(Lazy cross-post time: also at Naturally Selected)
One of the first and most important things a neophyte scientist learns–or at least, is taught–is the importance of keeping a comprehensive and accurate record. We all know it’s a good thing, and yet I’d wager most of us struggle with it. Who hasn’t scribbled a calculation or a measurement on a handy paper towel, with all the intentions of taping it in or copying it to our notebook, only for it to go missing somewhere twixt lab and office? Or for it to turn up, months later, made illegible by a mix of coffee rings and Coomassie stains?
Everything should be recorded. The practicalities of this, however, together with the effort involved in keeping a proper index, often mean that some things, often things that seem of only minor consequence, go unrecorded. And this is a real pain when you come to write the paper and realize you can’t remember whether you cloned your fragment using NotI or KpnI; or both; or perhaps it was EcoRV…
Electronic notebooks and other wild ideas might help, but there’s nothing quite like having to trawl through random pieces of scribbled-on tissue while writing up Materials and Methods to bring home the importance of a well-kept notebook.
Unless it’s a Nature retraction.
In my trawl through the F1000 database for retracted papers, I stumbled across this comment:
Unfortunately, however, a proper data notebook is not available as evidence to support our findings, which constitutes non-adherence to ethical standards in scientific research.
Note there is no suggestion of fraud or other shenanigans, and indeed, “There are several independent papers supporting” the findings (also see this one and here.) But,
In accordance with the recommendations from the National Institute of Advanced Industrial Science and Technology, K.T. therefore wishes to retract this paper.
A harsh lesson. Supervisors and students should learn it well.
This is not a case of Nature asking for a retraction, right? That’s impressive – i doubt many people would retract a paper because they couldn’t find a lab book.
An old supervisor (and institute director, and FRS, and Louis-Jeantet prize winner) of mine not only didn’t retract a paper when he found out it was not backed up by evidence, and in fact was an artefact of the use of the wrong mutant of a protein, he published a couple more papers about the fact that it was an artefact of the use of the wrong mutant of a protein.
I wonder if this is our mysterious hero K. T. being a scientist of outstanding integrity, or there was someone inside his institute leaning on them to retract for some reason. Cynical, moi?
Tom, I do wonder whether there is something else to this tale. But I’m not sure entirely what. I found three papers retracted by KT (fingering HK’s note-keeping abilities as the problem) but no whiff of an investigation. I’m keeping an open mind.
In 1982, instead of retracting your wrong paper, you could make a second one. Blessed times!
http://www.ncbi.nlm.nih.gov/pubmed/6177348
Remind me to tell the story of the time we published a paper–it was correct, but we drew the wrong conclusion (we were trying to identify a binding site). We followed that up with a better paper with the right conclusion (because I’d managed to get crystals with the ligand in place), but the first journal (higher IF) didn’t want it…
Notebooks are important, and proper documentation is critical. However, that story just doesn’t sound right. Suppose there is a problem with some information missing in a lab notebook. If the lab can again repeat the experiment (and of course document the new expt properly), doesn’t that solve the problem? Sure, there may have been some negligence, but the point of the matter is to be able to independently reproduce the results. Isn’t that the whole point of writing things up? If that can be done, it seems to me that the notebook issue simply becomes an internal lab issue (that needs to be dealt with before the researcher does any additional expts).