the most abundant of all domain groups [21, 66, 67, 70, 72, 79, 82, 86, 94, 95], contrary to their expected rarity.,,,
    Evolutionary biologists of earlier generations have not anticipated [164, 165] the challenge that (ORFan) singletons pose to contemporary biologists. By discovering millions of unique genes biologists have run into brick walls similar to those hit by physicists with the discovery of quantum phenomena. The predominant viewpoint in biology has become untenable: we are witnessing a scientific revolution of unprecedented proportions.
    http://vixra.org/pdf/1105.0025v1.pdf

    So how many ORFan genes are actually in humans??? Nobody really knows the exact number yet. One thing is certain though, Darwinian presuppositions have been a major hindrance in elucidating the true genetic similarity/dissimilarity between humans and chimps!

    ReplyDelete
    Replies
    1. "So how many ORFan genes are actually in humans???"

      Depends what you call an ORFan gene. I looked at the paper that Cornelius cites as having 60 de novo protein coding genes.

      Now, granted that I only looked at the very first one ("ZNF843"), it was quite easy to find the corresponding DNA on the chimpanzee chromosome, with approximately 98.5% identity.

      So whether it is protein-coding in humans and non-coding in everything else doesn't really concern me. What concerns me is whether it has an evolutionary history: clearly this one does.

      Like I said, I've only done one. I'd happily take bets on the majority of these de novo genes having an evolutionary history (chimpanzee > 95% and/or gorilla > 90%).

      Any takers?

      Delete
  9. dear dr hunter.

    i think i have very strong evidence for design in nature

    a) we know that a self replicate robot that made from dna need a designer

    b) from a material prespective the ape is a self replicate robot

    a+b= the ape need a designer

    or even a self replicat watch.the evolutionist always says that a watch need a designer because it cant self rplicat. so if we will find a self replicat watch we need to say that is made by itself


    plus: if a self replicate car cant evolve into an airplan, how can a bacteria can evolve into human ?

    the evolutionist says that small steps for milions years become a big steps. but according to this a lots of small steps in self replicat car (with dna) will evolve into a airplan.

    but there is no step wise from car to airplan

    evolutionist say that common similarity is evidence for common descent. but according to this 2 similar self replicat car are evolve from each other

    about the egg yolk protein. actually we have a lots of examples for convergent evolution . even on pseudogenes:

    http://www.answersingenesis.or.....3/mistakes

    so there is no evidence for evolution here.

    the vitellogenin actually have another function:

    http://www.sciencedaily.com/re.....092428.htm

    One of these molecules is a protein called vitellogenin. “Simply put, the more vitellogenin in bees, the longer they live. Vitellogenin also guides bees to do different social tasks, such caregiving or foraging


    ReplyDelete
    Replies
    1. dcscccc: ah, so now you've switched to saying that the egg yolk gene data is _not_ evidence for a common ancestor. Smart move! You're following my point #2 above: "Never agree with anything an evolutionist says, no matter how obvious."

      Delete
    2. "i think i have very strong evidence for design in nature

      a) we know that a self replicate robot that made from dna need a designer "

      That's laughably circular. Are you being serious?

      Delete
  10. This is extremely interesting. I remember blogging with someone on telic thoughts a few years back about orphan genes. He proposed that if there were more than 2 or 3 new orphans it would present as a significant problem to the darwinian model. He added that these 2 or 3 should have side-show functionality, not primary functionality.

    While I agree with the math presented here, I also think that a discovery of 2 or 3 unimportant orphans would realistically support the current theory. The orphan gene argument seems very strong to me.

    ReplyDelete
  11. Dcscccc says ” plus: if a self replicate car cant evolve into an airplan, how can a bacteria can evolve into human ?”

    Where did you see a self replicating car? I’d like to study it.

    BTW, bacteria did and don’t evolve into humans.

    Took several, billions of years before single-celled life discovered it could gang up to cooperate and increase their chances of survival. From there on, the road to present diversity was wide open.

    I’d enjoy the debate more if people knew what they are talking about. With a religious faith that is of course not necessary.

    ReplyDelete
  12. shoud've been "did not and do not"...

    ReplyDelete
  13. So single-celled life could "think" and form "co-ops" - that's more than could be said about the life-forms occupying most governments these days!

    But your "theory" fails when you say they needed to "Increase their chances of survival" - as if they weren't already surviving? How much more "survival" did they gain?

    And why did just some of them move on to this new Nirvana and leave so many of their bothers ans sisters behind in their "un-evolved" state? Doesn't sound like a very fair cooperative since their relatives still remain in their single-celled bodies today instead of enjoying all the "survival" benefits of being a full blown human-being like their "evolved" cousins.

    Fundamentally, Evolution Theory is a "religious faith"...

    "This reasoning is not new and it is not science. It is based on personal religious beliefs that are not open to debate. Imagine if you believed these things. Imagine that you believed, with Venema, that common mutations, for example, rules out any possibility of the species having been created in any sort of direct sense. Then of course you would be an evolutionist. Even though evolutionary theory fails every test. With evolution, the religion drives the science."

    ReplyDelete