Delimitation of allopatric populations into species remains subjective and largely arbitrary. Many cold-adapted species from the subarctic and Central and Southern European Mountain systems display frequent allopatry with disjunct distributions of their populations. The same concerns Holarctic species which are many and which almost regularly show various degrees of differentiation between the continents. In this study, we analyze high-throughput target enrichment data for ten groups of arctic-alpine and Holarctic lepidopteran species sampled from four main regions across the Holarctic realm – Fennoscandia, North America, Alps and Altai. We first aimed to assess whether the genetic differences in the nuclear genome reflect observed DNA barcode divergences and second, whether the gap between population and species-level differences can be dissected using genomic data. We compared the phylogenetic trees and uncorrected pairwise genetic distances obtained from target enrichment and the mitochondrial cytochrome oxidase I (COI) barcodes for each of the study species. Additionally, we also performed a suite of population genetic and species delimitation analyses to further shed light on patterns of intraspecific variation using a large number of nuclear markers. We observed that in about one half of the cases, DNA barcodes tended to show phylogenetic relationships similar to the target enrichment markers. We report varying levels of nuclear genetic differentiation among the populations analyzed, starting from low differentiation of geographically separated populations to the deeper separation of some Nearctic population and further arctic-alpine disjunction. Given that no single consistent pattern emerged across different case studies, we demonstrate that the delimitation of allopatric populations into species could be done much more efficiently and in a consistent manner if based on a large set of universal genetic loci, which would help in reaching standards for taxonomic delimitation of allopatric populations.

The nuclear genomic data was generated using target enrichment method. Barcodes were generated through standard barcoding lab protocol and sanger sequencing. The raw target enrichment data was processed using TEnriAn pipeline (Mayer et al. 2021) which is specifically designed for processing of target enrichment data. The downstream phylogenetic and other analyses were carried out using commonly used programs such as IQtree (version 2), STRUCTURE, etc., and species delimitation analyses were carried out using programs tr2 and SODA. 

The barcode sequences were processed using Geneious Prime to generate the multiple sequence alignments and phylogenetic trees were generated using IQtree 2. 

For other analyses such as PCA and generating P-distance plots, R packages adegenet and ggplot2 were used. 

To open the Maximum likelihood phylogenetic trees and ASTRAL trees (ending with .treefile or .tre), FigTree program would be needed. Nucleotide alignments can be viewed using any alignment editor such as AliView, Geneious. The vcf files can be read using vcfR package in R.