README.md

In qpmnguyen/microbe_set_trait: Reproducible analyses for evaluating trait-based taxon sets

Reproducible analyses for untitled manuscript on trait-based taxon sets.

Quang Nguyen

Dependency management

All R analyses were performed under R version 4.1.2 and the corresponding Bioconductor version 3.14. Specific package versions are managed via renv and details can be found in the corresponding renv.lock file.

All python analyses are performed under Python version 3.8.9. Specific package versions are managed via Miniconda. The specifications for the primary virtual environment is in the microbe_trait_env.yml file. Since we also use QIIME2 and picrust2, separate conda environments were utilized for each software respectively due to dependency conflicts. Specifications for the QIIME2 2022.2 environment can be found in the qiime2-2022.2-py38-linux-conda.yml file (for different OS, please refer to the main QIIME2 documentation for instructions to install version 2022.2 in your machine).

Analyses

All analyses are in Jupyter Notebooks located in the analysis folder. - Placeholder

Pipelines outside of jupyter notebooks

Most analyses are located within reproducible jupyter notebooks. However, there are some computations that require more resources and therefore are ran as stand-alone scripts or as part of more reproducible workflows.

R

Intensive computing steps in R were performed using the targets package. The entire pipeline was broken down into smaller "steps" for ease of processing. The file run.R contains code to choose between different pipelines to run.

Individual mini-pipelines include: - Placeholder

Function files

R

• metaphlan_db.R defines functions to process the MetaPhlAn3 marker information files.
• db_preprocess.R defines functions to pre-process the database into BiocSet format.
• hmp_preprocess.R defines functions to pre-process the HMP data data files (newest release).

Python

• Placeholder

Additional notes on reproducibility:

In order to conveniently access the MetaCyc pathway class hierarchy, the package pythoncyc is used. Since this package is not distributed via pip or conda, please create a new conda or virtual environment, and then use the manual installation instructions when your desired conda environment is activated. Due to the perks of the pythoncyc library, users need to also install the associated PathwayTools program (v. 25.1) and run it with the python API open (see instructions in the link above). This step doesn't need to be repeated (hence not included in the major pipelines) and the product of this step is saved as a binary file called metacyc_parse.rds in the databases folder.

qpmnguyen/microbe_set_trait documentation built on April 11, 2024, 6:07 p.m.