Search This Blog

Thursday, April 14, 2011

When to Retract

Retraction Watch posted a piece on why people/groups retract papers. The reasons people give all point in the direction that what was stated may not be true. What about the papers that are not retracted? Where is the line scientists cross that leads them to correct their mistakes officially?

Previously I've written about the peptide sequence GETRAPL. GETRAPL is the peptide sequence despayed on a contaminating bacteria phage that is found in New England Biolabs phage display kits. If you don't believe me, and you care, you have a few options. A) you can search and search for information on the sequence. B) you can call New England Biolabs and ask them if this is so. C) you can purchase a kit and use it until you start seeing white plaques in the presence of beta galactasidase. You will find GETRAPL!

That is the curious thing about modern science and our near religious faith regarding peer reviewed papers. Rather than going into a lab and reporting your results, people would prefer to find information from others. This is not uncommon. We went to war on faulty CIA information regarding weapons of mass destruction. We didn't retract Colin Powells speech at the U.N. We were wrong, but the war was just. Right? This is akin to the retraction of a paper. We can be wrong but that is not what gets a paper retracted. If you are a scientist who has spent the last 10 years studying the amazing effects of GETRAPL you are not about to retract all of your papers. You would rather just let the subject fade away while you find a way of burying the inaccurate information.

PubMed published an article on a peptide database that came across the GETRAPL confusion.
We suggest that one of the utilities for PepBank is to search the peptide sequences of interest to the user with BLAST or Smith-Waterman algorithms to find any important similarities to the known peptides collected in our database. In this example, the search can be used to remove a relatively nonspecific binder GETRAPL.


Even the last sentence is misleading. The reason GETRAPL ends up on your radar is because the phage displaying the peptide grows faster than the other phage. It has nothing to do with specificity. The sequence has been published and patented and beaten to death by cargo cult scientists. They needed to find something and they did.

So the question is simple: Do they have to retract? The Cargo Cult Scientist studies the ways in which errors occur in the minds of individuals highly educated in the ways of science. Where do they go wrong? Do their peers really get a chance to investigate and add to the story?

GETRAPL is a minor, insignificant story in terms of its impact on curing AIDS or targeting specific cells. It is however a modern day N-Ray story to the Cargo Cult Scientist. After all of the papers about GETRAPL, I have only one conclusion when I see the sequence. The authors used New England Biolabs phage display 7mer kits. That's all I can be sure of. Back in 2009, two years after the PepBank exposed GETRAPL we still have papers like the following:

Recognition of Patterned Molecular Ink with Phage Displayed Peptides
Yue Cui, Anupama Pattabiraman, Bozhena Lisko, Samantha C. Collins, and Michael C. McAlpine*
Department of Mechanical and Aerospace Engineering, Princeton University, Princeton, New Jersey 08544
Received October 2, 2009; E-mail: mcm@princeton.edu
Published on Web 01/06/2010

An aliquot of phage display library
(New England BioLabs, Ph.D. 7) (Figure 1a) was incubated with
a C8-functionalized Si surface (Figure 1b) and then eluted from
this surface to collect the bound phage (Figure 1c).First, fluorescent characterization for the binding of this phage to
C8 and Si substrates was investigated. This was accomplished by
exposing the substrates sequentially to (1) amplified single-colony


Eureka! I've done it! I've proved that New England Biolab libraries contain GETRAPL.

Education cannot teach people to Google GETRAPL. It is not considered science. Yet it is information. Education cannot teach people to design experiments. It will always be a subjective pursuit. Should Cui et. al. retract "Recognition of Patterned Molecular Ink with Phage Displayed Peptides"? They are wrong. They may have higher degrees in Aeronautical Engineering and work at an Ivy League school, but they have made a mathematical mistake. The data set was too small. The information from other scientists was ignored. They failed to consider alternative explanations. They are wrong. The research, even if correct, wouldn't be all that important. It is the fact that they erroneously attached such significance to their honest mistake that makes it a Cargo Cult Science paper.

No comments: