Population genetics of Varroa destructor

This repository contains code used in analyses in:

Ryals D, de Medeiros BAS & Farrell BD. Migratory beekeeping facilitates genetic admixture 
in populations of the honeybee parasite Varroa destructor. In review

All scripts were run on the FASRC Cannon cluster. All directories are completed in order of their leading numerals. Similarly, bash scripts (e.g., 3_pca/2_make_bam.sh) are designed to run in ascending order of their numerals within directories. Additional subdirectories and large data files are excluded from this repository, but all can be reconstructed by running these scripts. R markdown scripts were run outside of the cluster on a comsumer laptop after bash scripts and often rely on their products.

Data

• varroa_data.csv includes sample collection and library preparation data for all samples included in the final analysis
• The varroa genome assembly is not included and must be downloaded as a fasta file to 0_ref
• Raw reads from this project are available in NCBI bioproject XXXXX (link) and must be downloaded to 1_raw_reads following the name convetion in varroa_data.csv

Directories

• 0_ref
  • reference genome for Varroa destructor, linked here
• 1_raw_reads
  • raw illumina reads from this project
• 2_ipyrad
  • scripts for assembling raw reads using ipyrad
  • params-P*.txt parameter files contain barcode information for demultiplexing samples from each pool
  • params-varroa.txt combines all pools for remaining ipyrad steps
• 3_pca
  • bash scripts assemle and deduplicate reads, generate liklihoods via angsd, filter out sites not within Hardy-Weinberg equilibrium, and calculate a PCA
    • varroa.pca.cov is the final output, included here for reference
  • PCA_view.Rmd visualizes the PCA and combines it with map information (8_mapping) to produce figure 1
    • this means scripts in 8_mapping must be run before figure 1 can be generated
  • pop_view.rmd simply counts samples per populaiton, and has no downstream products
• 4_gendist
  • bash script calculates pairwise genetic distance between samples
    • varroa.dist is the output, included here for reference
  • gendist.Rmd calculates pairwise geographic distance between samples, and combines with the product of the above script to generate the Isolation by Distance test. It also computes mantel tests for significance.
• 5_fastsimcoal
  • bash scripts calculate 2D site frequency spectra for populations, create a filestructure for fastsimcoal results, run simulations, and finally compile simulations that resulted in maximum liklihood outcomes
  • template.est and template.tpl are example parameter files which each need to be provided for each population in the (respective population subdirectory), completed with population sample sizes (as in subdir pops_bystate)
• 7_diversity
  • scripts within calculate various summary statistics for genetic diversity
  • global and by-population statistics are calculated seperately in their respective directories, following similar workflows
  • statistics are calculated within 10000bp windows, then averaged across populations and globally in thetas.Rmd
  • heterozygosity is calculated per individual across all sites (pops/3_het_est.sh) and across only sites passing the Hardy-Weinberg filter created during PCA (het_hwe/0_het_est_hwe.sh). These are averaged across populations and globally in estimate_heterozygosity.Rmd
• 8_mapping
  • script for producing map in figure 1a
  • map subdirectory contains (large) data file not included in this repository. This can be downloaded locally in commented-out section of mapping.Rmd
• 9_tables
  • sample_stats.Rmd breaks samples down by extractions, sequencing, inclusion in final analysis, and population
  • *.csv files generated in 7_diversity and sample_stats.Rmd are (manually) compiled into tables.xlsx, which stores data for all tables in manuscript