Caution when Citing these Structures
The University of Alabama at Birmingham (UAB) requested on Dec. 8th for the removal of 12 structures from the PDB. The 12 structures are spread across 9 journals including Nature, Cell and the Journal of Molecular Biology.
Here are some highlights from the request concerning these structures:
1BEF is a synthetic version of 1JXP (I wrote about 1NS3 being the structure used, 1JXP and 1NS3 contain the same origin, orientation, space group and unit cell).
My favorite:
1CMW had its B factors derived by subtracting 16.00 from values found in the structure 1TAQ
1DF9 which replaced 2QID with the update including the removal of 155 waters some of which had hydrogren bonding distances less than 1 Angstrom.
1G40 is without any water despite diffracting to 2 Angstroms also there is a strange update to the unit cell parameters in Feb. of 2007 (fits into the time line previously posted)
1G44 contains 36 chemically impossible close contacts
1L6L has either 2036 residues or 2366 residues, 1011 waters or 1522 waters, etc…
2OU1 also involved heavy drinking
1RID and 1Y8E contain poor geometry, strange B factors and very high solvent content
2A01 has great electron density that correspond to physically impossible features
2HR0 has already been trashed.
————–
I noted that there were 449 citations that referred to these dubious publications on Dec. 10th. A week has passed since the original request made by UAB. Currently, only 1 structure (1BEF) along with the corresponding paper has been retracted. Now according to Google Scholar, not even a week later, 6 more citations have been added.
I hope that other publishers are able to review the data involving these structures quickly and act accordingly as to prevent the further spread of inaccurate information.
Note: there is going to be a delay from when this information was announced to when papers (patents, books, etc) work their way through the system, but the sooner the better
Eric
December 17th, 2009 at 1:45 PM #
Thanks for playing Sherlock Holmes for us. A more thorough exposition of Murthy’s evidence trail, ala Janssen, et al., might make a very interesting paper.
Martin Stoermer
December 17th, 2009 at 9:41 PM #
might make a very interesting paper
true but most scientists I feel wouldn’t like to be drawn into a $%&*fight in print (unless of course it added to their cite count and impact factors…)
Sean
December 17th, 2009 at 9:57 PM #
@Eric – thanks for taking the time.
@Martin
a paper will be out on the issues with these structures