Important note

The code in this package is intended to specifically run the analysis as it appears in Eng, Aeschlimann, van Veenendaal et al., PLOS Medicine 2019. We will be releasing a generalized version of our multilayer non-negative matrix factorization (NMF) workflow with a subsequent paper.

Data access

For data access, please contact Dr. Susanne Benseler, chair of the Canadian Association of Paediatric Rheumatology (CAPRI), who will forward your request to the CAPRI Scientific Protocol Evaluation Committee/Data Access Committee.

Installation

Python

This analysis requires Python 3.6 or higher.

We recommend managing your Python installation using conda, especially as we provide an environment.yml file for your convenience.

R

Please install R 3.5.0+ with the following packages:

• car
• ggbeeswarm
• lmtest
• moments

Circos

To generate the wheel figure, Circos 0.69+ is needed. The most straightforward way to do so is to install coreutils through Homebrew and then issuing the command

brew install circos

coreutils (macOS only)

On macOS, coretools must be installed (due to the gln command, required for properly linking inputs and outputs). The most straightforward way to do so is to install coreutils through Homebrew and then issuing the command

brew install coreutils

Running the analysis

To preview what rules will be run, type

snakemake --dry-run everything

To run the analysis, type

snakemake everything

If you wish to run the analysis on a cluster, type

snakemake --cmd '<command>' everything

where <command> is the command you would use to submit a job (e.g., qsub).

To clean all output files, type

snakemake --delete-all-output