Preserved T. rex or bad science?

Science has published several articles in the past year or so, with much publicity, announcing that researchers had found preserved soft tissue from Tyrannosaurus rex that contained fragments of the original collagen proteins from T. rex. The researchers leading the research, J.M Asara and M.H. Schweitzer, claimed (Science 2007 Apr 13;316(5822):280-5) that the T. rex protein fragments were more closely related to birds than to reptiles. Very plausible, of course - and if correct, yet another bit of evidence linking modern birds to dinosaurs.

However, even though birds almost certainly are descended from dinosaurs, this series of articles is a textbook case of how even the best journals - Science, in this case - can publish bad science. The mass spectrometry in the Asara et al. articles was poorly done, and even they had to admit - in a letter published in Science last fall - that at least one of their protein fragments was falsely identified. Despite this admission - and they only had 7 fragments to begin with - Science published a second article this past winter by the same authors. What did that article (Organ et al., Science 2008 Apr 25;320(5875):499) contain? Nothing new about the T. rex material, but instead some additional analysis of crocodile and ostrich proteins and a repeat of the claim that T. rex is closer to birds. The title of the article, though, is "Molecular phylogenetics of mastodon and Tyrannosaurus rex" - astonishing, if you expect Science to publish only the best science. But maybe not so astonishing if seen in the light of the publicity that Science is always hungry for. Still, it's too bad.

But why am I so skeptical? Well, don't take it from me. In a comprehensive examination of the T. rex material and methods, Buckley et al. reported - in Science, no less (2008 Jan 4;319(5859):33) that the "ancient" DNA reported by Asara and colleagues failed all the standard tests for such material. In other words, it's a contaminent.

Now, the most recent news on this story, published in PLoS ONE (
Kaye TG, Gaugler G, Sawlowicz Z 2008 Dinosaurian Soft Tissues Interpreted as Bacterial Biofilms. PLoS ONE 3(7): e2808 doi:10.1371/journal.pone.0002808), shows that the most likely source for this DNA is a bacterial biofilm. Not really surprising, but it's great to see this correction of the record. Unfortunately, the Science articles got so much play in the press that they will likely be hard to "erase" from the public's memory. I suspect that far more scientists will have heard of the "T. rex tasted like chicken" stories (yes, that's how some of the press stories were headlined) than the new PLoS ONE article.

For a more detailed discussion of the biofilm story, see Tara Smith's recent blog post on this same topic.

So my conclusion about the original (2007) article in Science: those 7 protein fragments weren't from T. rex at all. They were bacterial contaminents. Asara, Schweitzer, and colleagues were simply wrong - perhaps wishful thinking clouded their vision. The article never should have passed peer review, and Science should make an effort to correct the record. Sometimes even the best journals screw up, and that's what happened here.

3 comments:

  1. I work in a Mass Spec lab (Bafna/Pevzner at UCSD) that has looked extensively into the claims of this paper, including a re-examination of the original spectra used to make the claim. Your summation that this should never have been published is correct. They spectral annotations published in the science paper are highly questionable and several of them have been redacted. Moreover, when I was at ABRF, the authors presented this work. A member of the audience I talked questioned them heavily about their methods for protein preparation before mass spec, basically saying that you can't get results with the published methods on the protein collagen.

    ReplyDelete
  2. I don't see why the fact that the tissue looks bacterial is surprising. We've known for years that dinosaurs were just big prokaryotes thanks to Mark Boguski's famous analysis of the "Jurassic Park" sequence in the early 1990s.

    http://www.inf.fu-berlin.de/lehre/WS05/aldabi/aufgabe5_12.html

    ReplyDelete
  3. Sam, I hope you and your colleagues publish a technical rebuttal of the mass spec evidence. It would be very valuable to have additional refutations of this work appear in the literature. Schweitzer is standing behind it - she's quoted in Carl Zimmer's news piece in Science this week, and one of her chief defenses is basically this: if the material is bacteria, then how come the proteins resemble bird proteins?

    So you guys should get your work out there and add to the debate. I look forward to seeing your contribution.

    ReplyDelete

Markup Key:
- <b>bold</b> = bold
- <i>italic</i> = italic
- <a href="http://www.fieldofscience.com/">FoS</a> = FoS

Note: Only a member of this blog may post a comment.