Repository files navigation

TheOhmNet algorithm learns feature representationsfor nodes in any (un)directed, (un)weighted multi-layer network. Pleasecheck theproject page for more details.

To runOhmNet onhuman brain multi-layer network with nine layers,run the following command from the project home directory:

python2.7 main.py --input "data/brain.list" --outdir "tmp" --hierarchy "data/brain.hierarchy"

To checkOhmNet's running options, use:

python2.7 main.py --help

Results are saved to output directory specified by theout_diroption.

The output fileleaf_vectors.emb contains feature representationsfor nodes at the level of leaves in the hierarchy (i.e., leaves in thehierarchy correspond exactly to network layers).

The first line has the following format:

total_num_of_nodes_in_layers dim_of_representation

The nexttotal_num_of_nodes_in_layers lines are as follows:

node_id dim1 dim2 ... dimd

where node_id is formatted as network_layer_name__node_id, and dim1, ... , dimd isthed-dimensional representation learned byOhmNet.

The output fileinternal_vectors.emb contains feature representationsfor nodes at higher levels in the hierarchy (i.e., internal levels in thehierarchy contain feature representations at intermediate/higher scales).

The first line has the following format:

total_num_of_nodes_in_hierarchy dim_of_representation

wheretotal_num_of_nodes_in_hierarchy is equal to (size_hierarchy -num_layers) * total_num_nodes.

The nexttotal_num_of_nodes_in_hierarchy lines are as follows:

node_id dim1 dim2 ... dimd

where node_id is formatted as hierarchy_element_name__node_id, and dim1, ... , dimd isthed-dimensional representation learned byOhmNet.

If you findOhmNet useful for your research, please consider citingthe followingpaper presented at ISMB/ECCB 2017:

@article{Zitnik2017,  title     = {Predicting multicellular function through multi-layer tissue networks},  author    = {Zitnik, Marinka and Leskovec, Jure},  journal   = {Bioinformatics},  volume    = {33},  number    = {14},  pages     = {190-198},  year      = {2017},  publisher = {Oxford Journals}}

Please send any questions you might have about the code and/or thealgorithm tomarinka@cs.stanford.edu.

Note: This is a full Python implementation ofOhmNetalgorithm. A C++ implementation will be released as part of SNAP software.

OhmNet is tested to work under Python 2.7.

The required dependencies for OhmNet areNumPy >= 1.12, andNetworkX >= 1.11.

OhmNet is licensed under the MIT License.