Computational protocol: High phylogenetic diversity is preserved in species-poor high-elevation temperate moth assemblages

Similar protocols

Protocol publication

[…] As the number of observed species usually provides only a poor representation of α-diversity in mobile insects, we chose Hurlbert rarefaction for the comparison of the α-diversity in geometrid moth assemblages between sampling plots. These rarefaction techniques have proven robust in comparing diversity values in a wide range of ecological studies.For species identification, the first step was based on distinguishing morpho-species based on their wing patterns. In a second step, 152 selected specimens representing these morpho-species were submitted to DNA barcoding, which allowed discrimination to molecular operational taxonomic units (‘MOTUs’) based on a sequence divergence threshold of 2% using the Kimura 2-parameter (K2P) distance. This also roughly corresponds to the BIN system implemented on the Barcode of Life Data System (BOLD) database. If individuals of a species varied below the 2% threshold, the sequence with the shortest distance to the other respective species was included in the subsequent analysis. DNA barcoding allowed species identification and attribution to a Linnean binomen in 127 of the 152 MOTUs. These were supplemented by 3 MOTUs of species that we identified, but where we then used data already stored on BOLD for our subsequent analysis rather than sending specimens for barcoding ourselves. Although we acknowledge that some distinct species may share very similar COI sequences below the 2% threshold, we believe that this effect will be negligible for our limited samples from one small study region. Species were then further identified based on the comparison of MOTU information from BOLD database, where all our sequence data were also deposited.Based on the K2P distance, the expected phylogenetic diversity (PD, sum of total phylogenetic branch length) for all species recorded at individual plots was then estimated using Faith’s index. Additionally, the standardized effect size of the mean pairwise phylogenetic distance (MPD) was calculated to evaluate whether the respective species assemblage is clustered or over-dispersed across the phylogenetic tree. In this approach, a low and negative value indicates clustering on the phylogenetic tree, whereas high and positive values indicate an evenly or over-dispersed phylogenetic pattern. The value of this index represents the reverese of the relatedness index. It was calculated based on 1000 random iterations drawn from the recorded geometrid species pool.All calculations and statistical analysis were conducted in R. We used the ‘vegan’ package to calculate rarefactions, ‘Picante’ for the calculation of Faith’s PD and MPD and ‘ape’ for calculations of phylogenetic distances. […]

Pipeline specifications

Software tools Picante, APE
Application Phylogenetics