I would like the net readership's opinion of the following situation.
I am not trying to start trouble, only to find out whether people
believe the case in question to be a problem or not.
The paper in question is:
MJ Mahan, JM Slauch, and JJ Mekalanos.
1993. Science 259:686-688
Selection of bacterial virulence genes that are specifically induced
in host tissues.
It's an excellent and clever paper. In brief, the paper describes
a strategy for creating a library of bacteria which contain a clever
promoter-trap insert which should not cause a mutation. The host bacteria
(in this case Salmonella) are auxotrophs (in purine biosynthesis), and
the trap, if activated, will restore them to prototrophy. The library
can then be injected into animals, and recovered later from the animals.
The recovered bugs can then be screened on minimal medium to identify
constituitive prototrophs, and the remaining recovered bacteria represent
insertions next to promoters which are activated within the host.
The paper states that 15 insertions and their flanking DNA were sequenced,
and these broke down into five different genes. Two of these genes
"showed no significant homology to sequence in GenBank". 4 paragraphs
of the paper are devoted to describing the sequences found and interpreting
Where are the sequences? No accession number is listed in the paper.
No actual sequence data is presented in the paper. A search of
NCBI's RETRIEVE server with the authors found nothing.
Two of the sequences are novel. The other one's might not include anything
novel, but then again maybe a little bit is. I believe that they should
have been submitted. Science (like most journals) requires "archival
data sets (such as sequence and crystallographic data) will be offered
for deposit to the appropriate data bank".
So, what do y'all think?
Department of Cellular and Developmental Biology
Department of Genetics / HHMI
robison at biosun.harvard.edu