Population Genomics and Biogeography of the Northern Acorn Barnacle (Semibalanus balanoides) Using Pooled Sequencing Approaches

Abstract

The northern acorn barnacle (Semibalanus balanoides) is a robust system for the study of evolutionary processes in the intertidal. S. balanoides has a well-characterized ecology, a wide circumboreal distribution, and a life history characterized by tractable environmental stressors at various ecological scales. In this chapter, we discuss a variety of topics concerning the development of S. balanoides as a model in ecological genomics as well as inferences of demography and historical phylogeography. In addition, we introduce two novel genomic tools for S. balanoides: the complete mtDNA sequence and the second draft of the nuclear genome (Sbal2). Using these tools, we conducted a reanalysis of previously described mtDNA haplotypes, a and b, as well as genome-wide levels of variation and population structure across the North Atlantic using pooled sequencing approaches. Analyses of sequence data from older and more recent Illumina platforms revealed the effects of technical bias in the estimates of population genomic metrics. We found concordant levels of nuDNA and mtDNA genetic variation with no evidence of demographic bottlenecks. We observed low genome-wide F _ST values across the Atlantic, suggesting a large number of ancestral polymorphisms and shared standing variation across the basin. Comparisons of genome-wide estimates of F _ST with those derived from a discriminant analysis of principal components uncovered population-structure-informative SNPs. This suggests the existence of latent population structure across broad scales, despite the capacity for extensive planktonic dispersal. Noticeably, our samples collected in Iceland displayed higher similarity to North American populations than to the rest of Europe. We hypothesize this is consistent with a periglacial refugium in Iceland concomitant with a barrier to gene flow caused by the North Atlantic current. Lastly, we discuss challenges and opportunities for the improvement of genomic tools in barnacles. Our reflections in this area are easily generalizable to most natural populations.

Data Availability

Companion to this book chapter, we have made available the reference sequences for Sbal2 (DDBJ/ENA/GenBank accession PHFM00000000) and the complete mtDNA of S. balanoides (GeneBank accessions MG010647, MG010648, MG010649). Other genomic resources have been made available in FigShare (figshare.com). The gene feature file (GFF, v3) of Sbal2 from the Drosophila AUGUSTUS model, the UniProt BLAST hits file, and the nuDNA and mtDNA SNP tables are available at DOI: 10.6084/m9.figshare.6682995. The additional COX I sequences from our analyses can be found in GeneBank accessions MG925538–MG925662 and MG928281–MG928323. We note, however, that files corresponding to the 2017/2018 pool-seq datasets from Iceland, Maine, and Rhode Island, as well as the transcriptome, form part of larger analyses which are currently active in our research group. These datasets will be released in the future with their corresponding publications.

Glossary

Admixture

Refers to the process in which previously isolated populations begin interbreeding.

Ancestral polymorphisms

Genetic variation present in two (or more) species, subspecies, or populations that appeared prior to divergence.

Assembly graph

A graph that can be traversed to create an assembled DNA sequence. The most commonly used assembly graph is a De Bruijn graph.

Contig

DNA or RNA sequence, typically assembled from multiple overlapping short sequence reads.

Coverage

Indicates the number of times that a particular genomic region was sampled by mapped reads produced by a sequencing experiment.

COX I

Cytochrome c oxidase I, a gene encoded in mtDNA involved in the electron transport chain. This gene is commonly used in population genetic studies and species identification or DNA barcoding.

D-loop

The mitochondrial DNA control region, also known as the displacement loop. It contains the sequences for the origin of replication and transcription of the mtDNA molecule. Its high rate of DNA substitution makes it suitable for analyses of closely related populations and species.

De Bruijn graph

A directed graph representing overlaps between k-mers present in a set of reads. Nodes are represented by k-mers and edges by (k − 1)-mers.

De novo genome assembly

The process of stringing together overlapping DNA sequence reads to make longer DNA sequences, called contigs. Perfect genome assembly would produce 1 contig for each chromosome.

Effective population sizes (N _e)

The effective number of breeding individuals in a population, equivalent to the idealized population size in which the effects of stochastic sampling on allele frequencies (i.e. genetic drift) are similar to the real population of interest.

High-throughput sequencing (HTS)

Highly parallelized DNA sequencing that produces millions to 100s of millions of DNA sequences of varying length (50–250 bp for the Illumina platform; 1,000 to >20,000 bp for the Pacific Biosystems (PacBio) and Oxford Nanopore platforms.)

Homoplasy

A condition where a character is shared by a set of species or populations that is not shared by their common ancestor. In DNA terminology, it may refer to the independent mutation (or back-mutation) to the same nucleotide state in two populations.

Incomplete lineage sorting

The process by which a phylogenetically informative marker is shared among species or populations in which other markers have diverged to fixation in each population.

Indels

An insertion or deletion in a DNA sequence.

K-mer

A DNA sequence of length k. In genome assembly, k-mers are generated by splitting reads into smaller pieces of length k.

Long reads

DNA-sequences longer than 1,000 bp.

Mapping reads to a reference

The process of identifying a subsequence or multiple subsequences in the reference genome that matches or approximately matches a read.

N50

A statistical measure of the average length of a set of sequences (or contigs). N50 measures the length N such that 50% of all bases are contained within sequences with length less than or equal to N.

Panmictic

An idealized demographic model in which all members of a population mate randomly, resulting in panmixia.

Pool-seq

An experimental approach for the quantification of genetic variation in populations through the pooling and subsequent sequencing of multiple individuals.

Reduced representation libraries

An experimental approach to quantify genetic variation in populations by sampling a reduced (~10%) portion of the genome to high coverage.

Reference genome

A set of genomic sequences that represents the genome of a population or species. These sequences may include DNA from multiple individuals.

Sbal1

The first generation of the Semibalanus balanoides genome.

Sbal2

The second generation of the Semibalanus balanoides genome.

Sequencing bias

The introduction of sequencing artifacts by the unequal sampling of DNA sequences due to characteristics of the target sequence, such as GC content.

Short reads

DNA sequences with lengths ranging from 50 to 200 bp.

Single-nucleotide polymorphisms (SNPs)

A genomic variant occurring at a single-nucleotide position in genomic sequences.

Standing genetic variation

Allelic variation that currently exists within populations as opposed to new variants arising by de novo mutation.