Compass Advanced Usage Tutorial

Compass allows users to customize various features when using compass.

usage: compass [-h] --data FILE [--model MODEL] [--species SPECIES]
               [--media MEDIA] [--output-dir DIR] [--temp-dir DIR]
               [--torque-queue QUEUE] [--num-processes N] [--lambda F]
               [--num-threads N] [--and-function FXN] [--num-neighbors N]
               [--symmetric-kernel] [--input-weights FILE]
               [--penalty-diffusion MODE] [--no-reactions] [--no-metabolites]

Below we describe the features in more detail:

• and-function: Which function used to aggregate AND associations(min, median, mean). Default min
• input-weights:
• lambda: Smoothing factor for single-cell data. Should be set between 0 and 1
• media:
• model: Metabolic model to use (RECON1_mat, RECON2_mat, RECON2.2). Default is RECON2_mat
• no-metabolites: Flag to disable calculation and output of uptake/secretion scores in addition to reaction scores.
• no-reactions: Flag to disable calculation and output of reaction scores in addition to uptake/secretion scores.
• num-neighbors: Either effective number of neighbors for gaussian penalty diffusion or exact number of neighbors for KNN penalty diffusion. Default is 30
• num-processes: Number of processes for Compass to use. Ignored when submitting job onto a queue
• num-threads: Number of threads to use per sample. Default is 1
• output-dir: Final directory to output concatenated reactions.txt file
• penalty-diffusion: Mode to use to share reaction penalty values between single cells (gaussian, knn). Default gaussian
• species: Species to use to match genes to model (homo_sapiens, mus_musculus). Default is homo_sapiens
• symmetric-kernel:
• temp-dir: Directory to store partial results for completed samples in a dataset
• torque-queue: Name of the torque queue to submit to