Fig. 2: Variant filtering workflow and delimitation of cryptic species.
From: Demography and adaptation of a species specific pollinator associated with an invasive fig tree
Filtering workflow from single-nucleotide variants (SNVs) to single nucleotide polymorphism (SNP) datasets for different analyses (a), and results showing the composition of cryptic species using both sequences of COI gene (b) and whole-genome SNPs (c–e). a Datasets and the downstream analyses are highlighted by blue and purple boxes (see Methods). DP: read depth; MISS: missing rate in samples; MAF: minor allele frequency; QD: quality by depth; FS: FisherStrand annotation to indicate strand bias; MQ: mapping quality; MQRankSum: rank-sum test for mapping qualities; ReadPosRankSum: rank-sum test for relative positioning of reference versus alternate alleles within reads; r2: linkage disequilibrium. b GenBank accession IDs for the downloaded COI sequences of the three cryptic species and the outgroup species (E. altissima) used in the maximum likelihood phylogenetic (ML) tree are listed in Supplementary Table 3. c–e Results from ML tree, principal component analysis (PCA) and Admixture (K = 2) show that only E. verticillata sp.1 were found in the sampling locations in the nonnative range, with one (Asian) and two (The Americas and the Mediterranean) populations in the sampling locations in the native and nonnative ranges, respectively. Full names of sampling locations are listed in Supplementary Table 2.
