Addition and Improvements
- New analysis/visualization programs:
- LDA.py (Uses Linear Discriminant Analysis to visualize group-wise compositional differences)
- diversity.py (Visualize alpha diversity as KDE-smoothed histograms)
- New API module
- graph_util (Adds convenience methods for data visualization/graphing)
- biom_calc module:
- Switched to using the official biom-format python library for parsing and handline BIOM files
- Added arcsine square root transformation to the API and added as an option to several scripts:
- iTol.py
- LDA.py
- PCoA.py
- Support for the new UniFrac format in QIIME 1.9
- Adds the following command-line options:
- Set font size
- Set distance between axes and their labels
- Set the azimuth and elevation for 3D plots
- Set the graph styling to look like ggplot2 defaults
- Add Sample ID annotations to graphed points
- Replaces the --dpi option with --fig_size to set the figure size directly
- LDA.py: Added the option to save the LDA-transformed data out as a csv file.
- diversity.py
- Added group coloring by specify a column in the mapping file
- Support for multiple diversity metrics
- Significance testing is now performed to compare alpha diversity between groups
- Wilcoxon Signed Rank Test for two-group comparison
- Kruskal-Wallis H-test for comparing three or more groups
- biom_relative_abundance.py: Support for BIOM-format input.
- biom_calc.py: Updated raw_abundance() function to handle BIOM format version 2.x input files.
- iTol.py: Updated BIOM file handling for iTol script (v2.x files).
- PhyloToAST documentation moved from separate repository into the 'docs' folder in the main code repository
- Overhauled the main page for the project documentation
- Added step-by-step instructions to the documentation for many of the executable scripts
Bug Fixes
- Prevents duplicate entries from being passed on into the output from restrict_repset.py.
- Fixes the --save_calculations option in diversity.py. Now prints out all metrics in a single column with the second column being group membership.
- Adds pad_taxonomy method (otu_condense.py) to ensure all taxonomy strings have all levels down to species.
- Cleans up error message for line parsing in prune_taxonomy. Makes file error message more readable.