Robert Sheldon, ID proponent, defending the arsenic bacteria paper? Oh dear God.

You’ve got to be kidding me – Robert Sheldon, Uncommon Descent blogger and rocket physicist, has come out with a post on The Procrustean, his personal conservative political blog, defending the arsenic-metabolising GFAJ-1 paper. It truly is a perfect storm of ID proponent meets terrible scientific methodology.

The scientific backlash against the recent NASA-funded paper, which claims that a strain of Halomonadaceae bacteria called GFAJ-1 can metabolise arsenic as a substitute for phosphorus, has been swift and punishing, as anyone involved with the science blogging community would probably know.

How can Robert dismiss the paper’s criticisms so easily? Well, it boils down to a misrepresentation of the criticisms being made and a disregard for the importance of rigorous scientific technique.

The alacrity of the refutations to NASA’s breathless press release have been surpassed only by their vitriol. “Should not have been published“, “scathing attack“, “big idea with big holes“, “arsenic cowards” etc.  But to my googling eyes, there really are two, and only two refutations given to the Science paper,

a) The technique was sloppy, and the arsenic might be just a contaminant, not a constituent of the cell, making the phosphorus levels low, but still consistent with phosphorus-starved normal life.

b) Arsenic bonds are 100x less stable than Phosphorus bonds, so the claims for As replacing P in DNA are theoretically impossible.

He never addresses a) properly, a major critique of the paper – if the arsenic detected was simply contamination, the conclusion of the paper goes out the window without a second thought. Robert ignores the technical side of this criticism, as laid out by Rosie Redfield, and instead claims that it’s all okay, because the main author of the paper, Felisa Wolfe-Simon, was a post-doc and shouldn’t be expected to have good technique:

…the authors of the paper knew their results were a bit sloppy. After all, the first author is a post-doc who graduated last year! You can’t exactly expect her to have 20 years experience in laboratory technique under her belt.

That’s not a valid excuse! If the results are sloppy, you shouldn’t get published in Science! As Rosie Redfield said: “If this data was presented by a PhD student at their committee meeting, I’d send them back to the bench to do more cleanup and controls.” Journals don’t have to accept papers just because their conclusions are exciting, they also care about the quality of the data behind the conclusion, and if the data are not up to scratch then you don’t get the privilege of being published.

He then, strangely enough, attacks Rosie, insinuating she had a vested interest in critiquing the paper:

So what precisely is Rosie Redfield’s beef? Perhaps that she is also a woman doing good science with 5 times the experience of Ms Felisa, but none of the press. Just by criticizing Felisa, Rosie got 30,000 hits on her website. So perhaps we can forgive Rosie for being so negative. All of us gotta do what we gotta do to survive in this fame-crazed ratrace for grant monies.

I highly doubt she wrote a technical critique of a paper on her blog so that she could somehow score a grant.

What about Robert’s opinion of b), “Arsenic bonds are 100x less stable than Phosphorus bonds, so the claims for As replacing P in DNA are theoretically impossible”? Well firstly, before I quote him, you must know that this isn’t a prominent problem that scientists have had with the paper. Alex Bradley, the blogger Robert quotes as a source of this criticism, never mades this claim – in fact, he made a different one that was far more persuasive and pertained directly to the experimental technique used by Felisa and her colleagues:

If this DNA did not hydrolyze in water during the long extraction process, then it doesn’t have an arsenate backbone. It has a phosphate backbone. It is normal DNA.

In essence, any arsenate-based nucleotides in the genome of GFAJ-1 would have hydrolyzed under the conditions of the experimental techniques used, and this would have manifested in the electrophoresis images as many fragments of DNA – but fragments were not observed, leading Alex to conclude that the genome of GFAJ-1 is composed exclusively of phosphate-based nucleotides.

Robert didn’t see it that way, for some reason, and misrepresents what Alex was claiming:

Well what about Alex Bradley’s claim that arsenic-laced DNA is unstable?

It’s a theoretical claim, based on some other chemistry work he’s done. And like most chemistry claims, it captures about 10% of what the cellular biochemistry is doing. You really, really cannot use testtube chemistry to understand what the cell does, mostly because the cell is chock-full of nanomachinery that operate against entropy, against chemical gradients, against energy gradients. Cell walls, pores, pumping stations, directed transport, enzymes, etc, all conspire to stabilize things that should be unstable, and to destabilize things that should be stable. As far as I can tell, Alex knows that the bulk arsenates are 100x less stable than phosphates, and predicts that arsenic DNA will come unglued.  Maybe. Or maybe there’s a stabilizing protein that keeps those nasty water molecules away that would otherwise hydrolyze the bond. We don’t know until we look, and we won’t look until a paper like this comes along that claims an impossible result.

So Alex, how about showing some humility and saying something like “we would be greatly surprised if the cell had a way of stabilizing As-DNA, which theoretically should be unstable in water”?  My sense is that the Science paper was more humble than Alex, which contradicts his implicit claim of hypocrisy.

But the DNA had been experimentally extracted from the cellular biochemistry, including any hypothesised proteins that could stabilise the arsenate-carbon bonds! Isolated in an aqueous solution, the arsenate nucleotides, had they existed, would have hydrolyzed in approximately 10 minutes, something that wasn’t observed by the researchers. In aqueous solutions, the instability of arsenate-carbon compounds relative to related phosphate-carbon compounds is well known, this isn’t an ad hoc claim made by Alex or any of the other critics of the paper.

Robert’s post wouldn’t be complete without some tangential remarks about another area of science he finds questionable, of course:

And speaking of hypocrisy, why is it that Jonathan Eisen thinks that blogs are a great forum for scientific debate, and says that if one uses a press-conference to advance a Science paper, then one should engage the blogosphere as well? Perhaps because he is writing a blog instead of a paper submitted to Science? (I now view most unjust critiques as projection, and thus a post-modern form of confession.)

Ah, you say, but Science doesn’t have to publish his letter to the editor, so it is possible for Science to bias the discussion. Indeed it is. Which is why “peer-reviewed publishing” might not be the “gold-standard” of science that Anthropogenic Global Warmists have been claiming.

I’m at a loss. I’m just glad that nobody takes ID proponents seriousl- Oh. Wait. They do.

The take-home lesson from this post? If an ID proponent comes out in support of a paper, it would probably be best to take a far more critical look at it.

Update: Rosie Redfield has commented on Robert’s post. In case it gets taken down, here it is in full:

Hi sweetie!

You’re right, badmouthing other scientists is very lucrative, and it’s nice to see you trying to follow in my footsteps. I don’t think you’re quite up to my standard yet, but if you’d like to come over to my blog maybe I can give you some suggestions.

Who knows, we might even team up and get some juicy NIH grants by impugning the motives of the researchers they fund!

I have no words.

When life gives you questionable data, make sensational claims, non-thorough NASA!

Remember the amazing arsenic-metabolising Halomonadaceae, GFAJ-1? Of course you do, it was everywhere in the news only a few days ago. Well, it turns out the basic conclusion of the Science paper, that the bacteria were using arsenic metabolically instead of phosphorus, is highly questionable based on the data Felisa Wolfe-Simon et al. collected. Why do I say that? Rosie Redfield, a microbiologist in British Columbia, has written a critical review of the paper from a microbiology perspective, revealing a stunning lack of evidence that arsenic was even present in the macromolecules of the Halomonadaceae cells. I recommend you check it out.

Why am I writing a completely new blog post about this? In science, corrections and falsifications are as important, if not more important, than confirmations, and mainstream science journalism often fails to mention in any great detail or extravagance when a previously-hyped claim or discovery turns out later to be nothing special or based on dodgy data. I’m breaking that trend in my own ineffective way! Oh yes, things will change now… er, yeah.

Update: Alex Bradley, over at We Beasties, has another critique of the paper, from a chemistry perspective. If GFAJ-1 really did have arsenate-containing nucleotides in its genome, why didn’t they hydrolyse before the electrophoresis was complete?