Example Data

There are two ways to download the input data files:

• navigate to where you want the data, then clone the repository that contains both this tutorial and the files:

  git clone https://github.com/crsl4/PhyloNetworks.jl.wiki.git

  The repository will appear as a directory named PhyloNetworks.jl.wiki. Inside, you should find a directory data_results:

  cd PhyloNetworks.jl.wiki
  ls
  cd data_results
  ls
  

• or download and uncompress a tarball: .zip or .tgz

These tutorial files are organized in a directory data_results/ that has:

• 6 example data sets, with a separate directory for each. These are all simulated data, for which we know the true network.

  • 1 data set has 15 taxa (n15.gamma0.30.20.2_n300).
  • 5 data sets have 6 taxa (baseline.* and n6.gamma0.30.2_n300), to run fast during the tutorial. Among those, 4 (baseline.*) were generated under a network with 1 reticulation and different numbers of genes (from 30 to 1000 genes), and 1 data set was generated under a network with 2 reticulations (n6.gamma0.30.2_n300).

  These examples will provide ways to look at the effect of the number of taxa, the number of genes, and the number of reticulations.

• a folder scripts/ containing all the scripts we will use in the tutorial.

Within each dataset folder, you will find several directories:

• input/: contains input data. One file has the true gene trees (simulated from a network) and the other file is a tarball containing all the alignments (e.g. 1000 alignments if the data set has 1000 genes). All the other folders and files can be recreated using the scripts. The main results files are provided, though, to allow participants to pick up the tutorial at any step.
• bucky-output/: contains the main output file from running MrBayes + BUCKy, which is a table listing quartet concordance factors (CFs). Also contains a file with the species estimated from quartet CFs using Quartet MaxCut.
• raxml/: contains one file with all the best RAxML trees, one for each gene; and a directory bootstrap/ containing one file with 100 bootstrap for each gene.
• astral/: contains one file that lists all the bootstrap tree files from RAxML, and another file with the results of ASTRAL. These results consists of 102 trees: first 100 bootstrap species trees, then their consensus, then the species tree estimated from the original data annotated with bootstrap support.
• snaq/: contains various files for the various estimated networks: best networks and bootstrap networks.

For 4 of the 6 data sets, the main results of each step are provided. Participants can continue on the next step even if the previous step did not work on their laptop.