Aim: Seascape genomics studies aim to understand how environmental variables shape species diversity through genotype-environment associations. Identifying these effects on lecithotrophic larval species that live in intertidal zones is particularly challenging because they are subject to environmental heterogeneity and anthropogenic events. Here, we evaluate how biotic and abiotic features in the Southwest Atlantic littoral zone can affect a high dispersal species’ present and historical demographic.

Taxon: Perinereis ponteni.

Methods: We investigated population genetic diversity, connectivity, and past dynamics using 23,300 SNPs generated using Genotyping by sequencing. We tested whether environmental abiotic variables could explain the variance found in genotype frequencies using isolation-by-environment (IBE) and landscape association approaches. These data, combined with paleodistribution simulations and oceanic circulation modeling, were used to infer species demographic history and connectivity patterns.

Results: Along with high levels of connectivity detected, we found a genetic boundary in the southeastern region of Brazil around Cabo Frio (Rio de Janeiro) and a cline trend for some loci. The paleodistribution simulations reveal a spatial refuge in the southeast during the Last Glacial Maximum (21 kya), with the expansion of the northern region. We identified 1,421 SNPs with frequencies associated with eight environmental variables, most of which were related to temperature - the main environmental factor determining IBE.

Main conclusions: Perinereis ponteni, a polychaete with high gene flow capability responds to biogeographic barriers, highlighting the importance of biotic and abiotic factors in shaping population connectivity. Furthermore, the effect of temperature indicates that future climate change and ocean warming can hugely impact this species.

Genomic DNA was extracted from 61 specimens of Perinereis ponteni using the protocol by Doyle and Doyle (1987). DNA quality and concentration were assessed by gel electrophoresis, spectrometry (NanoDrop Lite spectrophotometer; Thermo Fisher Scientific), and fluorometry (Qubit 3.0 Fluorometer; Invitrogen). Approximately 100 ng of the extracted DNA per sample were sent to EcoMol Consultoria e Projetos Ltda (Piracicaba, Brazil), where GBS libraries were prepared according to the protocol described by Elshire, Glaubitz, Sun, Poland, Kawamoto, Buckler & Mitchell (2011) and modified by Nunes, Liu, Pértille, Perazza, Villela, de Almeida-Val ... & Coutinho (2017). The libraries were sequenced in 100 bp single-end fragments using the Illumina HiSeq 2500 platform at the Centro de Genômica Funcional ESALQ-USP.

Sequenced fragments were first filtered using Seqyclean v1.10.09 (Zhbannikov, Hunter, Foster & Settles, 2017), removing sequences with an average Phred quality score ≤ 20, contaminants, and adapters. After filtering, Ipyrad v0.7.30 was used for demultiplexing, quality and size filtering, fragment clustering, and SNP prospection. Reads with more than five Ns or shorter than 35 bp were discarded. The minimum read depth was set to six for calling consensus sequences within the samples, and the maximum depth was set to 10,000. The clustering threshold was set to 90%, and the maximum number of SNPs per locus was set to 30. A locus had to be present in 50% or more of the samples in order to be retained in the final dataset. All other parameters were maintained at their default values. SNPs were first visualized as an occupancy matrix using the divergent option in the Matrix Condenser tool; samples with ≥ 40% missing data were removed. The resulting variant call format (VCF) files were used as input to VCF tools v.0.1.16, with the option max-missing 0.5, followed by the PLINK software, which filtered SNPs for missing data (geno < 0.45), rare alleles (MAF > 0.01), and linkage disequilibrium (indep-pairwise 50 5 0.5). The resulting files were converted to specific program formats using PGDSpider v.2.1.1.5.

The files included are text files, ready for the analyses. You could open using vim at the terminal or any text editor.