API: ipyrad analysis tools

The ipyrad-analysis toolkit is a Python interface for taking the output files produced in a ipyrad assembly and running a suite of evolutionary analysis tools with convenient features for filtering for missing data, grouping individuals into populations, dropping samples, and more.

All of these tools share a common syntax making them easy to use without having to worry about creating different input files, or learn new file formats. They are designed for use within Jupyter notebooks, a tool for reproducible science. See the examples below.

# the analysis tools are a subpackage of ipyrad
import ipyrad as ipa

# a large suite of tools are available
tool = ipa.structure(data="./outfiles/data.snps.hdf5")

# all tools share a common syntax for setting params
# and distributing work in parallel.
tool.run()

• ipa.vcf_to_hdf5
  • Required software
  • Pre-filter data from other programs (e.g., FreeBayes, GATK)
  • A peek at the cleaned VCF file
  • Converting clean VCF to HDF5
  • Downstream analyses
  • Run a single PCA analysis from subsampled unlinked SNPs
  • Run multiple PCAs over replicates of subsampled SNPs
• ipa.treemix
  • Required software
  • Required input data files
  • Short Tutorial:
  • 1. Finding the best value for m
  • 2. Iterate over different subsamples of SNPs
  • 3. Save the plot to pdf
• ipa.pca
  • Required software
  • Required input data files
  • Input data file and population assignments
  • Enter data file and params
  • Run PCA
  • Run PCA and plot results.
  • Subsampling SNPs
  • Subsampling with replication
  • Advanced: Imputation algorithms:
    • No imputation (None)
    • No imputation but stricter filtering (mincov)
    • Kmeans imputation (integer)
  • Save plot to PDF
  • Advanced: Missing data per sample
  • Advanced: TSNE and other dimensionality reduction methods
  • Advanced: UMAP dimensionality reduction
• ipa.raxml
  • Required software
  • Short Tutorial:
  • Draw the inferred tree
  • Setting parameters
  • Cookbook
  • Generating new phylip alignments
  • What’s next?
• ipa.mrbayes
  • Required software
  • Setting mb param settings
  • Run mrbayes
  • Accessing results
  • Check the parameter values and convergence
  • Plotting mb consesus tree
  • Plotting mb posterior distribution of trees
  • Save figures
  • What else?
• ipa.tetrad
  • Required software
  • Input data
  • Initialize the analysis object
  • Call run
  • Show the full tree with bootstrap supports
  • Show the majority-rule consensus tree with bootstrap supports
  • Show variation over the bootstrap replicates
  • Continuing from a checkpoint
• ipyrad-analysis toolkit: STRUCTURE
  • Required software
  • Required input data files
  • Note: missing data in STRUCTURE analyses:
    • Approximate run times
    • Input data file and population assignments
    • Enter data file and params
    • Run STRUCTURE and plot results.
  • Analyze results: Choosing K
  • Analyze results: Barplots
• Cookbook
• ipa.sratools
  • Required software
  • Fetch info for a published data set by its accession ID
  • File names
  • Download the data
  • Check the data files
• ipa.baba
  • Load packages
  • Set up and connect to the ipyparallel cluster
  • A tree-based hypothesis
  • Load in your .loci data file and a tree hypothesis
  • Short tutorial: calculating abba-baba statistics
  • Look at the results
  • Plotting and interpreting results
  • generating tests
    • Further investigating results with 5-part tests
      • Full Tutorial
  • Creating a baba object
  • Linking tests to the baba object
  • Other parameters
  • Running the tests
  • The results table
  • Auto-generating tests
  • Running the tests
  • More about input file paths (i/o)
  • (optional): root the tree
  • Interpreting results
  • Running 5-taxon (partitioned) D-statistics
• ipa.bucky
  • Software requirements for this notebook
  • Cluster setup
  • Create a bucky analysis object
  • Write data to nexus files
  • An example nexus file
    • Complete a BUCKy analysis
  • Alternatively, you can run each step separately
  • Convenient access to results
• ipa.window_extacter
  • Required software
  • Required input data files
  • The scaffold table
  • Selecting scaffolds
  • Subsetting scaffold windows
  • Filtering missing data with mincov
  • Filtering missing data with imap and minmap
  • Subsample taxa with imap
  • Concatenate multiple scaffolds together
  • Consensus reduction with imap
  • Write selected window to a file
  • Accessing the output files