This archive includes all of the non-trivial scripts used to process data in “Assessing the utility of transcriptome data for inferring phylogenetic relationships among coleoid cephalopods” by Lindgren and Anderson.

ceph_hamstr_outtofastas.sh - Will make fasta files for each ortholog group from the HaMStR output. This was written by Kevin Horn and modified by FEA.

ceph_groups_to_pruned.sh - Will align each ortholog group fasta produced with ceph_hamstr_outtofastas.sh, infer a likelihood tree using FastTree and prune those trees using PhyloTreePruner. This script is a modification of Kevin Kocot's PhyloTreePruner wrapper script available at https://sourceforge.net/projects/phylotreepruner/files/wrapper_scripts/

ceph_pruned_to_bestmodel.sh - Was used after ceph_groups_to_pruned.sh and will back up the pruned fasta files, then convert them to phylip format and run ProteinModelSelect.pl on each. It will output a file containing the name of each ortholog group, the best model, and the alignment length. It too is a modification of Kevin Kocot's PhyloTreePruner wrapper script.

TSEBA.sh - This is a clunky script written by FEA that will take a set of fasta files and a matching set of single-OG RAxML bipartition files and sequentially process them through TreSpEx.v1.1.FEA.pl (a very slightly modified version of TreSpEx.v1.1.pl by Torsten Struck and available at http://annelida.de/onewebmedia/TreSpEx_v1_1.tgz) and BaCoCa.v1.104.r.pl (by Kück and Struck and available at http://annelida.de/onewebmedia/BaCoCa_v1.1.r.zip) as described in the manuscript. It will produce various files, but the most important product is a folder called “TSE-and-BaCoCa-filtered-concats”. This script requires the TreSpEx_LBH_filter.R and BaCoCa_filter.R R scripts (see below), as well as TreSpEx.v1.1.FEA.pl (included in this archive) and BaCoCa.v1.104.r.pl.

TreSpEx_LBH_filter.R - R script that reads output from TreSpEx.v1.1.FEA.pl and removes OGs that return a value equal to or greater than 1.5 times the interquartile range above the median for three of the indices calculated by TreSpEx: the standard deviation of the tip-to-root distance, the average patristic distance and the LB score; see manuscript for details. Written by Kevin Horn.

BaCoCa_filter.R - R script that reads output from BaCoCa.v1.1.r.pl and removes OGs with a p value of less than 0.05 in a chi square test of homogeneity and OGs that had an RCFV value that was 1.5 or more times the interquartile range above the median RCFV value for all OGs. Written by Kevin Horn.