More Recent Comments

Tuesday, October 02, 2007

Mythical PNAS Papers


 
Here's part of a Harvard University Press release issued yesterday.
Beyond a 'speed limit' on mutations, species risk extinction

Genomes of various organisms lose stability with more than 6 mutations per generation

CAMBRIDGE, Mass. -- Harvard University scientists have identified a virtual "speed limit" on the rate of molecular evolution in organisms, and the magic number appears to be 6 mutations per genome per generation -- a level beyond which species run the strong risk of extinction as their genomes lose stability.

By modeling the stability of proteins required for an organism's survival, Eugene Shakhnovich and his colleagues have discovered this essential thermodynamic limit on a species's rate of evolution. Their discovery, published this week in the Proceedings of the National Academy of Sciences, draws a crucial connection between the physical properties of genetic material and the survival fitness of an entire organism.
This sounds very interesting. The limit of six mutations per genome per generation is far less than the calculated mutation rates for mammalian genomes [Mutation Rates] so it looks like another genetic load argument in favor of junk DNA.

So, I set off to retrieve the article that, according to the press release was published in this week's issue of PNAS. But it wasn't. You can see for yourself by looking at the current issue on the website [Sept. 25, 2007].

Not a problem. I've encountered this discrepancy before. What they mean is the issue that's about to be published and the article is available online in prepublication format. All you have to do is check the "Early Edition" (in this case the Oct. 2, 2007 edition) by clicking on the link from the PNAS home page. Except that the paper isn't there either.

Thus, in spite of what it says in the press release, this paper has not been published by PNAS in either the paper issue or online. This is not the first time this has happened. Over the past few months I've tried to find half a dozen mythical PNAS papers that are prominently mentioned in press releases.

Wait a minute ... look at the fine print on the early edition page [Early Edition]. The version that I'm looking at right now says "Last updated October 2, 2007." Right below that is the following statement.
Because PNAS publishes daily online, you may read about an article in the news media on Monday or Tuesday, but the article may not publish online until later in the week. You may use the CiteTrack feature to set up an e-mail alert to notify you as soon as the article you are interested in publishes.
This is unacceptable. If PNAS can't guarantee that a paper will be available when the press release embargo is lifted then they should change the embargo date. Most other journals have a restriction on press releases that delays the promotion of a paper until it is published and we can see for ourselves whether the hype and the reality match. Apparently PNAS is aware of this problem but instead of fixing it by moving the embargo date to Friday they choose to ignore publishing etiquette. This is wrong.

10 comments :

Anonymous said...

I've also been noticing this problem lately. A few weeks back I was trying to look up a new paper on the PNAS website about taste sensors in the gut (which had been mentioned in a news release as being available on the PNAS site), yet it was nowhere to be found. PNAS sends me an email every week day with links to new Early Edition papers, and I had to wait about 3 or 4 days before the paper was finally available on the site.

Anonymous said...

Yeah, this seems to be a real problem with PNAS. I check their early edition pretty much every day and papers are often behind their press release dates. Sometimes they get it right (a BBC article on Sabretooth Tigers based on a PNAS article, appeared on time today), however I must have waited a week at least for this one:

http://www.pnas.org/cgi/content/abstract/0703913104v1

Anonymous said...

While on the subject of PNAS, is anybody up on their current reviewing policy? It's my understanding that an Academy member can publish pretty much anything in there without peer review, but articles submitted by the hoi-polloi do get reviewed conventionally. Or is it any article sponsored or recommended by an Academy member gets published without review?
I'm sure I could look this up but, well, I'm lazy.

T Ryan Gregory said...

I have the same problem (likely in looking for the same papers!) with PNAS vs. media reports. I have actually more or less stopped trying to find the paper being summarized as it rarely is to be found when they say.

Rosie Redfield said...

Me too! I've just sent them an email of complaint: pnas@nas.edu.

RPM said...

This may be the paper you're looking for: http://arxiv.org/abs/0705.4062

And, yes, it looks fishy. From what I can gather, they only looked at protein coding sequences. It appears highly unremarkable.

In regards to PNAS peer review, they have multiple tracks. If an academy member is author, he/she gets to choose who reviews the paper and ultimately decides whether the paper passed review. The same thing for if an academy member communicates the article for a non-academy member. Finally, one can directly submit a paper to PNAS and it goes through the conventional peer review process.

PonderingFool said...

In regards to PNAS peer review, they have multiple tracks. If an academy member is author, he/she gets to choose who reviews the paper and ultimately decides whether the paper passed review. The same thing for if an academy member communicates the article for a non-academy member. Finally, one can directly submit a paper to PNAS and it goes through the conventional peer review process.
*******************************
In addition, academy members only get a certain number of slots per year. There is talk of reducing this. A number of faculty members I know hold their slots to be able to fast track papers in order to beat a competititor to the punch (i.e. first to publish) because a track I submission can greatly accelerate the time between submitting and getting published. Basically it gives academy members a leg up on combetition that are non-members.

Anonymous said...

ponderingfool: you're close but not exactly correct. The Info. For Authors pages have the full rundown:
http://www.pnas.org/misc/iforc.shtml#submission

Anyway: I know at least one longtime Academy member who has *never* directly communicated a paper to PNAS - all his stuff goes through full blind peer review, same as the papers submitted by the rabble. *Very* cool, that dude.

Torbjörn Larsson said...

An irritating effect of commercial publication, which the email alert function doesn't quite make up for. This is another reason why I like the arxiv for physics and other open source initiatives.

The authors bungled that one though, they didn't link to the paper's supplement.

From what I can gather, they only looked at protein coding sequences. It appears highly unremarkable.

I don't know the context of the field, but I will comment because it looks superficially like some work I have had the opportunity to contribute with in material processing. And I can't resist comment on a paper which refers to fitness spaces, folding spaces and stability spaces. :-P

The authors claim that theirs is the first work approaching the idea of describing folding stability from first principles, ie others have made some ad hoc descriptive model. Superficially it looks legit, independence from other fitness effects and protein interactions admits separating out the desired solution from a parabolic diffusion equation for "stability space". A simple model of first order effects, and assuming these models are correct and valuable they are presumably better than earlier ad hocs. Much better.

IANAB, but I'm a bit bemused by the nice fit for the predicted behavior to data, as the figure contains selected proteins. (Described in the missing supplement.) [But if it's correct, I guess I also have an increased trust in Larry's description of protein folding (with subsequent stabilisation) as "entropically driven". :-)]

But while they don't apply their model on functional RNA, I don't see why it couldn't be used there too, assuming it is correct and useful.

Greg Laden said...

I agree (with Larry, in the post) and have had that happen to me as well with PNAS as well as, recently, another journal. Can't remember which journal it was, but exactly the same thing...