Wasn’t it enough that I read Nessa Carey’s terrible Junk DNA book? It scarred me, it did. But there’s another one out, John Parrington’s The Deeper Genome: Why there is more to the human genome than meets the eye
I am totally mystified by all these educated people who are completely lacking in basic knowledge of the field they are writing books about. And weirdly, this book got a positive review in Nature!
A scientist and journalist, Parrington covered the ENCODE story for The Times in 2012; his book enriches those accounts with historical and scientific context. The science is better than the history. He provides a fine discussion of recent support for McClintock’s often-overlooked late work on how stress can activate transposition, but he perpetuates the myth that at first no one thought transposition was real. The contested point was actually McClintock’s interpretation of mobile elements as controllers of gene action. Parrington’s strongest chapters survey the emerging view of gene regulation, including DNA folding, epigenetics and regulatory RNA. Overall, this is a faithful, engaging portrait of the twenty-first-century genome.
Jaw dropped. WTF flag fully deployed. Eyes rolled back so far they’re peeping out my ears.
Read Larry’s arguments. If you have trouble following them, though, don’t worry: apparently that means you’re fully qualified to be hired by Oxford University!
But PZ, that’s censorship.
I refuse to read it too. Since I have zero authority, not being a reviewer or the government, my lack of reading isn’t censorship. It is only not bothering with bad science.
My outer D&D nerd read this as The Deeper Gnome.
I have been reading the comments section on Larry Moran’s posts and the argument against Junk DNA seems to be coming entirely from ID types. I would have thought there would be some argument (against Junk DNA) from people with a more systems point of view. Am I wrong (along with being ignorant)? If not, could someone non-ignorant point me to literature on this topic.
G’Deeper Gnome?
http://scienceblogs.com/pharyngula/2015/01/06/the-genetic-load-problem/
Thanks Nerd (hopefully that’s not too familiar). Would you have a link to the Dan Graur paper? The link in PZ’s note no longer works.
Empty, if you google “dan graur junk dna” you get links to this:
http://gbe.oxfordjournals.org/content/early/2013/02/20/gbe.evt028
Using the term “dan graur genetic load” you get links to this:
http://sandwalk.blogspot.com/2015/01/a-lesson-on-genetic-load.html
And that is only first page of each.
@7 empty
The paper linked to in PZ’s note has apparently been removed by its author. Perhaps because Graur recognized the arguments contained therein were specious.
No, ENCODE shill, with specious claims. You haven’t shown with evidence that the genetic load equation is invalid. It requires you and the project to figure out what false positives are first. Have you done that yet?
@10 Nerd
What equation? Be more specific please.
Thanks NoR,DOMT. I should really learn this google thingy. By the way your second link leads back to the removed paper. However, the first was extremely useful. I still haven’t gotten through that so I haven’t tried the google thingy yet. But I will. Thanks again.
Only dumb plays that dumb. You aren’t being rational, rather emotional in your response. You know the equation (under mathematics). It refutes the ENCODE conclusions, and has been around since the ’50s. Pretend ignorance is last resort of the incompetent. Thanks for playing. You acknowledged you have squat.
@13 Nerd
Your link (under equation) leads to a not found error. Want to try again?
It is explicitly from the linked PZ note
and the mean fitness of a population, w, is (1-µ)^n, where n is the number of loci, or genes, in the genome.
It’s derived from Haldane JBS. 1937. The effect of variation on fitness.
Am. Nat.
71:337–49
You can read it in the wikipedia page https://en.wikipedia.org/wiki/Genetic_load in the Mathematics block .
Basically mean fitness is sum(p_i*w_i) where the i_th allele is has the fitness and frequency w_i and p_i respectively. (w_i is equivalent to 1-µ).
This paper had a nice summary on page 116: http://agrawal.eeb.utoronto.ca/files/2013/06/afa_mcw_2012.pdf
This is apparently well-known among biologists. (Of which I am not one.)
Don’t know why Graur’s article was deleted but Nerd’s link to Graur’s “Immortality of Television” article (http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3622293/) seems to be the same idea, albeit a lot longer than 2 1/2 pages.
@15 woozy
Thanks
@10 Nerd
If the genetic load equation that you are referring to is the one described by @15 woozy then it’s validity is not in question. I said that perhaps (a speculation on my part) Graur’s post had been removed because the author recognized the speciousness of his arguments contained therein. I inferred that I found his arguments specious. However, the speciousness of his arguments are related to the assumptions he makes and the conclusions he draws, not in the equation itself.
His assumptions are based on those equations, and so is the conclusion he draws. They are the assumptions and criticisms used by geneticists, developmental biologists, and evolutionary biologists, and many other scientists not buying the ENCODE presuppositions. Which are wrong.
@17 Nerd
You’re right. His conclusions are based on those equations and they are specious conclusions because they’re based on flawed assumptions not backed by data. Grau starts with an equation for fitness into which he plugs values for n and mu. Where does the value of mu (mutation rate) come from? Also, where does he get the number 3 x 10^6 as a value for n (number of genes) ? ENCODE says nothing about n nor mu. Those are Graur’s numbers chosen to fit his presupposition that much of the human genome ‘must be’ junk.
Holy crap, Chris61, you really don’t know this stuff at all. µ is a measured parameter in a great many species: you can actually measure the rate of mutation for specific genes. It is typically on the order of 10-10 nucleotides per replication. The number 3 x 106 is hypothetical for purposes of testing the ENCODE hypothesis. We know the genome is 3 x 109 nucleotides long, and if the average ‘gene’ or functional unit is 103 nucleotides, it can accommodate 3 x 106 genes.
The formula for genetic load tells us that the human genome cannot possibly contain that many genes. It can contain at most 20-30 thousand genes without imposing an impossible genetic load.
Whereas this: from Encode comentaryEncode commentary.And this is a PROXY for function: from Greg Landen:
. The histone binding is PROXY for activity, not proof of activity.
Chris61, please show the paper where the PROXY for binding, and fuctionality, is shown to be actually in vivo activity and function by the normal use of the term, which is that the gene is turned on during the lifetime of the organism.
If you can’t do that, either don’t respond, or acknowledge that the information isn’t available…..
Dang, my #20 is A mess. The definition used by ENCODE is :
This is the histone binding.
It is a proxy for activity, not proof of in-vivo activity, merely an indication of potential genetic activity, and this is a problem ENCODE can’t address.
So, where is the 1:1 correspondence between histone binding and secondary methods showing genetic functionality, that is the gene is expressed during the lifetime of the organism???
@19 PZ
If the measured mutation rate per nucleotide is 10^-10 (on average) how do you get to a mutation rate per gene of 10^-6 – 10^-5 per gene? How do you account for genetic code redundancy and the fact that mutations outside of coding sequence affect gene function? Population genetics deals with idealized genomes and idealized genes. It is a useful mathematical construct for certain purposes but it doesn’t negate the usefulness or conclusions of ENCODE any more than ENCODE challenges evolution.
@21 Nerd
As there is no term for histone binding in the genetic load equation I have no idea what you are talking about.
#22, chris61
Huh, I just did a quick look and see figures of more like 10^-8 per nucleotide, and if there about about 10^3 nucleotide pairs per gene, one might reasonably come up with a mutation rate per gene of around 10^-5. But then, I’m not a geneticist, so what do I know?
Nerd was not talking about the genetic load equation at all. Nerd was talking about ENCODE, as is quite clear from their 6-line post. A cynical person might suspect you’re being deliberately obtuse.
Nerd brought up the genetic load equation. And then brought up histone binding. Graur’s paper said nothing about histone binding, it was based on the concept of genetic load and whether ENCODE’s conclusions were invalidated by that concept. So I genuinely have no idea what Nerd’s talking about as it pertains to Graur’s arguments about ENCODE.