## The 80/20 of “Junk DNA”

In a recent debate with PZ Myers, a prominent old-school Darwinist, my rival asserted that “90 percent of our DNA is junk.”

I insist very little of it is junk. I have no doubt some is broken, corrupted, redundant or whatever – but if the broken stuff were fixed, we would all be healthier.

We know exactly what 3% of it does – it codes for proteins.

In 2012, the ENCODE project team announced that at least 80% of our DNA has some discernible function. Exactly what most of that does remains a mystery.

Meanwhile, some parts of DNA are clearly far more essential than others. One experiment deleted 1% of a mouse genome with no obvious defect.

80/20, the Pareto Principle, offers us a very useful shortcut for predicting how useful it is. 80/20 applies to practically all complex systems. The 80/20 Curve (www.8020curve.com) allows you to calculate, quickly and easily, how things are going to stack up.

If you have 10,000 files on your hard drive and they take up 10 gigabytes, how big is each file? 80/20 tells you with remarkable accuracy. If you have 1,000 customers and your company has a million dollars revenue, how much money do you get from each customer?

80/20 will tell you that too, and when you compare it to your sales reports, you’ll find it’s scary accurate. Ditto with populations of countries.

This is the basis of my book 80/20 Sales & Marketing. The Appendix shows a range of examples, from donations to a church to wealth in the Forbes 400 to dairy production in Wisconsin.

So what does the 80/20 curve predict about DNA? It says 20% of the coding sequences do 80% of the work. To get a more exact answer, you go to “specify the total output of all members” and enter:

Number of members: 100
Total output of all members: 100

What these numbers are saying is simply that 100% of the DNA (100 members) collectively gives us 100% of the DNA’s function; the total adds up to 100%. Now the tool will generate a curve that tells you the relative contribution of different parts.

Enter the parameters and you get a curve like this:

Mouse over the curve and you get to see the contribution of each percentile section.

• The top 1%, or the 99th percentile, contributes 14.2%.
• The 98th percentile contributes 7.8%.
• The 97th percentile contributes 5.5%.

Add those three together, and it says the part of DNA we fully understand, the 3% that codes for proteins, tells 27.6% of the story.

So almost three fourths of the story lies elsewhere, in the “non-coding” regions.

You can use the feature “By rank” on the bottom left of 8020curve.com and it will predict how much function comes from the least important 50% of the genome:

18.2%.

In other words, the least vital half of our DNA contributes 18%. The most vital half of our DNA contributes 82% of function. The vital stuff really is vital. The tiniest error in a Hox gene spells disaster. Cystic Fibrosis comes from one missing codon.

80/20 also predicts that the bottom 1% of our DNA contributes 0.27%. That some parts are nearly negligible.

My experience with 80/20 suggests that all of the bottom 10% will be even less useful than the curve predicts. The bottom ten segments are each probably less than 0.1% of the function. Since most systems can at least work with 80% of their parts, I’m guessing you could lose half your DNA and still stay alive.

Now of course you wouldn’t want that. Any more than you want to have your tonsils taken out or get your appendix removed. (Those were labeled “vestigial organs” for a long time. It turns out they aren’t so useless after all.) You could even survive with one kidney (like my dad did) but it would suck.

We have good reasons to believe that mysterious portions of DNA provide redundancy. For example the genetic code itself is 3:1 redundant. For most amino acids there are three codons that generate the same amino acid instead of one. This turns out to be a highly optimized scheme (optimal to the tune of one in a million) for minimizing copying errors. We know that cells can replace damaged DNA with segments copied from other chromosomes.

We know for example that salamanders often have many duplications in their genomes and have huge amounts of extra DNA. Same is true for onions. So what’s going on here?

As I describe in chapter 16 of Evolution 2.0, hybrids (as in horse + donkey = mule) double the number of chromosomes. Then a process called Hybrid Dysgenesis deals with instability in the genome and removes sequences that are unnecessary. The organism and its descendents begin to use re-arranged segments of the genome for other purposes. Over time the genome gets smaller and functions are added.

A similar thing happens with retroviruses. It now looks very much like mammal placentas were built using large pieces of code from viruses; and viruses are actually a MAJOR source of biodiversity.

Anyone who proclaims your genome is 90% is junk is in the lazy end of the science community. They are lying on hammocks, making egotistical pronouncements about things they barely understand. Junk DNA is junk science.

Mostly the only guys still trying to defend “Junk DNA” are atheists doing damage control. The creationists had been insisting it’s useful since the 1970s. ENCODE put egg on the Darwinists’ faces, and evidence continues to flow in the wrong direction for the Junk DNA crowd.

I have yet to meet anybody who is willing to voluntarily delete 90% of their DNA.

I’m with Eibi Nevo, the Israeli scientist who said, “The future of evolutionary biology lies in better understanding regulation of the non-coding areas that have been wrongly or unjustifiably called junk DNA.”

I estimate we’ll spend the next 100 years figuring out what it all does. Every year the percentage of so-called “junk” will decline until nobody cares to talk about it anymore.

How useful any one piece of DNA is spans a huge range… from a hybrid where half the chromosomes are mostly switched off, to non-coding DNA whose function is mysterious.

Since the definition of “junk” is:

noun
1. any old or discarded material, as metal, paper, or rags.
2. anything that is regarded as worthless, meaningless, or contemptible; trash.

…and since scientists widely acknowledge that quite a bit of “junk DNA” isn’t really junk anyway… how about we recognize that the derisive term “junk DNA” needs to be junked?

Like I said, it may take 100 years to confirm all this… but I’m willing to bet that 100 years from now, good ol’ Pareto won’t be that far off.

Perry Marshall

Where Did Life And The Genetic Code Come From? Can The Answer Build Superior AI? The #1 Mystery In Science Now Has A \$10 Million Prize. Learn More About It, Here – https://www.herox.com/evolution2.0

### 20 Responses

1. Chris Wallis says:

Hi Perry, after listening to your debate with PZ and reading your blog, im left wondering if you have actually bothered to to read any of the many evolutionary and population genetic arguments for junk DNA?
What is your response to the onion test?
Why is only 8% of the code conserved in mammals?
Why does natural selections power to remove slightly deleterious alleles depend on the effective population size?
Also, you mentioned that the “Darwinian” assumption of junk and randomness is unscientific, this is precisely the opposite, only the null hypothesis of “no function” can be used to demonstrate function. This is basic statistics.
If you are genuinely interested in the junk DNA debate, id recommend you read these papers and books:

https://gbe.oxfordjournals.org/content/early/2013/02/20/gbe.evt028.full.pdf+html

Lynch, Michael, and Bruce Walsh. The origins of genome architecture. Vol. 98. Sunderland: Sinauer Associates, 2007.

Regards, Chris.

• Chris,

Thank you for posting. I appreciate you being here.

We know for example that salamanders often have many duplications in their genomes and have huge amounts of extra DNA. Same is true for onions. So what’s going on here?

As I describe in chapter 16 of Evolution 2.0, hybrids double the # of chromosomes and then a process called Hybrid Dysgenesis deals with instability in the genome and removes sequences that are unnecessary. The organism and its descendents begin to use re-arranged segments of the genome for other purposes. Over time the genome gets smaller and functions are added.

A similar thing happens with retroviruses. It now looks very much like mamalian placentas were built using large pieces of code from viruses; and viruses are actually a MAJOR source of biodiversity. Eibi Nevo discusses this at http://budurl.com/g3y4.

So – is this extra DNA junk? Well certainly it is unnecessary from the standpoint of the immediate operation of the organism. But it is also a library of re-organizable code that can and will get used or else discarded in the future. It’s kind of like “whale feet” – I insist the whale is keeping them around for a reason. Organisms retain this code in an orderly way.

Red flags go up all over the place any time I hear someone say things like

“I would be quite proud to have served on the committee that designed the
E. coli genome. There is, however, no way that I would admit to serving
on a committee that designed the human genome. Not even a university
committee could botch something that badly.”
—David Penny

That is a huge over-reach. All these statements I find, upon investigation, are based on the atheistic, “random mutations plus natural selection” model of evolution which information theory shows could not possibly be more wrong. Notice what I said in this article – you could probably get away with deleting 50% of the human genome and still have an immediately functional human. But it would be significantly compromised.

The sanity check for this is:

“OK, so can I delete 90% of YOUR genome?”

And I’m asking you that question. WOULD YOU ALLOW ME TO DELETE 90% OF YOUR DNA?

Most of the people who make statements like the one by David Penny-

1) Do not understand the work of people like Barbara McClintock & James Shapiro
2) Adhere to a neo-Darwinian understanding of evolution, which is demonstrably false
3) Do not build things for a living like engineers do – most of them have never designed a single product and put it into production in their life. They are woefully ignorant of design, even as they sneer at people who see design in biology. Their naive notions of biology and biological evolution would massively fail if they applied them to an actual software program or automobile. This is the reason for the “Salem Hypothesis” which says that engineers and computer programmers tend to be skeptical about old-school Darwinism. That’s because it violates all their professional experience. Their skepticism is correct – evolution does not come from random mutations and natural selection. It comes from a wide array of natural genetic engineering tools that are now very well known to biologists – transposition, horizontal gene transfer, retroviruses etc.
4) They have typical “biologist math background” which in my opinion is woefully inadequate – because when you start talking about math, most biologists tune out. Neo Darwinism is mathematically impossible.
5) Do not understand information theory, which is one of the most useful scientific discoveries of the 20th century
6) Have never even considered building cells or organisms from scratch, let alone DONE it

Bottom line is most junk DNA arguments show tremendous disrespect for nature, hubris and arrogance.

I freely concede that they have SOME points to make. There is SOME validity to the Junk DNA argument.

But the motivations behind it are usually anti-scientific. The term “Junk DNA” should be deleted from our vocabulary. It is entirely misleading and it is an non scientific agenda.

Until you can build an entire organism from scratch, or manually re-construct the entire genome AT WILL (which would have gravely concerning ethical implications) you have no right to claim that it’s junk.

A scientist’s job is to continue pursuing the mystery, not proclaim ‘this stuff is stupid, job done.’

20 years from now, David Penny will be embarrassed that he said that.

2. Chris Wallis says:

Hi Perry, thanks for the reply.

Your response unfortunately misses the point. Much of the massive range in genome sizes (350-fold in vertebrates) is not due to polyploidy and those with massive genomes are stable. The point of the onion test is explain why for example would two related fish that live in similar environments differ in their genome size by orders of magnitude? One of them “decided” to radically expand its genome? This is sheer speculation.
Onions and humans both have diploid (genome not duplicated) genomes and yet onions have a genome 5X larger than our own. What is all that extra DNA doing, and why can other species of onion do just fine without it? What reason does an onion have to radically self engineer that another onion does not? Did it not get the memo? You have no explanation for this at all, only hand waving and pleas of future unknown functions.
The most obvious and scientific answer to this is that the KNOWN process of genome expansion by random transposon proliferation is unconstrained by natural selection because it is too weak in small populations to outweigh genetic drift.

Also you didn’t respond to my question about genome conservation among species. One of the most useful and obvious tests for the function of a sequence is its level of conservation among similar and distantly related species. Important and essential genes tend to be highly conserved because a mutation will tend to destroy the function, surely from a computer and electrical engineering perspective this is obvious. If you have two separate lines of code, A and B, and any change in A causes the system to fail, but any change in B has no effect, what would you conclude?
In most mammals 5-10% of the genome is conserved, this means that 90-95% can change sequence without consequence to fitness (these changes accumulate at the background rate of random mutation). Imagine a novel where you can randomly alter 90% of the words without changing its meaning, if you still thought that 100% were required, then the onus would be on you to show that.
Estimates of the total sequence required for viability of a genome can also be done from the deleterious mutation rate and total genome size, using statistical methods, it has been estimated that humans sustain 2.1–10 deleterious mutations per generation. These data would suggest that at most 10% of the human genome exhibits detectable organism-level function and conversely that at least 90% of the genome consists of junk DNA. These figures agree with measurements of genome conservation (~10%, see above) and are incompatible with the view that 80% of the genome is functional in the sense implied by ENCODE.

The rest of your reply doesn’t seem at all relevant. This has nothing to do with “old school Darwinism” in actual fact, strict Darwinians appose neutral theory, they believe that natural selection would remove junk for the genome. Your other cheap slurs against biologists not knowing math (absurd if you read anything about population genetics) not being engineers and not understanding information theory are typically vacuous. None of those things, even if true, have any relevance to this debate.
Again, I recommend you read the links I gave you, also, this one is a good review and easy to understand for a non-scientist such as yourself. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4014423/pdf/pgen.1004351.pdf

Regards.

• Chris,

Thank you for your great questions, and for keeping it civil.

The root of these authors’ approach to this whole issue is found in this paragraph:

To understand the current state of the human genome, we need to examine how it evolved, and as Michael Lynch once wrote, ‘‘Nothing in evolution makes sense except in the light of population genetics’’ [55]. Unfortunately, concepts that have been generated by this field have not been widely recognized in other domains of the life sciences. In particular, what is underappreciated by many nonevolution specialists is that much of molecular evolution in eukaryotes is primarily the result of genetic drift, or the fixation of neutral mutations. This view has been widely appreciated by molecular evolutionary biologists for the past 35 years.

The paper is based on the above premise. The premise is wrong.

Population genetics is not responsible for major evolutionary events. At best it gives minor gradual events. Symbiogenesis and Hybridization are responsible for large / rapid evolutionary leaps. Population genetics does not take into account transposition or epigenetic changes that are triggered by signals from the environment.

This belief is also blended with an assumption that random events play a large part in evolution itself. Information theory shows this to be absolutely false. Evolutionary events have to conform to the syntax and semantics of the genetic language.

Chris you have to keep in mind that the core of what most biologists are taught about basic Darwinian evolution – the random mutation theory – is absolutely wrong. A very large number of assumptions in evolutionary biology hinge on a false model. These guys are all traditional Darwinians. That long paragraph above is the smoking gun.

Biologists’ ignorance of information theory is not trivial. It’s holding back science in a major way.

In most mammals 5-10% of the genome is conserved, this means that 90-95% can change sequence without consequence to fitness (these changes accumulate at the background rate of random mutation).

The reason 5%-10% of the genome is tightly conserved is that – see my 80/20 article above – 10% of the genome determines 50% of function and in almost all situations, that is the extent of operation that is “life-and-death” to the organism.

I know this very well from another “Darwinian” environment, Google AdWords. 2% of the advertisers get 50% of the traffic. It’s ferociously competitive. I wrote the world’s best selling book on internet advertising, the Ultimate Guide to Google AdWords, and have been involved in hundreds of millions of dollars of A/B testing of ads. Some parts of a Google ad are mission critical and you optimize them to the nth degree. Down to the last comma.

The rest of it – as long as you have something reasonable, any number of “acceptable” combinations of words or phrases or graphical elements will do. Nothing you do with those variables makes an appreciable difference, as long as you don’t screw something up.

So in a winning campaign, in which you might test thousands of ads, some components are highly conserved and others are not. THIS DOES NOT MAKE THE NON-CONSERVED ELEMENTS USELESS. It only means that while elements A B and C are highly conserved, the other 50% could be interchangeably given elements D, E, F, G, H, I and J and any of them would work OK. You also find out that K, L, M and N in those same positions would be lethal.

Then when you say:

“Imagine a novel where you can randomly alter 90% of the words without changing its meaning, if you still thought that 100% were required, then the onus would be on you to show that.”

The 90% you refer to DO change the meaning. But all those meanings in the population are merely acceptable. Most men do not choose their wife because of the beauty of her elbows – most any woman’s elbow would do. But that doesn’t mean that women’s elbows are meaningless with respect to attracting men. It doesn’t mean her elbows are junk.

I disagree with the statement: “the characteristics of the sequences constituting 98% or so of the human genome that is nonprotein-coding are generally well understood.”

The reason I disagree is that they say so many other things that are mathematically false. This group of authors, which includes Larry Moran (he is thanked near the end of the paper) have this idea that mobile genetic elements are things that just randomly and parasitically multiply through the genome. Larry Moran’s description of transposons is sophomoric and that’s putting it kindly.

First of all I have never seen any code anywhere, where you could just randomly and parasitically insert bits of code into other code willy nilly. (Jerry Coyne uses the exact phrase “willy nilly” to describe mutations in his book Why Evolution is True.)

Have you?

I base my assumptions on how all OTHER codes operate. True randomness is ALWAYS lethal so I feel very safe in predicting that many behaviors of transposons currently assumed to be random actually occurs for a reason.

Hypothesis A: There’s a reason why transposons do what they do, go where they go.
Hypothesis B: there’s not a reason why transposons do what they do, go where they go.

Hypothesis A is a scientific hypothesis.
Hypothesis B at best is a null hypothesis and at worse is an anti-scientific self fulfilling prophecy.

There is a flippant disrespect of living things that I find in all the “Junk DNA” arguments.

And nobody (including you) seems interested in answering the question “May I delete 90% of your DNA?”

May I, Chris?

There is a collection of statements that I’d like to call into question:

endogenous retroviruses, and cut-and-paste DNA transposons. Because of their capacity to increase in copy number, transposable elements have long been described as ‘‘parasitic’’ or ‘‘selfish’’ [22,23]. However, the vast majority of these elements are inactive in humans, due to a very large fraction being highly degraded by mutation.

Despite this, there is currently no evidence that the majority of highly repetitive elements are functional.

Once again, although some pseudogenes have been co-opted for organism-level function (for example see [37]), most are simply evolving without selective constraints on their sequences and likely have no function [38].

The majority of human DNA consists of repetitive, mutationally degraded sequences.

I do not think the tests in which these assumptions were made were anywhere near as exhaustive as they would need to be in order to justify the conclusions. I’ve owned a couple of software companies and software testing has to be exhaustive. In fact a good program for testing software is typically 3X more complicated than the software under test.

Are our tests of organisms 3X more complicated or exhaustive than the organisms themselves?

Ummm… no.

Most of the time we’re content if the new fish are just swimming around in the aquarium. If you have not subjected the organisms to all manner of shocks and environmental conditions and observed the effects over many generations, then you don’t know.

I just got a tetanus shot yesterday. I’m sure that changes RNAs or VDJs some strand of something which is stored somewhere in my immune system. My tetanus immunity is totally useless “junk” … EXCEPT when my body is fighting tetanus. Then suddenly it’s not junk.

It makes me cringe when people say “We can’t really see what this does, so it must be useless.”

So when they say:

Another large fraction of the genome consists of highly repetitive DNA. These regions are extremely variable even amongst individuals of the same population (hence their use as ‘‘DNA finger- prints’’) and can expand or contract through processes such as unequal crossing over or replication slippage.

These findings are consistent with the idea that eukaryotic genomes are filled with junk DNA that is transcribed at a low level.

They appear to be dismissing this and even contradicting themselves, since “repetitive” and “extremely variable” sound like opposites to me. Maybe I’m misunderstanding something.

The presence of these mutations is usually not harmful, because diploid organisms generally require only one functional copy of any given gene.

Why is this true? Because of the built-in redundancy and error correction in the genome. Redundancy is not to be confused with Junk. Totally different things.

I agree that 10% of the genome (which according to 80/20 is 50% of its function) is absolutely mission critical and completely determined by selection.

if most eukaryotic DNA is functional at the organism level, be it for gene regulation, protection against mutations, maintenance of chromosome structure, or any other such role, then why does an onion require five times more of it than a human?

I don’t know.

http://scienceoveracuppa.com/2013/08/18/continuing-the-debate-on-function-in-junk-dna-rethinking-the-onion-test/

I will personally make an educated guess that it’s analogous to files on my computer.

The “.docx” files on my computer right now range from 20.9 GB to 4KB.

The longest one with the most pages is 400 pages and it’s 10GB. The 20GB file has 13 pages. There is little correlation between # of pages and file size. Why? Because they have images and stuff like that. The images are in many cases not mission critical. In a fitness test some images which take up lots of space could be entirely optional. But they are not junk.

The most important part of most WORD documents is the text, and images are secondary. But the images take more space.

The core functions of an organism are stored in code that is INCREDIBLY efficient by human standards. Is it ridiculous to suggest that there are some functions in nature that are very inefficient in their use of coding space, but are still useful?

Now let me be clear that I am only speculating on the onion question, drawing analogies from familiar situations. But what I object to is the hubris of assuming something is junk when we are still decades or maybe centuries away from being able to build living things from scratch ourselves.

Craig Venter talked about using an existing Ribosome as being like stealing engine parts out of a Ferrari.

Also, can we once again both acknowledge that organisms have layers of redundant systems so they can deal with shocks and unanticipated situations? Redundancy is not junk.

3. Chris Wallis says:

Hi Perry, you wrote:

“Population genetics is not responsible for major evolutionary events. At best it gives minor gradual events. Symbiogenesis and Hybridization are responsible for large / rapid evolutionary leaps. Population genetics does not take into account transposition or epigenetic changes that are triggered by signals from the environment.”

This has to be one of the strangest, and most baseless assertions I’ve ever heard about population genetics, you really couldn’t be more wrong. It’s almost as if you are confusing it with some other field. Population genetics is ABSOLUTELY responsible for ALL evolutionary events, by definition any change, random or otherwise must spread and become fixed in a population if it is to continue, this isn’t just obvious, it’s mathematically proven. The parameters that dictate how any change can spread to fixation are the effective population size, selection co-efficient and mutation rate. This applies to transposition, duplication, insertion and other form of change you can dream up (epigenetics concerns regulation, not inheritance, you, like many are miss-using the phrase.) Experiments have repeatedly shown that the power of natural selection to remove slightly deleterious alleles is proportional to the inverse of the population size. Bacterial genomes, like the one Craig Venter synthetically synthesized are small and contain basically zero junk, this is because of their huge population sizes which allows selection to remove deleterious alleles. Humans and (some) onions have very low effective population sizes (historically 10,000 for humans) and are therefore unable to effectively remove slightly deleterious alleles.

“This belief is also blended with an assumption that random events play a large part in evolution itself. Information theory shows this to be absolutely false. Evolutionary events have to conform to the syntax and semantics of the genetic language.”

Sadly, this again is completely wrong, there is overwhelming experimental evidence that shows that the occurrence of mutations is independent of their effects, this is what is meant by random in a biological context. A famous experiment done by Luria and Delbrück showed with rigorous mathematics that mutations that allow E.coli to resist phage were independent of being exposed to phage. There are two explanations for this experiment (and thousands of others) some bacteria “engineered themselves” to be resistant to phage even though they were never exposed (no signal), while others did not for some odd reason. The other is that there is a background rate of mutation caused by copying errors that is independent of phenotype, so just by chance, some bacteria received mutations that enabled them to resist phage, and others did not. The second option is not only more obvious and sensible, it is simpler, more predictive and therefore more scientific. It also fits perfectly with similar experiments on many, many other organisms. Mutations are not directed, there is no bias towards advantageous mutations, most are neutral, some are deleterious and occasionally some are useful. It seems, that because of your lack of biological and scientific training, you have fallen in the typically naïve trap of assuming that the word random is the same in all contexts. In short, these are not aesthetic biases, biologists don’t want mutations to be random, we just have 100yrs of evidence that suggests that by and large, they are.

http://www.genetics.org/content/28/6/491.full.pdf+html

http://www.sciencedirect.com/science/article/pii/S0025556499000450

“Chris you have to keep in mind that the core of what most biologists are taught about basic Darwinian evolution – the random mutation theory – is absolutely wrong. A very large number of assumptions in evolutionary biology hinge on a false model. These guys are all traditional Darwinians. That long paragraph above is the smoking gun.”

I don’t think I need to point out the hypocrisy and sheer hubris of a non-biologist telling trained biologists what is wrong with their theories, especially when you’ve gotten so much basic stuff completely wrong. Also, I mentioned previously that “traditional Darwinians” strongly oppose neutral and nearly neutral theories of molecular evolution, they worship at the altar of natural selection, the people you quote above, including Lynch absolutely do not.

“Biologists’ ignorance of information theory is not trivial. It’s holding back science in a major way.”

Again, you are completely wrong. Biologist’s knowledge of information theory reflects the usefulness of information theory as proposed by Shannon and Kolmogorov to solving biological problems. There was initially huge interest and many cross-collaborations between biologists and information theorists such as John von Neumann around the time Shannon published A Mathematical Theory of Communication and geneticists were cracking the genetic code. Unfortunately these theories were shown to be basically useless apart from providing biologists with some useful analogies and languages. One reason is that both these informational theories say nothing about meaning, something obviously important in biology. I think that actually it is creationists that don’t understand information theory, they invent their own bogus versions of it that actual mathematicians and information theorists tear apart (or laugh at) and then accuse biologists of ignorance, meanwhile mathematicians have shown that information theory poses NO problems to evolution. Randomness and selection, mutation, recombination and reproduction can all increase information in the proper ways. Your assertion that information theory shows evolution to be impossible is nothing but that, an empty and groundless assertion.

I won’t spend too much time on the screed about advertising models, they show only a superficial resemblance to the models used to describe actual evolving populations. The grammatical rules of the genetic code are completely different to human languages and are a poor proxy. I will quickly address this:

“The 90% you refer to DO change the meaning. But all those meanings in the population are merely acceptable. Most men do not choose their wife because of the beauty of her elbows – most any woman’s elbow would do. But that doesn’t mean that women’s elbows are meaningless with respect to attracting men. It doesn’t mean her elbows are junk.”

This analogy is again false, conserved functional genetic elements can only withstand a small degree of mutational variation, much like variation in elbows. But you couldn’t randomly replace or change elbows with any old thing, there are still constraints to building a functioning elbow. However, there are NO sequence restraints on non-functioning alleles, they can be mutated and changed to almost anything (just think of the billions of combinations) with ZERO effect on any measurement of function or fitness.

“So in a winning campaign, in which you might test thousands of ads, some components are highly conserved and others are not. THIS DOES NOT MAKE THE NON-CONSERVED ELEMENTS USELESS. It only means that while elements A B and C are highly conserved, the other 50% could be interchangeably given elements D, E, F, G, H, I and J and any of them would work OK. You also find out that K, L, M and N in those same positions would be lethal.”

Again, you are still putting tight constraints on the “functional” elements in your flawed analogy. If 50% could be changed to ANYTHING, not just some other elements with known functions (D, E, F, G, H, I, etc) but instead with absolutely any combination, and it had no measurable effect on function, you would be reasonable to conclude they are functionless.

I also wanted to quickly address this hilarious statement regarding transposons.

“Hypothesis A: There’s a reason why transposons do what they do, go where they go.
Hypothesis B: there’s not a reason why transposons do what they do, go where they go.
Hypothesis A is a scientific hypothesis.
Hypothesis B at best is a null hypothesis and at worse is an anti-scientific self fulfilling prophecy.
There is a flippant disrespect of living things that I find in all the “Junk DNA” arguments.”

People that actually do scientific experiments and use statistics to test hypotheses know that there are no isolated hypotheses, all hypotheses have at least two parts, H0 and H1. H0 is the null hypothesis and it is essential, it makes testing possible. H0 suggests what one should find if there is no relation between the variables being tested. H1, H2 etc suggest that there is a relationship between variables being tested.
For example, H0: drug X has no effect on B. H1: drug X increases B. You then compare the results, if you cannot show that H0 is false (to high significance) you CANNOT say H1 is true.

Your simplistic caricature amounts to “transposons do something” which cannot be tested and thus cannot be scientific. Any experimental result is consistent with this, it is impossible to prove wrong. “Transposons do nothing” however, IS a scientific hypothesis because it can be tested.
I find it astonishing that this basic form of essential scientific reasoning has to be pointed out to you.

Finally, I’ll quickly address this statement you continue to repeat ad nauseam.

“And nobody (including you) seems interested in answering the question “May I delete 90% of your DNA?”
May I, Chris?”

This is simply too childish to take seriously, we are saying that currently, our best estimate of the proportion of functional elements in the human genome is ~10%. This is no dogma, it could turn out to be 12, 15 or 20% but not staking my own life on 10% is not evidence that the number is wrong. This threat is simple desperation.

There are more points that id like to refute but I feel this reply is long enough for now.

Regards.

• I will respond to your subsequent post as I have time.

• Jim Serafin says:

Antibiotic Resistance Is Prevalent in an Isolated Cave Microbiome –

“These studies support a hypothesis that resistance is an ancient and genetically rich natural phenomenon, deeply embedded in the microbial pangenome.”

http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0034953

• Jim Serafin says:

Actually the question and your answer points out the uncertainty of your own position. Otherwise you would test it.

A guy claims gravity has no effect at the top of The Empire State Building. You tell him prove it. Jump. He says that is childish. What is one to think?

4. RE: “…as Michael Lynch once wrote, ‘‘Nothing in evolution makes sense except in the light of population genetics’’ [55].

See The bioenergetic costs of a gene http://www.pnas.org/content/early/2015/10/29/1514974112.abstract

Lynch and Marinov have decided to remove the energenic cost of de novo gene creation and replace it with pseudoscientific nonsense.

Excerpt: “The general logic underlying this treatment is that the selective consequences of modifying a particular genic feature (e.g., number and size of introns, expression level, amino acid use, gene-copy number, etc.) is a function of the degree to which the overall energy budget is altered. Based on its phenotypic manifestations, a gene may have a multiplicity of advantages, but the energetic cost of replication, maintenance, and expression represents a minimum burden that must be overcome to achieve a net selective advantage. If a genic variant or a novel gene is to be efficiently promoted by natural selection, the net selective advantage (beyond the energetic cost) must exceed the power of drift (defined as 1=Ne for a haploid organism, where Ne is the effective population size)” (2, 8).

5. Blake Reas says:

Perry,

As I was listening to the debate one thing puzzled me. PZ asserted numerous times that you cannot use concepts of engineering in Biology. He didn’t give a good rationale. It seems to me that human beings are naturally able to detect design. This is where I think the Intelligent Design crowd has it correct. We use our ability to detect specified complexity all the time; we use it in archaeology, the ability to detect codes from the surrounding noise, etc. So, it would seem that someone like PZ would have to give an equally good rationale for denying information in biology. Second, P.Z. seems to assume that God would have to directly intervene and “tinker” in order for their to be information in biology, but that doesn’t seem necessarily the case to me. God could, conceivably, create a world with all the potentialities packed in, and as it unfurled those potentialities would become actualities (I am intentionally using Aristotelian concepts here). In the same way the acorn has all the potentiality of an oak, so the early universe (or singularity) would have all the potentiality of the present universe.

Am I getting PZ wrong here? I just couldn’t detect any rationale at all.

• You got it exactly right! Information is a major problem for Darwinists and the only easy way out is to just avoid it entirely.

6. Chris says:

Hi Perry, happy to wait for a thoughtful response.

I would like to quickly address the last comment. Information as defined by shannon and others is no problem for evolution at all, variation, replication, recombination, duplication and selection all increase information in the proper way. This is why creationists have invented an entirely unrecognised circular form of information that actual information theorist simply laugh at. Specified complex information is not accepted by mathematicians or information theorists, its just a question begging joke.

As for archeology, design is detected by using specific information of actual designers, their limitations and capacities to deduce design. Knowledge of the designer is first required for this process. Since the aim of creationists is to show the existence of a designer, they must do this first and then they can show that biology is designed.

• Chris,

I can get to this first. Will respond to your other salvo later.

It’s not clear from what you just said that you understand the severe problems that information theory poses for standard Neo-Darwinism.

Selection does not increase information. It always decreases information.

“Variation” ONLY increases information if the new information obeys the laws of the code. Darwinism claims the variations are random. Random variations increase information entropy and destroy information. This is one of the main points of Shannon’s work. Randomness is noise and noise destroys. ALWAYS. There are no exceptions to this. Well, I take that back, there are exceptions in the sense that there is an infinitesimally small chance that all the heat in the room you’re in right now will spontaneously move to one corner and you will freeze to death 10 seconds from now. But that is not going to happen. And neither are random mutations going to increase information in the genome.

The reason that positive evolutionary genetic changes happen (and I agree and affirm that they do; and I provide MANY examples of such evolutionary events in my book Evolution 2.0) is because of cell-directed genetic edits like horizontal gene transfer, transposition, symbiogenesis etc. Those systems do increase information.

DNA is a code.
All the other codes we know the origin of are designed.
Therefore we have 100% inference that DNA is designed and 0% inference to any other explanation.

You are free to reject that hypothesis, but you can’t claim your rejection of design in biology is scientific until you can produce a code without designing one.

My private equity investment group is offering \$3 million for a naturally occurring code. Specification is at http://www.naturalcode.org.

7. Chris Wallis says:

Hi Perry, you said

“Selection does not increase information. It always decreases information.”

Wrong, selection reduces noise, which according to Shannon increases information.

““Variation” ONLY increases information if the new information obeys the laws of the code. Darwinism claims the variations are random. Random variations increase information entropy and destroy information. This is one of the main points of Shannon’s work. Randomness is noise and noise destroys. ALWAYS. There are no exceptions to this. Well, I take that back, there are exceptions in the sense that there is an infinitesimally small chance that all the heat in the room you’re in right now will spontaneously move to one corner and you will freeze to death 10 seconds from now. But that is not going to happen. And neither are random mutations going to increase information in the genome.”

Its seems you have failed to read any relevant work by information theorists that work on these problems, I suggest Elsberry and Shallit’s paper to bring you up to speed.

Under both Shannon and Kolmogorov, information can be increased by random variation, duplication, deletion, recombination and selection. I realise this is counter-intuitive, but I’m afraid these are just facts. You should be more open minded.

Please show me one relevant academic paper that shows this to be false.

“The reason that positive evolutionary genetic changes happen (and I agree and affirm that they do; and I provide MANY examples of such evolutionary events in my book Evolution 2.0) is because of cell-directed genetic edits like horizontal gene transfer, transposition, symbiogenesis etc. Those systems do increase information.”

This is another example of ignorant speculation presented as fact. I already showed you that adaptive processes have been shown to occur via random mutations and selection, see Luria and Delbrück. I do agree that the processes mention above increase information, but they are not directed by the cell, teleology has little use in science, it makes more assumptions and cannot be tested.

“DNA is a code.
All the other codes we know the origin of are designed.
Therefore we have 100% inference that DNA is designed and 0% inference to any other explanation.”

Silly syllogisms also have no place in science, and that isn’t how inferences are made. I can easily dream up my own.

DNA is a code
All the codes we know the origin of come from organisms with material brains
Therefore we have 100% inference that DNA comes from a material mind and 0% inference to any other explanation.
Wow, I just proved that we invented DNA and that biology came before its own code! Philosophy is fun!
You are free to reject that hypothesis, but you can’t claim your rejection of materialism in biology is scientific until you can produce or find a non-material brain.

Maybe I’ll start a million dollar competition to find a non-material mind.

Regards.

• The above statement is a gross misinterpretation of Shannon’s theory – in fact turned it inside out like a sock.

Yes, randomness increases information – but here’s what you’re saying.

You record CD #1 onto a cassette tape. Then you play the cassette and record it back on to CD #2.

Yes, absolutely it does. It has tape hiss. It may have 2-3X more information than the original.

Is the signal better?

Nope.

Warren Weaver, in his preface to the U of Illinois printed book edition of Shannon’s paper, calls out this exact fallacy and refers people who use it, “jokers.” Flagrant abuse of the concept of information entropy.

“Please show me one relevant academic paper that shows this to be false.”

http://smile.amazon.com/Mathematical-Theory-Communication-Claude-Shannon/dp/0252725484/ref=sr_1_1?s=books&ie=UTF8&qid=1452558214&sr=1-1&keywords=mathematical+theory+of+communication

“information can be increased by random variation, duplication, deletion, recombination and selection”

Random variation destroys information. Selection destroys information. Duplication doesn’t increase information except when epigenetics or gene expression changes the transcription of the duplicated part. Yes, deletions can increase functional information. I can change “the flight is not taking off at 6:05” to “the flight is taking off at 6:05” and reverse the meaning by deleting the word “not.”

Recombination can increase information – recombination is not random. I cite a reference in the side annotations at http://www.evo2.org/pz-myers.

Dellbruck Luria experiment did not prove evolutionary mutations are random. See

http://shapiro.bsd.uchicago.edu/Shapiro.2013.Rethinking_the_%28Im%29Possible_in_Evolution.html

where the author says:

4. Virus infection cannot induce DNA changes giving heritable resistance. This opinion has long cited the famous 1943 Luria-Delbrück experiment (Luria and Delbrück 1943) as evidence that viral infection cannot induce heritable resistance determinants in the infected cells. However, it turns out that such a sweeping conclusion is not valid. Prokaryotes have the CRISPR acquired immunity defense system, and animals can generate so-called “piRNA” molecules after viral or mobile element infection. Both prokaryotes and eukaryotes have the capacity to incorporate fragments of genome sequence from invading nucleic acids and use these fragments to generate small RNA molecules that block invader reproduction. As recently as the end of the 20th Century, such adaptive genomic defense strategies would have been dismissed as inconceivable. Yet their existence and functions have been incontrovertibly documented in this century.

Syllogism: You said

DNA is a code
All the codes we know the origin of come from organisms with material brains
Therefore we have 100% inference that DNA comes from a material mind and 0% inference to any other explanation.

What you have stated is very nearly equivalent to the biological principle of biogenesis: life only comes from life.

8. Chris Wallis says:

Hi Perry, thanks again for the reply.

“The above statement is a gross misinterpretation of Shannon’s theory – in fact turned it inside out like a sock.
Yes, randomness increases information – but here’s what you’re saying.
You record CD #1 onto a cassette tape. Then you play the cassette and record it back on to CD #2.
Yes, absolutely it does. It has tape hiss. It may have 2-3X more information than the original.
Is the signal better?
Nope.”

Yet another question-begging straw-man. If you cannot notice a logical fallacy that obvious, then I can’t help you I’m afraid. Maybe an example, linked in the paper below, an experiment you say is impossible. Scientists created a large library, about the size of a large bacterial population, containing random strings of 80 amino acids. None of the sequences were designed or preordained for any function. Functional strings that bound ATP were then isolated after 8 generations of selection and amplification. Over 6% of the population showed ATP binding functionality. So which part of information theory precludes this?
http://www.ncbi.nlm.nih.gov/pubmed/11287961

“Random variation destroys information. Selection destroys information. Duplication doesn’t increase information except when epigenetics or gene expression changes the transcription of the duplicated part. Yes, deletions can increase functional information. I can change “the flight is not taking off at 6:05” to “the flight is taking off at 6:05” and reverse the meaning by deleting the word “not.””

Variation and selection both increase information. Epigenetics turns off genes, it has nothing to do with neo-functionalisation. Information is increased when the gene is duplicated and further increased by subsequent rounds of mutation and selection, most of the human gene families can be explained by this.

http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0000085

“Recombination can increase information – recombination is not random. I cite a reference in the side annotations at http://www.evo2.org/pz-myers.”

Tell that to people with down-syndrome and hundreds of others types of recombination caused genetic diseases.

“Dellbruck Luria experiment did not prove evolutionary mutations are random.
4. Virus infection cannot induce DNA changes giving heritable resistance. This opinion has long cited the famous 1943 Luria-Delbrück experiment (Luria and Delbrück 1943) as evidence that viral infection cannot induce heritable resistance determinants in the infected cells. However, it turns out that such a sweeping conclusion is not valid. Prokaryotes have the CRISPR acquired immunity defense system, and animals can generate so-called “piRNA” molecules after viral or mobile element infection. Both prokaryotes and eukaryotes have the capacity to incorporate fragments of genome sequence from invading nucleic acids and use these fragments to generate small RNA molecules that block invader reproduction. As recently as the end of the 20th Century, such adaptive genomic defense strategies would have been dismissed as inconceivable. Yet their existence and functions have been incontrovertibly documented in this century.”

Wrong again, no one says that ALL adaptions precede in this fashion, only that many do and that variation plus selection can be adaptive, something you say is impossible. You would do well to actually read the paper. The paper shows that mutations (CRISPR is not a mutation, it’s a form of recombination) are random with respect to outcome. We know exactly which mutations cause the phage resistance and they are demonstrably not directed by the cell. We can use random mutagens to achieve the exact same outcome. So I ask again, what is the explanation for this? Does the cell “know and direct” these mutations to certain genes because they “know” that later on scientists will infect them with Phage? Of course not, but this is exactly what you are proposing is necessary for adaptive evolution.
Also, contrary to Shapiro, far from being inconceivable, we have known since the 50s and 60s that lambda phage integration into the bacterial genome causes heritable resistance. Does the cell retain the integration because it “knows” it will prevent further infection, even though later on all its descendants will be destroyed because of it? Cells are so clever!
So, your statement that random mutation plus selection cannot be evolutionary is just plain false, your reference to CRISPR is also a red herring.

“Syllogism: You said
DNA is a code
All the codes we know the origin of come from organisms with material brains
Therefore we have 100% inference that DNA comes from a material mind and 0% inference to any other explanation.
What you have stated is very nearly equivalent to the biological principle of biogenesis: life only comes from life.”

Comically wrong, I think you have again missed the point of my reductio ad absurdum. By aping your own silly syllogism I showed that DNA must have come from a brain, but that is impossible because DNA is required to make brains. Just as intelligence comes from a material brain, it is therefore impossible that brains come from intelligence, its like saying liquid comes from swimming pools, but of course liquid is required before swimming pools even makes sense.

Im looking forward to a proper response to my earlier comments.

Regards.

• The Delbruck Luria experiment does not support your hypothesis. This link explains why:
http://www.huffingtonpost.com/james-a-shapiro/epigenetics-ii-cellular-m_b_1668820.html
It says:

“The empirical evidence against experience-related genome change is remarkably thin. Textbooks present the famous 1943 Luria-Delbrück experiment as the definitive demonstration that mutations must occur prior to selection.

Salvador Luria and Max Delbrück, who received Nobel Prizes for their pioneering roles in molecular genetics, set out to test whether virus infection could induce mutations to resistance.

The analysis was statistical in nature. (Their paper even includes an elaborate mathematical appendix by Delbrück, a converted theoretical physicist.) The argument was as follows. If infection induced mutations to virus resistance with a certain low probability, then both different samples from one culture and samples from independent cultures should display similar numbers of colonies that would fit some form of normal (Gaussian) distribution.

They set up a large number of independent cultures of bacteria, divided them each into multiple samples, infected all samples with viruses that killed infected cells, and measured the frequency (proportion) of resistant mutant bacteria in each sample that could produce a colony in the presence of the lethal viruses.

What they found was that the replicas of single cultures did produce normal distributions of resistant colony numbers, as expected, but the colony numbers from one culture bore no statistical relationship to the other independent cultures. Some cultures produced very low numbers of colonies, while others displayed “jackpots” of high frequencies of resistant cells.

They correctly interpreted this deviation from a normal culture-to-culture distribution of mutant frequencies to provide evidence for the stochastic occurrence of rare mutations to virus resistance at different times in the growth of each culture prior to selection. Early mutations would multiply into many resistant progeny (high mutant frequency), while later-occurring mutations would produce smaller populations of resistant progeny (low mutant frequency).

Given the lethal nature of the selecting virus Luria and Delbrück used, there was in fact no other possible outcome. Infection was invariably lethal, and only preexisting resistant mutants could survive. Nonetheless, this experiment was cited for over six decades as proof that virus infection could not induce a genetic change to resistance.

The “impossible” had happened. By a still-unknown mechanism, the bacterial cell managed to insert a short sequence from the infecting virus in its genome’s CRISPR region and use that newly acquired information to block the infection. There is now evidence that the CRISPR defense works by encoding a small interfering RNA molecule (siRNA) and using that siRNA to guide cleavage of the invading DNA.”

Barbara McClintock’s contribution was showing that variations are NOT random with respect to outcome. Quoting her Nobel Prize paper:

“The responses of genomes to unanticipated challenges are not so precisely programmed. Nevertheless, these are sensed, and the genome responds in a discernible but initially unforeseen manner.”

Population genetics:
1. the branch of genetics concerned with the hereditary makeup of populations.
2. the study of changes in gene frequencies in population of organisms and the effects of such changes on evolution and adaptation.

Well I suppose just about anything could fall under the population genetics umbrella, but in my experience the term normally does not refer to hybridizations between distant species, symbiogenesis events or epigenetics. If you feel that it does then population genetics is too broad a topic to be useful in this conversation.

Transposition hypotheses: H0: It makes no difference where transposons land. H1: Where transposons land correlates strongly to biological function, if the experiments are thorough. I’m with H1.

Any situation where living things are competing for resources and the consequences are success and failure, life and death, is natural selection. Darwin himself got his ideas about natural selection from Malthus who got his ideas from studying business. If you don’t think business is Darwinian… you’ve never owned one.
I’ve done thousands of natural selection experiments, from the level of micro advertising to success and failure of multi-million dollar companies. The fact is, 80% of what you find in any business is not mission critical to its success. 20% is.

Note that this is the whole point of this blog post above.

Your theory of conservation and argument for Junk DNA essentially says:

“I found Espresso Machine X in every single Starbucks, therefore Espresso Machine X is highly conserved and necessary. But I found chairs by brand A, B, C, D, G, H and J which are obviously not conserved, so the chairs at Starbucks are all junk.”

Why do you believe this? Because you believe that selection is the only directional component in evolution. This is demonstrably not true. It’s certainly not true in Starbucks and it’s well known that many variations are largely indifferent to selective pressures.

So no, you don’t get to dismiss insights from every day, real world, practical natural selection that puts literal food on peoples’ tables, in favor of an ivory tower theory that is for the most part untested.

specially when you’re not willing to be the test subject by deleting your own Junk DNA!

I said: “And nobody (including you) seems interested in answering the question “May I delete 90% of your DNA?”  May I, Chris?”

You said: This is simply too childish to take seriously…This threat is simple desperation.

You want me to take your Junk DNA hypothesis seriously? But you don’t want to delete any of your Junk DNA? That’s funny. From this point forward I will no longer take your Junk DNA assertions seriously. Another case of “don’t listen to what people say, watch what they do.” Ostensibly you don’t trust your own theory.

Nothing you have said indicates that you understand information theory. If you think you understand information theory, then begin by explaining to me, in detail, how natural selection increases information.

I said:
“The above statement is a gross misinterpretation of Shannon’s theory – in fact turned it inside out like a sock. Yes, randomness increases information – but here’s what you’re saying. You record CD #1 onto a cassette tape. Then you play the cassette and record it back on to CD #2. Does CD #2 have more information than CD #1? Yes, absolutely it does. It has tape hiss. It may have 2-3X more information than the original. Is the signal better? Nope.”

Then you simply evaded the question. Chris, please explain in exact detail how your theory of random mutation increasing information because it creates greater Kolmogorov complexity is any different than what I just described above.

You said:
Scientists created a large library, about the size of a large bacterial population, containing random strings of 80 amino acids. None of the sequences were designed or preordained for any function. Functional strings that bound ATP were then isolated after 8 generations of selection and amplification. Over 6% of the population showed ATP binding functionality. So which part of information theory precludes this? http://www.ncbi.nlm.nih.gov/pubmed/11287961

1) They excluded stop codons and other patterns, so it’s not fully random. Based on fully random mutations, one out of 20 nucleotides would be a stop codon. So it’s easy why they would want to get rid of that inconvenient problem.

2) The paper says: We therefore estimate that roughly 1 in 10^11 of all random-sequence proteins have ATP-binding activity comparable to the proteins isolated in this study…. this frequency is still low enough to emphasize the magnitude of the problem faced by those attempting de novo protein design.
In other words: Randomness cannot be relied upon to produce a new protein.

One chance in 1 in 10^11 is very small. We only have 10^11 humans in all history on earth. Randomness is too inefficient to produce evolutionary events. Based on your math we would only get one new PART of a protein 1 time with all the humans on earth. This paper does not support your thesis, Chris. It supports mine.

9. Biosemiosis.org says:

Chris Wallis: “DNA is a code. All the codes we know the origin of come from organisms with material brains. Therefore we have 100% inference that DNA comes from a material mind and 0% inference to any other explanation.”

I agree with Perry, you’ve merely restated the principle of biogenesis, and properly so.

So the first step in our search for answers tells us that life only comes from life. That is our universal experience. You then suggest we have nothing else to work with, but that is simply not true. Genetic translation is a semiotic system that uses spatially-oriented representations (Pattee) and a reading frame code (Crick). And as it turns out, those things require very specific material conditions in order to function. Those conditions are fully documented in the literature, and are not even controversial. They also happen to create a physical system that is exclusively identifiable among all other physical systems. The singular material conditions that enable the genetic code can only be found in one other instance anywhere in the cosmos, and that is in recorded language and mathematics — two unambiguous correlates of intelligence. So from a purely physical perspective, the unique conditions required by such a system, which would not appear on earth until the rise of human intelligence, are entirely evident in the organization of the first heterogeneous cell.

So our first step into ‘universal experience’ tells us that life only comes from life — vie de la vie — and our second step tells us that an unambiguous correlate of intelligence is empirically detected in the organization of every living cell on earth. And by virtue of the cell’s self-replicating nature, we now know that the minimum requirements for the origin of the system are established by what is physically necessary to record and translate the amount of information that the system needs to successfully describe itself into memory.

From this, we also know that in order to begin the cycle of life, the details of the system must be simultaneously encoded in the very information that it makes possible. Otherwise, Iife on earth would not exist.

What is completely absent, however, is even a hint of a non-intelligent origin of life.

http://www.Biosemiosis.org

10. Chris Wallis says:

Hi Perry, I hadn’t checked the site in a while, but thanks for the reply.

You said:
“Given the lethal nature of the selecting virus Luria and Delbrück used, there was in fact no other possible outcome. Infection was invariably lethal, and only preexisting resistant mutants could survive. Nonetheless, this experiment was cited for over six decades as proof that virus infection could not induce a genetic change to resistance.”

Oh dear, you are really struggling with this simple concept, and quote mining James Shapiro and the huffington post? You obviously again failed to read the paper, and instead have relied on someone else who has. The cultures were taken from independent SINGLE colonies that were NOT resistant to phage, new mutations arose during the growth phase prior to selection that changed non-resistant cells into resistant cells. It’s really pretty obvious, what would be the point of choosing a single cell already resistant to phage to measure spontaneous phage resistance? I think your lack of background is making this a waste of my time. I asked some simple questions that you still cannot answer, how did these single colonies “know” before they were exposed to phage to pick the right mutation? What happened to the millions that didnt choose the right mutation? Was their magic teleology machine broken? Why does increasing the random mutation rate with UV or other mutagens increase the proportion of cells resistance to phage? Or do the bacteria tell the UV radiation to make specific non-random changes?

Also, again, thousands of similar experiments on lethal and non-lethal selections have shown the same thing, and this paper is just an example. It is not proof of anything other than random mutations can cause resistance to phage. It is also now known the exact mutations required to produce phage resistance, they occur in transport proteins and it has absolutely nothing to do with CRISPR, it is a complete red herring.

“Population genetics:
1. the branch of genetics concerned with the hereditary makeup of populations.
2. the study of changes in gene frequencies in population of organisms and the effects of such changes on evolution and adaptation.
Well I suppose just about anything could fall under the population genetics umbrella, but in my experience the term normally does not refer to hybridizations between distant species, symbiogenesis events or epigenetics. If you feel that it does then population genetics is too broad a topic to be useful in this conversation.”

I already explained why population genetics is indispensable for studying ANY kind of evolutionary change, including transposition, hybridisations and any other process you flout but clearly misunderstand. In your experience? Ha, you have none, and that is precisely the point.

“Transposition hypotheses: H0: It makes no difference where transposons land. H1: Where transposons land correlates strongly to biological function, if the experiments are thorough. I’m with H1.”

Pitifully wrong again, H0 is the null hypothesis, you can’t just choose H1, you use H0 to test H1! That is the point of a null, its essential!

“Your theory of conservation and argument for Junk DNA essentially says:
“I found Espresso Machine X in every single Starbucks, therefore Espresso Machine X is highly conserved and necessary. But I found chairs by brand A, B, C, D, G, H and J which are obviously not conserved, so the chairs at Starbucks are all junk.””

You really have a talent for saying “essentially your argument is……” and then completely miss-representing it, I wonder if this is intentional or just random? You should avoid business metaphors and stick to the scientific literature. You also used this straw man in a previous reply, im happy to refute it again.

Sequence conservation is an obvious proxy for function because it tells you which parts can and cannot be changed and to what degree, it isn’t an on or off phenomenon, it’s variable. In your example, chairs are conserved, they CANNOT be randomly changed to bowling balls, uranium, dirt or Coca-Cola. They have to function as chairs, changing brand doesn’t change the function, just another red herring im afraid. In the genome however, if you can randomly change the sequence of some DNA to ANYTHING without consequence, then that is a good argument it has no function.

“Why do you believe this? Because you believe that selection is the only directional component in evolution. This is demonstrably not true. It’s certainly not true in Starbucks and it’s well known that many variations are largely indifferent to selective pressures.”

Its really very hard for me to understand how you could write something that is just so obviously wrong and contrary to my entire argument. I said that natural selection CANNOT remove neutral or slightly deleterious alleles, they accumulate as junk because of genetic DRIFT. The reason junk exists is because selection cannot remove it! Selection is absolutely not the only force that shapes genomes.

“So no, you don’t get to dismiss insights from every day, real world, practical natural selection that puts literal food on peoples’ tables, in favor of an ivory tower theory that is for the most part untested.”

This is just special pleading, you have walked into an academic debate you have no experience with, if you cannot stomach academic discourse and instead prefer simplistic pop-psychology, then I suggest you stick to advertising. I won’t even bother to address the typically weak and unsubstantiated slur that evolution is untested.

“You want me to take your Junk DNA hypothesis seriously? But you don’t want to delete any of your Junk DNA? That’s funny. From this point forward I will no longer take your Junk DNA assertions seriously. Another case of “don’t listen to what people say, watch what they do.” Ostensibly you don’t trust your own theory.”

I already showed how absurd and childish this is, my unwillingness to perform experiments on my self has absolutely nothing to do with the validity of any of the arguments I’ve put forward. You’re best argument is a non-sequitur.

“Nothing you have said indicates that you understand information theory. If you think you understand information theory, then begin by explaining to me, in detail, how natural selection increases information…….
Then you simply evaded the question. Chris, please explain in exact detail how your theory of random mutation increasing information because it creates greater Kolmogorov complexity is any different than what I just described above.”

I evaded nothing, and nothing you have said indicates you have read or understood the paper by Elsberry and Shallit that I linked. The paper shows:

“under accepted interpretations of the term “information”, such as Kolmogorov complexity, it is easy for algorithms (including genetic algorithms) to generate outputs having more information than that contained in the input. For example, it can be shown that the Kolmogorov complexity of xx exceeds that of x for infinitely many strings x. In other words, a simple program that duplicates strings (i.e., maps x to xx) may increase Kolmogorov information.”

“simple randomized algorithms (those with access to a source of genuinely random bits) may generate outputs with arbitrarily more Kolmogorov information than the total information contained in the algorithm and input combined. For example, consider an algorithm PAL, defined as follows: on input n, it generates a randomly chosen palindrome of length 2n by starting with the empty string and then randomly applying either f0(x) = x0x or f1(x) = x1x n times. The resulting string will nearly always have Kolmogorov complexity close to n, but the Kolmogorov complexity of the input and program combined is bounded by (log2 n) + c for some fixed constant c.”

“Thus information may well increase in genetic algorithms, in one standard sense of the word.”

If you want the gritty details on the mathematics, again I suggest you read the paper.

“They excluded stop codons and other patterns, so it’s not fully random. Based on fully random mutations, one out of 20 nucleotides would be a stop codon. So it’s easy why they would want to get rid of that inconvenient problem.”

Completely irrelevant, you have again completely miss-interpreted the paper or even the reason I cited it. I asked which part of information theory precludes a large pool of random sequences being selected for a novel function? You say random variation plus selection cannot produce new functions, the paper shows the opposite. Now you’re saying its not impossible but unreliable? I can cite you a hundred papers were this exact method has been used to find new functions for proteins. We use broken polymerase enzymes to introduce random mutations into existing proteins which are then selected for new activities routinely in my lab, how hard is this to understand? Stop codons are simply eliminated by selection.

“The paper says: We therefore estimate that roughly 1 in 10^11 of all random-sequence proteins have ATP-binding activity comparable to the proteins isolated in this study…. this frequency is still low enough to emphasize the magnitude of the problem faced by those attempting de novo protein design.
In other words: Randomness cannot be relied upon to produce a new protein.”

No, in other words, you didn’t understand the paper, randomness is used in directed evolution experiments to develop new proteins every day, do a google scholar search, there are thousands of examples.

“One chance in 1 in 10^11 is very small. We only have 10^11 humans in all history on earth. Randomness is too inefficient to produce evolutionary events. Based on your math we would only get one new PART of a protein 1 time with all the humans on earth. This paper does not support your thesis, Chris. It supports mine.”

Logic isn’t really a strong point of yours is it? Your own reasoning would suggest that an average bacterial population of 10^11 that divides every 30 min would produce 48 new parts every single day! Hardly very small, I think you should really have a proper think about what your thesis actually entails.

• I will respond to this as I have time. Thanks for your patience.