Comment on Bohmann et al. Strategies for sample labelling and library preparation in DNA metabarcoding studies