cellmaps_ppi_embedding package

Submodules

cellmaps_ppi_embedding.cellmaps_ppi_embeddingcmd module

cellmaps_ppi_embedding.cellmaps_ppi_embeddingcmd.main(args)

Main entry point for program

Parameters:

args (list) – arguments passed to command line usually sys.argv[1:]()

Returns:

return value of cellmaps_ppi_embedding.runner.CellMapsPPIEmbedder.run() or 2 if an exception is raised

Return type:

int

cellmaps_ppi_embedding.exceptions module

exception cellmaps_ppi_embedding.exceptions.CellMapsPPIEmbeddingError

Bases: Exception

Base exception for CellMapsPPIEmbedding

cellmaps_ppi_embedding.runner module

class cellmaps_ppi_embedding.runner.CellMapsPPIEmbedder(outdir=None, embedding_generator=None, inputdir=None, skip_logging=True, name=None, organization_name=None, project_name=None, provenance_utils=<cellmaps_utils.provenance.ProvenanceUtil object>, input_data_dict=None, provenance=None)

Bases: object

Class to run algorithm

Constructor

Parameters:

• outdir (str) – directory where ppi embeddings will be saved

• embedding_generator (EmbeddingGenerator) – Object responsible for generating the embeddings. Must implement get_next_embedding(), typically an instance of a subclass of EmbeddingGenerator, such as Node2VecEmbeddingGenerator or FakeEmbeddingGenerator.

• inputdir (str or None) – Input directory that contains ppi edgelist file and its RO-Crate metadata file.

• skip_logging (bool) – If True skip logging, if None or False do NOT skip logging

• name (str or None) – Optional display name for the dataset. If not provided, the name will be inferred from the RO-Crate metadata or provenance dictionary.

• organization_name (str or None) – Optional name of the organization generating the dataset. Used in provenance tracking. Falls back to RO-Crate or provenance input if missing.

• project_name (str or None) – Optional name of the project associated with the dataset. Used in provenance tracking. Falls back to RO-Crate or provenance input if missing.

• provenance_utils (ProvenanceUtil) – Utility class used for RO-Crate generation and FAIRSCAPE dataset, software, and computation registration. Defaults to a new ProvenanceUtil.

• input_data_dict (dict or None) –

  Dictionary of parameters and their values that capture the configuration used to generate the embeddings. This is serialized in the task metadata for reproducibility. If not provided, one is auto-generated from available parameters.

  Example:

  {'outdir': '/output/path', 'inputdir': '/input/path'}
  

• provenance (dict or None) –

  Optional dictionary specifying provenance metadata. Required if inputdir does not contain an RO-Crate. Used to describe the input edgelist, dataset authorship, and context.

  Example:

  {
      'name': 'Example PPI Dataset',
      'organization-name': 'CM4AI',
      'project-name': 'Network Embedding',
      'description': 'Node2Vec embeddings of protein-protein interactions',
      'keywords': ['PPI', 'embedding', 'node2vec'],
      'edgelist': {
          'name': 'PPI Edgelist File',
          'author': 'Krogan Lab',
          'version': '1.0',
          'data-format': 'tsv'
      }
  }
  

PPI_EDGELIST_FILEKEY = 'edgelist'

generate_readme()

static get_apms_edgelist_file(input_dir=None, edgelist_filename='ppi_edgelist.tsv')

Parameters:

• edgelist_filename

• input_dir

Returns:

Returns:

get_ppi_embedding_file()

Gets PPI embedding file in output directory

Returns:

Return type:

str

run()

Run node2vec to create embeddings

Returns:

class cellmaps_ppi_embedding.runner.EmbeddingGenerator(dimensions=1024)

Bases: object

Base class for implementations that generate network embeddings

Constructor

DIMENSIONS = 1024

get_dimensions()

Gets number of dimensions this embedding will generate

Returns:

number of dimensions aka vector length

Return type:

int

get_next_embedding()

Generator method for getting next embedding. Caller should implement with yield operator

Raises:

NotImplementedError: Subclasses should implement this

Returns:

Embedding

Return type:

list

class cellmaps_ppi_embedding.runner.FakeEmbeddingGenerator(ppi_downloaddir, dimensions=1024)

Bases: EmbeddingGenerator

Fakes PPI embedding

Constructor

Parameters:

dimensions (int) – Desired size of output embedding

get_next_embedding()

Generator method for getting next embedding. Caller should implement with yield operator

Raises:

NotImplementedError: Subclasses should implement this

Returns:

Embedding

Return type:

list

class cellmaps_ppi_embedding.runner.Node2VecEmbeddingGenerator(nx_network, p=2, q=1, dimensions=1024, walk_length=80, num_walks=10, workers=8)

Bases: EmbeddingGenerator

Constructor

NUM_WALKS = 10

P_DEFAULT = 2

Q_DEFAULT = 1

WALK_LENGTH = 80

WORKERS = 8

get_next_embedding()

Returns:

Module contents

Top-level package for cellmaps_ppi_embedding.