Wolbachia

Klopfstein, Seraina, Kropf, Christian & Baur, Hannes, 2016, Wolbachia endosymbionts distort DNA barcoding in the parasitoid wasp genus Diplazon (Hymenoptera: Ichneumonidae), Zoological Journal of the Linnean Society 177 (3), pp. 541-557 : 547-550

WOLBACHIA INFECTIONS AND CORRELATION ANALYSIS

Twenty out of 54 Diplazon individuals or eight of the 16 species tested positive for Wolbachia infections, ( Table 2). To test whether the infections coincided with a failure in DNA barcoding, we used a method correcting for phylogenetic relationships by comparing an evolutionary model assuming independent with one based on dependent evolution. Barcoding was considered successful if the species were recovered at the most sensitive threshold of 0.5%. Both the ML approach and the Bayesian approach significantly preferred the dependent over the independent model of evolution in both ML model testing (likelihood ratio test statistic = 10.29, P <0.0358) and Bayesian testing [Bayes factor as 2 X (difference in the logarithm of the harmonic mean) from stan- dard MCMC = 5.14, from reversible-jump MCMC = 2.84; Bayes factors are considered significant from a value of 2 and highly signficiant from a value of 5; Kass & Raftery, 1995].

To examine the extent to which this method was sensitive to taxon sampling, we also analyzed the dataset under the assumption that the aberrant specimen of D. hyperboreus , which was retrieved apart from the specimen with the typical morphology in the ITS2 tree, actually represents a different species. This change had a drastic impact on the

Species

0.2

0.1) % 50.9 (1 PC 0.0

–0.1

–0.2

–0.3 –0.2 –0.1 0.0 0.1 0.2 Isometric size

result, which thus has to be considered with caution. With this new dataset, the significance disappeared, with the likelihood ratio test statistic dropping to 5.47 and the P value increasing to 0.24.