The seven deadly sins of DNA Barcoding (3)

It is time for sin number three in the series on the Collins and Cruickshank paper.

The use of the term ‘species identification’

The authors criticize the universal use of the term 'species identification' despite the fact that sometimes two different subdisciplines of DNA Barcoding are meant.

What's the problem here?

There are indeed two different applications for DNA Barcoding that shouldn't be confused. The primary motivation to do DNA Barcoding is the prospect of being able to identify an unknown organism based on a short sequence. I concur with the authors that this process of querying a DNA Barcode reference library with sequences of an unidentified specimen should bear the name species identification.

However, Collins and Cruickshank are concerned that species identification is often confused with species discovery. Building a DNA Barcoding reference libraries often exposes new species. When looking at this just semantically this should indeed be termed species discovery as identification represents the process of assigning a pre-existing name to an individual organism.

I have to admit that I never stumbled over this when reading DNA Barcoding papers although I have to admit that I might have overlooked it and figured out the main purpose of the study differently. That being said one needs to stress that I am an insider when it comes to DNA Barcoding and it is rather easy for me to figure such details out on the fly while someone not so much into the topic might be confused. What I have noticed quite often though is a confusion of species identification (now in its correct definition) and DNA Barcode library building. Those are completely different. Actually the former wouldn't work without the latter and this should be clearly separated in a publication. This is also related to the first sin as formulating a clear hypothesis based on an array of data points would go a long way to avoid this conflict.

Overall not a particular bad sin as in most cases I would think it is only a matter of semantic accuracy and not a real problem of study design.