Overview

This repository contains the code used for the population genetic analyses of Quercus tardifolia and other Texas red oaks. The goals of these analyses are:

1. determining how genetically distinctive Q. tardifolia (QUTA) is
2. ff QUTA is found to be a hybrid, determine the hybrid class of QUTA individuals
3. understanding the taxonomic entities which exist in this group

The genetic data for these analyses came from the Hipp Lab at the Morton Arboretum. The code in this repository is exploratory and in development.

Repository layout

There are four primary folders in this repository (ordered alphabetically, in approximate processing order):

1. Genotyping contains the BASH script and input files used to perform demultiplexing (when necessary), QC, reference alignment, and genotyping. There's also one R script, which was used to assess the level of polymorphic error between replicate samples. Barcode files used during multiplexing are included in the 1_Demux-QC folder.

2. Clustering contains the scripts used to run STRUCTURE, build neighbor-joining trees, and perform discriminant analysis of principal components (DAPC) and principal components analysis (PCA). This folder is divided into subfolders based on STRUCTURE clustering runs (prelim, Clust2, or Clust3; Clust3 includes the most relevant results to the manuscript), as well as a folder for DAPC outputs. Note that the R scripts used for visualizing the STRUCTURE results are contained in this folder.

3. hybridAnalyses contains the scripts used to validate the hybridization patterns found during the clustering steps. There are two subfolders: one containing the scripts used to run the species networks applying quartets (SNaQ) software, and one containing the R script and output of the F-stats analysis (which presently isn't included in the main results for this project.) There is also an R script used to run the nQuire software (for assessing polyploidy) and for conducting heterozygosity analyses (e.g. using triangulaR). Note that the GravHypoTar_HzAnalysis.R script includes the code for the triangulaR plots.

4. Mapping contains two R scripts.

   1. QUTA_sampleMapping.R was used to generate an HTML map of all the samples; this map was largely used internally.
   2. The GravHypoTar_RangeMapping.R script was used to build a range map of Q. gravesii and Q. hypoleucoides, and show sampled individuals on that map.