X-posted Ancient DNA publishing

From
Date
Cc
Reply-To
The retrieval and sequencing of DNA from paleontological materials is a
growing subdiscipline.  Perhaps paleontologists would like to see some of
the commentary on this topic.  Little of it pertains to the K/T boundary!!
(Just to add a new subject to the list).

               RE>Ancient DNA publishing standards          7/14/95

This letter is intended to solicit some general discussion regarding
publication standards for ancient DNA sequences. Although I am relatively new
to this field, I have a Ph.D. in molecular biology and have been using
standard recombinant methodologies (PCR, cloning, mutagenesis, gene knockouts,
etc.) since 1981 in oncogene research. As an amateur paleontologist, I have
followed with interest the development of methodologies to detect and sequence
DNA from fossil sources. Over the last two years I have attempted to obtain
multi-million year old DNA from amber and Mesozoic samples and have
collaborated with Mary Schweitzer on her T. rex project.

What disturbs me is the easy with which laboratories can publish DNA sequences
from amber, Mesozoic bone samples and even "resurrect" bacteria without
confirmation from independent laboratories using duplicate specimens. At
considerable personal expense I  have been unable to detect any trace of DNA
in Dominican amber-preserved stingless bee tissue, even when using PCR primers
designed directly from the published sequences. Perhaps this reflects my
technique (in gene knockout studies we routinely detect DNA from a single cell
of a single copy gene by PCR so I am not an unfamiliar with high sensitivity
PCR) or examination of an insufficient number of samples (twelve, most with
lots of tissue), or that the amber was from a different mine. I am willing to
give the benefit of the doubt, but before I can seriously conclude that the
published studies are not reporting contaminating sequences, independent
confirmation is essential. DeSalle has published "fossil termite" 18s
sequences in Science (257:1933), but in Experientia (49:908) he reports
(perhaps more openly?) that a duplicate sample also yielded sequences from
Drosophila, Gryllus and two fungi as well as "ancient" termite. Is it possible
to have confidence in termite tissue that yields more Drosophila DNA than
termite? Cano's lab has reported viable bacteria (100's of isolates) from
Dominican amber. If this is such a straight forward technique why was no
independent confirmation sought by the journal on such an astounding claim?? I
have seven years experience in bacteriology and vaccine development and have
known situations where bacillus contamination is nearly impossible to
eliminate, particularly where the microbe is routinely studied. Scott
Woodward's "Dino DNA" is another case in point. Here is a respected researcher
with excellent credentials and accomplishments in analyzing ancient human DNA,
who gave in to the excitement and published DNA sequences phylogenetically
indistinguishable from human, but extracted from Mesozoic bone fragments.
I do not intend to suggest that these published reports are incorrect, only
that they are completely unbelievable until independently confirmed. Using the
standards followed by most scientific disciplines, these studies would never
have been published.

So my question is: why are ancient DNA researchers so reluctant to obtain
independent confirmation of >10,000 year old sequences? Is it perhaps that it
has been unsuccessful?
Until rigorous publication standards are developed and adhered to, ancient DNA
studies will be just another series of cold fusion experiments to careful
scientists in the biomedical field.
Any comments?
Mark Marshall
Associate Professor of Medicine, Biochemistry
and Molecular Biology
Indiana University School of Medicine
Mark_Marshall@iucc.iupui.edu

Cross-posted with the permission of Mark.

Jere

Jere H. Lipps,  Director
Museum of Paleontology
University of California
Berkeley, California 94720 USA

Voice:  510-642-9006.  Fax:  510-642-1822
Internet:  jlipps@ucmp1.berkeley.edu