More Recent Comments

Wednesday, August 03, 2016

More junk science in Science

The latest issue of the journal Science (Aug. 1, 2016) has an article on a recent paper by Aires et al. (2016) published in Developmental Cell. Here's the abstract of the paper ...

Vertebrates exhibit a remarkably broad variation in trunk and tail lengths. However, the evolutionary and developmental origins of this diversity remain largely unknown. Posterior Hox genes were proposed to be major players in trunk length diversification in vertebrates, but functional studies have so far failed to support this view. Here we identify the pluripotency factor Oct4 as a key regulator of trunk length in vertebrate embryos. Maintaining high Oct4 levels in axial progenitors throughout development was sufficient to extend trunk length in mouse embryos. Oct4 also shifted posterior Hox gene-expression boundaries in the extended trunks, thus providing a link between activation of these genes and the transition to tail development. Furthermore, we show that the exceptionally long trunks of snakes are likely to result from heterochronic changes in Oct4 activity during body axis extension, which may have derived from differential genomic rearrangements at the Oct4 locus during vertebrate evolution.
... those ignorant of history are not condemned to repeat it; they are merely destined to be confused.

Stephen Jay Gould
Ontogeny and Phylogeny (1977)
The results were written up by a freelance journalist named Diana Crow [‘Junk DNA’ tells mice—and snakes—how to grow a backbone]. She writes ...
‘Junk DNA’ tells mice—and snakes—how to grow a backbone

Why does a snake have 25 or more rows of ribs, whereas a mouse has only 13? The answer, according to a new study, may lie in "junk DNA," large chunks of an animal’s genome that were once thought to be useless. The findings could help explain how dramatic changes in body shape have occurred over evolutionary history.

Scientists began discovering junk DNA sequences in the 1960s. These stretches of the genome—also known as noncoding DNA—contain the same genetic alphabet found in genes, but they don’t code for the proteins that make us who we are. As a result, many researchers long believed this mysterious genetic material was simply DNA debris accumulated over the course of evolution. But over the past couple decades, geneticists have discovered that this so-called junk is anything but. It has important functions, such as switching genes on and off and setting the timing for changes in gene activity.
Sandwalk readers will see all the mistakes and misconceptions in these paragraphs. She's talking about regulatory sequences that were never, ever, thought to be junk. The paper being discussed has nothing to do with junk DNA and the results do not in any way alter our understanding of developmental gene regulation.

If you look carefully at the abstract, you'll see the word "heterochronic." This is one of Stephen Jay Gould's favorite words. He wrote about it in Ontogeny and Phylogeny.
I wish to emphasize one other distinction. Evolution occurs when ontogeny is altered in one of two ways: when new characters are introduced at any stage of development with varying effects upon subsequent stages, or when characters already present undergo changes in developmental timing. Together, these two processes exhaust the formal concept of phyletic change.; the second process is heterochrony. [my emphasis ... LAM] If change in developmental timing is important in evolution, then this second process must be very common.
This was written in 1977—that's almost 40 years ago! These ideas were around for decades before Gould wrote his book1 and they have been shown to be correct by numerous studies in the 1980s.

What's going on here? Science is supposed to be one of the leading science journals. How could it publish an article that misrepresents the field so badly? Do the editors send these "Latest News" articles out for review?

1. Ed Lewis shared the Nobel Prize in 1995 for his contribution to "the genetic control of early embryonic development" [The Nobel Prize in Physiology or Medicine 1995].


Jass said...

What's going on here? Science is supposed to be one of the leading science journals. How could it publish an article that misrepresents the field so badly? Do the editors send these "Latest News" articles out for review?

Larry, you have to either wake up or grow up, or both. I know your beliefs interfere with this progressive thinking but the "filed of science" you refer to represents the leftovers of that either neo-Darwinist or progressive Darwinists that couldn't move on. There are very few left.

The world of science is moving one. You are being ignored and left behind. Coyne realized it just in the right time...

Unknown said...

Velhovsky, your comments betray your ignorance. Dr. Moran's critique is valid, regardless of neo-Darwinism's accuracy. The editors of Science do not appear to be doing their jobs. If the DNA region in question is demonstrably functional, then it is not junk, regardless of what might have been previously thought. Regulatory elements were never considered junk, so why does the article claim they were?


PZ Myers said...

Hox genes specify segment identity, not the number of segments, just to throw in one more complaint about that abstract.

I don't get the Science blurb at all. Are they trying to argue that Oct4 is "junk DNA"?

W. Benson said...

Larry, Stephen Jay Gould's affirmation in Ontology and Phylogeny is incorrect. Biological evolution occurs when gene frequencies change, not "when characters are introduced at any stage of development," whatever that means. Everyone knows that. Heterochrony, by the way, is a term originally coined and defined pretty much in its modern sense by German naturalist (and atheist: religious folk hated him) Ernst Haeckel in, I think, his 1874 book "Anthropogenie".

Donald Forsdyke said...


There is much more to critique about Science than the stray musings of the occasional freelance journalist. Last week Science proudly announced "Immunology Journal Debuts."

The launching of a new journal, Science Immunology, by a major publisher is a major event (1). A newly appointed editor’s choice of first article for an opening issue can expect wide attention. First impressions count. Science Immunology’s eye-catching title – “The Discontinuity Theory of Immunity” – a collaboration between philosopher and basic scientist, should serve this purpose (2).

However, the article appears to repeat, in more concise form, the authors’ earlier article in Nature Reviews Immunology (3). Here it was claimed that “the discontinuity theory gathers under a simple explanation a range of phenomena.” While it was conceded that the theory “echoes the concept of” the earlier work of Burnet and others on “immune surveillance,” the reader was given the impression – as in the Science Immunology article – that this “theoretical framework” was somehow novel.

Yet, writers on immune surveillance have long taken as obvious the intermittency of immune challenges. For example, a paper in a leading medical journal in 1968 (4) pondered:

"In what ways do self-determinants differ from foreign determinants? Self- determinants can vary over a wide range of concentrations (compare, for example, serum-albumin with the serum levels of some protein hormones). In theory, foreign determinants can also vary over an equally wide range of concentrations. The only definitive statement which can be made about foreign determinants is that they are unlikely to be constantly present; for much of the time their concentration is likely to be zero [i.e. they are discontinuous]. Self-determinants are likely to be constantly present at some concentration higher than zero [i.e. they are continuous]."

The first article in Science Immunology contains a valuable literature update, but the “simple explanation” seems to take simplicity too far. A naïve philosopher who wrote that the universe can be explained in terms of a balance between the forces of good and evil should not expect his work to mark the debut of a new philosophy journal.

(1) Colmone AC, Sallusto F, Abbas AK (2016) Promoting immunology: The future is here. Science Immunology 1: aag2713.

(2) Pradeu T, Vivier E (2016) The discontinuity theory of immunity. Science Immunology 1: aag0479.

(3) Pradeu T, Jaeger S, Vivier E (2013) The speed of change: towards a discontinuity theory of immunity? Nature Reviews Immunology 13: 764-769.

(4) Forsdyke DR (1968) The liquid scintillation counter as an analogy for the distinction between "self" and "not-self" in immunological systems. Lancet 291: 281-283.

Larry Moran said...

We mustn't forget that these journals have to make money, or at least not lose money. We often think of them as motivated by the same goals and ideals as typical scientists but that's just not true.

The problem comes when the editors of the journals get confused about this. They often claim that their motives are pure and all they're trying to do is advance science and science education. Their behavior belies that claim. They seem to care very little about truth and accuracy.

It's the hypocrisy that bothers me more than anything.

Science Immunology was not created in order to advance the field of immunology. It was created to siphon grant money and subscription fees into the AAAS bank account.

Larry Moran said...

You should probably read the book before making any more comments.

Brian said...

I can't get over the collective amnesia/ignorance of the last 30 years of gene regulation research. I was in graduate school getting a PhD in molecular biology during the heady days of promoter and enhancer bashing in the early 90's. We saw intergenic regions as a goldmine of potential cis-acting regulatory sequences that were bound by potentially novel transcription factors. We certainly didn't see it as junk a priori. We would joke with each other about what we would name any novel factors we found. We used phylogenetic footprinting (my lab produced some of the earliest multi-pairwise alignments of large sequence blocks at the time) to find conserved blocks of sequence(we were working on globin gene clusters then), and test them for function. Everyone was doing this all over the world, and countless intergenic regulatory systems were characterized at thousands of gene loci. This was just before targeted chromatin modification was being identified as a control mechanism (which was in ~1995). It's like it was all a dream.

Donald Forsdyke said...


I think Larry is a little harsh. The new editors have respectable scientific backgrounds and are, I believe, truly seeking to advance science.

My comment above, questioning their judgement, was submitted as an eLetter to Science (28 July) in response to "Immunology Journal Debuts." Readers can judge for themselves whether it is impolite or extreme. However, a week after submission, the eLetter has not gained acceptance for publication in Science.

Quite often these days when political problems arise, we find that an initial offence is less serious than the subsequent cover-up. It seems that, in not encouraging forthright comments on this debut article, the editors of science are seeking to protect the new editors of Science Immunology.

Larry Moran said...

For me it was the 1980s when my lab did promoter bashing. We identified a strong regulatory region (noncoding DNA!) that we could paste onto the beta-galactosidase gene from E. coli. When we inserted this gene into the mouse genome we could turn them blue!

Kothary, R., Clapoff, S., Darling, S., Perry, M. D., Moran, L. A., and Rossant, J. (1989). Inducible expression of an hsp68-lacZ hybrid gene in transgenic mice. Development, 105(4), 707-714. [PDF]

Jon Binkley said...

I guess ignorance about the history of regulatory regions is sort of understandable for science journalists -- they probably don't teach lac operon and lambda repressor to undergrads anymore -- but it's dismaying to see it in working scientists. Does no one search literature older than 5 years any more?

Donald Forsdyke said...

Regarding my "Cover up?" note (Aug 4), the "editorial staff of Science" informed me (28th July") that "Your eLetter, if accepted, should be viewable within a few days." Well, today (August 8th, incidentally being William Bateson's birthday), the eLetter has become viewable. The staff may have a liberal interpretation of "a few days." Alternatively, my complaints on blogs such as this may have generated second thoughts on acceptability. The delay is regrettable, since many readers have now moved on to the next issue of Science. But better late than never.

Unknown said...

"Ontogeny" not "ontology" guys.

Larry Moran said...

Oops! Thanks. I fixed the post.

Have you read the book?

Unknown said...

No, but I think I will.