Compass User Guide¶
Welcome to the user guide for Compass, if you haven’t already installed Compass visit the install guide.
In this page
Further resources:
Quick Start¶
Once Compass has been installed you can run it the algorithm on an input gene expression matrix by opening a command line in the directory with the input file and entering:
compass --data expression.tsv --num-processes 10
If your gene expression matrix is in a matrix market format (.mtx file) you can use the --data-mtx option. You’ll also need to provide a tab-delimited file with labels for what genes are in each row and optionally add in a sample names file:
compass --data-mtx expression.mtx genes.tsv sample_names.tsv --num-processes 10
Once Compass has finished running, it will output a matrix reactions.tsv with reaction penalties where higher scores correspond to a reaction being less likely.
Next steps¶
For a more detailed example of running Compass, visit the tutorial.
If you want to see all the settings you can customize visit here.
More details on the micropooling setting (useful for 10x-scale datasets) can be found here here.
For tools to postprocess and analyze the results of Compass, visit here
Cloud computing like AWS can help Compass scale because each cell can be handled in parallel, so we provide a guide for running Compass on AWS.