metaQuant

Quantitative analysis of the function and taxonomy of microbiomes and their interaction.

Setup

The newest version of metaquant should be downloaded from this site. The dependencies are most easily satisfied with conda, and the environment can be created as follows:

conda create -n metaquant python=3.5 pandas ete3 goatools wget numpy statsmodels biopython

Note that the bioconda and conda forge channels must be enabled, as described on the bioconda website.

Tests

To run unittests for the project, run the following from the root directory:

python -m unittest discover tests

Roadmap

High Priority

• visualizations
• documentation
• better arg checking
  • deal with sample that is completely missing values
  • deal with missing values better
  • taxonomy rank checking
  • move ncbi checking to IO
  • check that supplied columns are in the dataframe
  • check that sample info provides a list, or is coerced to a list
  • don't return above phylum
  • don't return BP, MF, or CC
  • just pass args from cli?
• configure Travis CI on Github
• add option for specific rank in TF
• move threshold to calculating mean values as well
• make COG more similar to other databases
  • make class
  • add is_in_db method

Lower Priority

• benchmarking and optimization
• use flake8 for codestyle
• refactoring
  • move every analysis function to one to cut down on duplication?
  • could implement databases as instantiations of abstract classes. Any advantages?

Done

• unified database structure and 'adding up'
  • unify this with classes
  • work on EC in particular
  • how do we implement for COG cats?
• add handling if description is not found in database (EC)