Tuesday, August 18, 2009

Factor this

Wolfgang G. Stock of the Department of Information Science at the Heinrich-Heine-University in Düsseldorf has an interesting essay titled "The inflation of impact factors of scientific journals" coming out in ChemPhysChem (subscription required). In it he makes the point that impact factors (IF), h-indices and even eigenfactor scores for journals are flawed and really not very informative. For starters, he points out-

"For example, the British library holds more than 40000 scientific serials and adds about 800 new journals each year, whereas the two most comprehensive multidisciplinary databases, namely Elsevier's Scopus and Thomson Reuter's Web of Science (WoS) cover only 16000 (Scopus) and 10000 (WoS) periodicals."

Note that because of the differences in periodicals surveyed by Scopus and WoS, the IF's obtained from these two sources often disagree on the IF for a given journal.

Stock notes-

"All indicators that work with relative frequency measures (ie, all Group 2 indicators) suffer from serious statistical problems. It is a precondition for calculating average values (in our case: average cites per publication) that there is a Gaussian distribution... In journal informetrics this is not the case."

Seems like a serious flaw to me...

He asserts-

"In no case is it possible to use a journal impact factor on the article level to evaluate the influence of an article, an author or an institution."

Of course we all knew that.

Wish the bean-counters did...

Friday, August 14, 2009

PSIfi careers

Recently there was quite a bit of discussion in the blogosphere regarding a former SLAC National Accelerator Laboratory technician destroying $500,000 worth of proteins crystals. This incident occurred at the Joint Center for Structural Genomics, a part of the NIH-funded Protein Structure Initiative (PSI). You can read about this incident over at DrugMonkey's place.

A couple of the comments following DM's post made the point that these PSI labs are little more than large protein structure factories. 'Tis true. According to the global repository of all that is worth knowing, Wikipedia, PSI labs produced >3400 new structures between the inception of PSI in 2000 and November 2008, with ~1900 being new folds. One can argue whether or not the scientific output is worth the (so-far) >$700 million cost - many people have, including prominent structural biologists such as Greg Petsko. I don't really want to get into that here. What I do want to talk about is one of the human costs of PSI.

PSI employees a great number of scientists. Many of these are employed as postdocs. The problem with this is that many of these "postdocs" aren't really receiving research training.* Sure, they solve many structures, but that in and of itself is not research. The PSI labs are trying to solve as many representative protein structures as possible and each (at least initially) tend(ed) to focus on a single organism (or maybe a couple) as the source of material. A postdoc in a PSI lab could solve a whole bunch of structures, but the only relationship between the proteins would be the organism they were derived from. Many of the proteins were/are of unknown function. And the vast majority of the publications reporting the structures are short letters or notes with many authors... This is not research and it is not research training.

We're not associated with PSI, but do happen to have very good facilities for x-ray crystallography at Big State U. Consequently, in recent faculty searches we have had quite a few applications from PSI-trained crystallographers. Despite long publication lists, none of them have ever made it to the interview pile. Why? Well, an unusually large portion haven't written a coherent research plan (unusually large compared to applications from non-PSI trainees). I suspect they simply don't know how to construct a worthwhile/fundable research project. They haven't received the training. Many of the remainder suggest decent systems to work on (proteins of medical relevance), but propose nothing more than crystallography, which nowadays is not enough to get funding. And they have no preliminary data. None.

You could argue that a postdoc who takes a position in a PSI lab hoping to move on to an academic career is naive. They are. Judging by the applications we've seen, there are a bunch of naive postdocs out there. However, I think the PSI labs share some responsibility here. They should make it very clear to postdoc applicants that the positions they are applying for are unlikely to provide the training necessary for an academic career. Perhaps some of them do. Even better would be to call the positions by a name other than "postdoctoral scholar/fellow/trainee" - something employing "technician" seems more appropriate.

Petsko has stated "As a structural biologist, I want to train people who use structure determination as part of what they do. It is not the end in itself, nor should it be, not any more."

Sounds about right to me.

* Many, not all. I am aware that some PSI postdocs receive excellent research training.

Thursday, August 13, 2009

Thursday, August 06, 2009

iPhone apps

I have a question for my iPhone-wielding compadres: what apps do you find truly useful? Other than the ones that come standard.

I've grown partial to the following:

Evernote - particularly useful since it automagically syncs with my laptop via the internet. Good for recording notes, saving web pages etc. A useful catchall and organizer for information.

TripIt - useful as long as you're comfortable with them potentially having access to your travel info. It's also available on LinkedIn if you happen to have an account there. I like the built-in flight status and map features.

UrbanSpoon - for when you're traveling and have a need for good Thai food...


We had one humungous mother of a thunderstorm come through a couple of days ago. Knocked out the power in the building I work in.

It fried my office phone.