Demographically explicit scans for barriers to gene flow using gIMble

Abstract

Identifying regions of the genome that act as barriers to gene flow between recently diverged taxa has remained challenging given the many evolutionary forces that generate variation in genetic diversity and divergence along the genome and the stochastic nature of this variation. Here we implement a composite likelihood approach for quantifying barriers to gene flow. This analytic framework captures background selection and selection against maladaptive alleles, i.e. genomic barriers, in a model of isolation with migration (IM) as heterogeneity in effective population size (Ne) and effective migration rate merespectively. Variation in both effective demographic parameters is estimated in sliding windows via pre-computed likelihood grids. We have implemented genomewide IM blockwise likelihood estimation (gIMble) as a modular tool, which includes modules for pre-processing/filtering of genomic data and performing parametric bootstraps using coalescent simulations. To demonstrate the new approach, we reanalyse data from a well-studied sister species pair of tropical butterflies with a known history of post-divergence gene flow:Heliconius melpomeneandH. cydno. Our example analysis uncovers both large effect barrier loci (including well known wing pattern genes) and a genome-wide signal of polygenic barrier architecture.

Chromosome fissions and fusions act as barriers to gene flow betweenBrenthisfritillary butterflies

“…We use gIMble (Laetsch et al 2022), a recent implementation of a blockwise likelihood calculation (Lohse et al 2016), to infer the demographic history of speciation between B. daphne and B.…”

“…Variant calls were filtered using gIMble preprocess (Laetsch et al 2022), with the following options: --snpgap 2 --min qual 10 --min depth 8 --max depth 3, where --max depth is in units of mean coverage.…”

“…To investigate the effect of rearrangements on reproductive isolation, we followed the approach of Laetsch et al (2022) by inferring effective population sizes (N e ) and the effective migration rate (m e ) in windows along the genome. We assume that the species split time is fixed to the MCL estimate under the IM →Bda model (Table 1).…”

Chromosomal inversions and the demography of speciation inDrosophila montanaandDrosophila flavomontana

“…We analysed mean genetic divergence (d XY ) and fitted models of species’ divergence with and without long-term interspecific gene flow between and within species using gIMble (Laetsch et al 2022). This analytic likelihood method uses the joint distribution of mutation types in short sequence blocks, the blockwise site frequency spectrum (bSFS) across subsamples of pairs of individual genomes to fit a series of models of speciation history.…”

“…The resulting BAM-files were used for variant calling with freebayes v1.3.1-dirty (Garrison and Marth 2012). Raw variants were processed with gIMble preprocess (genome-wide IM blockwise likelihood estimation toolkit; (Laetsch et al 2022)). In brief, non-SNP variants were deconstructed into allelic primitives, where remaining non-SNPs were removed in addition to any SNP variant within two bp of a non-SNP.…”