Species assignment from trace DNA sequences: an in silico assessment of the test used to survey for foxes in Tasmania

Diagnostic DNA tests have become important for species detection from environmental samples and are increasingly applied to the analysis of ecological systems and in wildlife management. The availability of reference DNA sequences from many taxa enables the development of diagnostic PCR primers. Where there is a high risk of false‐positive PCR amplification, or where even a low rate of false positives has serious management implications, DNA sequencing is crucial for accurate specimen identification. The ability of DNA sequencing to discriminate among target and non‐target species must be explored for each system. The red fox Vulpes vulpes is an invasive pest in Australia. A fox‐specific PCR and sequencing test has been applied to a systematic survey of scats collected in Tasmania, where the management of a recent fox incursion remains controversial. We investigated the risk that DNA sequences obtained using this test might be mistakenly assigned to fox in cases of non‐specific amplification, or mistakenly assigned to another species when fox DNA was correctly amplified. We conducted an analysis of barcoding efficacy using cytochrome b sequences from 74 vertebrates. In our analysis, no non‐fox sequences were identified as fox (false positives) and no genuine fox sequences were misidentified (false negatives). This two‐stage DNA test, including PCR screening and sequencing steps, can reliably discriminate fox DNA from that of other Australian species. Synthesis and applications. DNA tests are attractive to wildlife managers interested in detecting species that are cryptic or difficult to identify. When DNA data directly influence management decisions, it is important to understand the limitations of the genetic markers and the likely causes of failed identifications or erroneous species assignments. We show that short cytochrome b sequences can provide high specificity for vertebrate species assignment. We highlight the importance of developing appropriate reference sequence data bases for each study system, and of evaluating the potential for misidentification of different taxa.