Data Import and Export (Python API)

1. Importing a network from a Table

If you have an interaction matrix stored in a tsv or csv format, you can load it into a Network object. Additionally we require:

  1. obsmeta_file - A csv file containing the taxonomy information for the microbes that are a part of the matrix

  2. meta_file - A json file containing the experimental metadata for the network

  3. cmeta_file - A json file containing the details about the computational processing done on the data

from micone import Network, NetworkGroup
network = Network.load_data(
    interaction_file,
    meta_file,
    cmeta_file,
    obsmeta_file,
)
network_group = NetworkGroup([network])

Additionally, you can specify the matrix of pvalues and thesholds if any to be applied to the data

Note

  1. The indices of the interaction_file and obsmeta_file must match

  2. The order of indicies of interaction_file and pvalue_file must match

2. Importing a network from an edge list

If you have interaction data stored in the form of an edge list

from micone import Network. NetworkGroup
network = Network.load_elist(
    elist_file,
    meta_file,
    cmeta_file,
    obsmeta_file,
)
network_group = NetworkGroup([network])

Note

  1. The headers of the elist_file must contain “source”, “target”, “weight”

3. Exporting your Network object into JSON file

network_group.write("network.json", pvalue_filter=True, interaction_filter=False)

This JSON file is compatible with the MIND database and can be uploaded and visualized directly on the website.