Data and code from: Spatial and temporal admixture patterns from farm oysters supplementing wild population recruitment
Data files
Apr 28, 2026 version files 103.24 MB
-
NYC-oyster-admixture.tar.gz
103.24 MB
-
README.md
4.54 KB
Abstract
Aquaculture is a fast-growing global industry providing important dietary and economic benefits. Unlike other coastal aquaculture, bivalve aquaculture diet depends on natural productivity rather than introduced nutrients, avoiding detrimental environmental effects. In fact, farmed bivalves provide ecosystem services during grow-out through their water filtration, nutrient sequestration, and habitat provisioning. For shellfish species that are farmed amidst declining wild conspecific populations, recruitment supplementation has been hypothesized as an additional ecosystem service, but there are also risks if domestication has resulted in low relative fitness in the wild. This study uses population genomics to contrast wild eastern oyster populations in regions with and without aquaculture farming, providing the first documentation of farm-source recruitment in an open-coast estuarine system.
Using double-digest restriction-site associated DNA (ddRAD) with subsequent confirmation and refinement based on low-coverage whole genome sequencing (lcWGS) data, we find no admixture in most of the Hudson River (where aquaculture is prohibited) but frequent low-level introgression among individuals in western and central Long Island Sound. Almost all admixed oysters are later-generation backcrosses, consistent with historical recruitment supplementation from oyster spawning on farms and low recent rates of immigration into wild stocks. We discuss both demographic benefits and possible evolutionary impacts of this introgression from domesticated oyster strains.
Short Running Title: Admixture between farmed and wild oysters
1. Overview
This zipped repository contains the data and code for Hudson River Raritan and Long Island Sound oyster admixture manuscript "Spatial and temporal admixture patterns from farm oysters supplementing wild population recruitment".
We assessed admixture using double-digest restriction-site associated DNA (ddRAD) data across spatially extensive samples. Subsequently, low-coverage whole genome sequencing (lcWGS) was applied to a subset of repeat samples plus additional aquaculture reference strains and the Connecticut wild population. Simulations were used to explore the farm-to-wild immigration history that could explain the observed patterns of admixture.
2. Contents of the dataset
This repository includes the essential processed data (VCF and Beagle files) used for downstream analyses, along with the full analysis workflow. Raw sequence reads are deposited in the SRA (BioProject PRJNA1150351). Pipelines for processing raw reads to VCF/Beagle files are provided, along with scripts for all downstream analyses. Some intermediate and large files referenced in the code are not included, as they can be regenerated from the raw data using the documented workflows.
After decompressing NYC-oyster-admixture.tar.gz, you will see folders below:
ddRAD/: Scripts processing & analyzing ddRAD-Seq datadDocent_workdir/: conigs and input bam list used in dDocent runfiltering_workdir/&LD_pruning/: filtering scripts andREADME.mdsummarizing filtering stepsdiversity/: scripts calculating diversity metrics and plotting thempca/: script to run PCA analysisrelatedness/: script to run relatedness analysisstructure/: scripts and parameters to run STRUCTURE analysisvcf/: VCF files containing SNP data of 380 samples. Two files correspond to 8,132 “ddRAD filtered SNPs” and 5,625 “ddRAD LD-pruned SNPs” used in the manuscript.sample_lists/: bam list, vcf list, popmap, per-sample sequencing depth files
lcWGS/: Scripts processing & analyzing low coverage WGS dataproc_scripts/: scripts for data processing that take in raw fastq files to generate analysis-ready bam files, and aREADME.mdguiding through the subdirectories (from01_fastqcto11_downsample).angsd_scripts/: scripts for calling genotype likelihoods, running ngsld, running ngsrelate, running PCA, running ngsadmix.results_pca/: genotype likelihood file of 136 samples for 230,092 “lcWGS LD-pruned SNPs” (".beagle.gz")results_ngsadmix/: admixture analyses output (".qopt" files)sample_lists/: fastq list, bam listsnp_lists/: list of 230,092 “lcWGS LD-pruned SNPs”
figures/: Collection of scripts generating figures in the manuscriptmap_cood/: lat and lon coordinates for each sampling site
nemo_simulation/: scripts performing Nemo simulations and analyzing simulation resultsREADME.md: instructions on how to run Nemo + STRUCTURE pipeline using scripts in this foldersimulation/: series of ".ini" for each Nemo config files and bash script to parallel batch run Nemo simulations.
misc/: scripts for comparison between ddRAD and lcWGS, coding region test, and checking GC content.
Recommended Citation
Chen, Y., Kutsumi, Y., & Hare, M. P. (2025). Spatial and temporal admixture patterns from farm oysters supplementing wild population recruitment. Molecular Ecology. Data available at Dryad: [DOI: 10.5061/dryad.08kprr5hv]
Access
Raw sequence reads are deposited in the SRA (BioProject PRJNA1150351).
All scripts can also be accessed at the Github repository https://github.com/yuqingc7/NYC-oyster-admixture
Acknowledgements
Scripts processing lcWGS data are mostly modified from Therkildsen Lab scripts for low coverage whole-genome sequencing data processing and analysis.
