The “Infallible” DNA Test

Foxes (Vulpes vulpes) and pine marten (Martes martes) produce scats (faeces) that look quite similar. There are a list of rules, suggestions, smells and descriptions used by field naturalists to help identify and differentiate the two scats, but a number of studies have shown that even the “experts” get it wrong, and that DNA identification is the best way to differentiate the two species. This was established in Davison’s 2002 paper “on the origin of faeces…”

Pine marten photographed in Co. Waterford by Maurice Flynn

Pine marten photographed in Co. Waterford by Maurice Flynn

A recent study by Baines et al. (2014) from the Game and Wildlife Conservation Trust, and published in Wildlife Biology has examined this issue in the Scottish Highlands. The study aimed to assess if pine marten and foxes had expanded in forested sites across the Scottish Highlands since an initial study, which was conducted in 1995. In 1995, the study was based on the direct observations of the field collectors, and at that time, it was not possible to use DNA analysis to verify field identified scats. With this in mind, Baines et al. decided to genetically test a subset of the samples from the most recent survey to verify the species, and to come up with a correction factor based on the field versus genetically identified results.

Baines and colleagues selected woodlands where capercaillie (Tetrao urogallus) were present, as some previous research had shown that both pine marten and fox have a negative impact on their abundance. In addition to the study sites used in 1995, Baines et al. included an additional five sites to increase the sample size for DNA analysis.

414 scats, representing 30% of the overall scats collected were used for DNA analysis. These included 217 scats that the collectors could only tentatively identify in the field (52.4%), and 88 scats (21.3%), where the collectors were fairly sure they had correctly assigned the species, and an additional six scats (1.45%) that they were fairly sure were neither fox nor marten. But this is where the details get a little fuzzy, because after this point the authors say the following “The scats sent for DNA analysis were a randomly selected subset from within each certainty category”, and that was the end of the details provided regarding the DNA analysis.

As someone who has tested a large number of scats over the years for the presence of pine marten and fox DNA, I really wanted to know how the analysis was conducted, as the overall success rate was pretty good at 75%. This is a good result for this type of DNA sample, as scats have low quality and quantity DNA, and as most of the samples chosen for DNA testing couldn’t be identified in the field, I would have assumed that they might be old or of generally low quality, and therefore less likely to provide good quality DNA.

The only reference point to a lab was mentioned in the acknowledgements of the paper where the Forest Research was acknowledged for undertaking the DNA analysis of the scats. I would be quite satisfied if the authors had cited a previously published study that had described the tests used to identify these species, but no references were provided, and the reader is left to assume that DNA testing is somehow infallible.

There are plenty of problems associated with DNA testing of scats. First of all, we don’t know if the samples were tested using a specific species test, or if each sample was DNA sequenced. A specific test is based on the use of primers which are designed to specifically target the species of interest, where you get a yes/no type answer, while DNA sequencing can be used to obtain the letters (ATGC) that make up the unique and specific DNA of the target species. Specific tests have to ensure that the primers will actually work with all the animals in that population, as a test designed for animals in one part the species’ range may not actually work with animals in another part of their range due to small genetic differences simply caused by geographic variation, but have not been incorporated in the test design.

I also think there is another issue here regarding authorship. The DNA evidence is an integral part of this study, yet the lab that did the work were merely acknowledged. Had the lab that actually conducted the DNA analysis been involved in the writing of the manuscript, the queries I have outlined would have been readily and easily addressed.

Which leads me on to my next issue; why did the reviewers and editors not question the lack of clarity and detail in this study? Considering the results of this DNA work were subsequently used to inform some kind of abundance estimate for both species, surely the reviewers and editors wanted more information regarding how that work was conducted? I can only assume that all of those people came from a non-genetics background, and simply did not question the analysis. This raises other issues about peer-review of multi-disciplinary studies, and when extra reviewers, or at least a reviewer from a different background should be sought to cover all aspects of the study.

Seventy seven percent of the of samples that were DNA tested were correctly identified by the field collectors, and 69 scats were misidentified in the field, with 64 of those being incorrectly labelled as fox, while the DNA result indicated that they had been pine marten. 116 samples had been labelled as fox in the field, while only 52 of those were genetically identified as fox. 195 scats had been labelled as pine marten in the field, but in total, 254 samples were genetically pine marten. The authors have also assumed that the scats that were not tested had been correctly identified in the field, which pretty much defeats the purpose of the DNA test in the first place.

Based on these results, the study appeared to show that there was a 3.9 fold increase in pine marten “sign index” (i.e. scats) between the study conducted in 1995 and the most recent survey conducted in 2009. The authors do conclude that this may not be an adequate way to estimate actual pine marten and fox abundance, and suggest that future studies should incorporate genotyping to genetically identify individual animals. The overriding implication in this study is that capercaillie are declining in the woodlands where this study was conducted, and unfortunately for the pine marten, their apparent population increase has been somehow correlated with the capercaillie decline. The author’s closing sentence is as follows “A greater appreciation of the factors that determine predator abundance may be necessary to inform conservationists and forest managers alike whether more funding for legal fox control, together with a licensed removal of protected martens, may help reverse the decline in capercaillie”. Other factors that might affect capercaillie abundance were not discussed such as habitat loss, climate change, and the several other factors that may also be implicated in the capercaillie’s decline. Studies such as this could be used to support the (licensed) removal of pine marten from such areas, and I personally think that greater scrutiny is needed for studies that appear to be chasing agendas.
David Baines, Nicholas Aebischer, Allan Macleod & John Woods (2014). Pine marten Martes martes and red fox Vulpes vulpes sign indices in Scottish forests: population change and reliability of field identification of scats Wildlife Biology, 19, 490-495.

Angus Davison, Johnny D. S. Birks, Rachael C. Brookes, Tony C. Braithwaite, John E. Messenger (2002). On the origin of faeces: morphological versus molecular methods for surveying rare carnivores from their scats The Journal of Zoology, 257, 141-143. 

%d bloggers like this: