Command Line Interface¶
DiffuPy Command Line Interface
diffupy¶
DiffuPy
diffupy [OPTIONS] COMMAND [ARGS]...
diffuse¶
Run a diffusion method for the provided input_scores over a given network.
- param input
Path to a (miscellaneous format) data input to be processed/formatted.
- param network
Path to the network as a (NetworkX) graph or as a (diffuPy.Matrix) kernel.
- param output
Path (with file name) for the generated scores output file. By default ‘$OUTPUT/diffusion_scores.csv’
- param method
Elected method [“raw”, “ml”, “gm”, “ber_s”, “ber_p”, “mc”, “z”] or custom method FUNCTION(network, scores, kargs). By default ‘raw’
- param binarize
If logFC provided in dataset, convert logFC to binary. By default False
- param threshold
Codify node labels by applying a threshold to logFC in input. By default None
- param absolute_value
Codify node labels by applying threshold to | logFC | in input. By default False
- param p_value
Statistical significance. By default 0.05
- param format_output
Elected output format [“CSV”, “JSON”]. By default ‘CSV’
- param kernel_method
Callable method for kernel computation.
diffupy diffuse [OPTIONS]
Options
- -i, --input <input>¶
Required Input data
- -n, --network <network>¶
Required Path to the network graph or kernel
- -o, --output <output>¶
Output file
- -m, --method <method>¶
Diffusion method
- Options
z | gm | ml | ber_p | raw | mc | ber_s
- -b, --binarize <binarize>¶
If logFC provided in dataset, convert logFC to binary (e.g., up-regulated entities to 1, down-regulated to -1). For scoring methods that accept quantitative values (i.e., raw & z), node labels can also be codified with LogFC (in this case, set binarize==False).
- -t, --threshold <threshold>¶
Codify node labels by applying a threshold to logFC in input.
- -a, --absolute_value <absolute_value>¶
Codify node labels by applying threshold to | logFC | in input. If absolute_value is set to False,node labels will be signed.
- -p, --p_value <p_value>¶
Statistical significance (p-value).
- Default
0.05
- -f, --format_output <format_output>¶
Choose CSV or JSON output scores file format.
- Default
csv
kernel¶
Generate a kernel for a given network.
- param network
Path to the network as a (NetworkX) graph to be transformed to kernel.
- param output
Path (with file name) for the generated scores output file. By default ‘$OUTPUT/diffusion_scores.csv’
- param log
Logging profiling option.
diffupy kernel [OPTIONS]
Options
- -g, --graph <graph>¶
Required Input network
- -o, --output <output>¶
Output path to store the generated kernel pickle
- Default
/home/docs/.diffupy/output/kernel.json
- -l, --log¶
Activate debug mode