Compute degree-weighted path count similarity.

Use the degree-weighted path count metric to compute a meta-path based similarity score. Node that, in this implementation, type-specific degrees are used (except for the last step of the meta-path).

get_dwpc(
  x,
  y,
  paths_x,
  paths_y = NULL,
  reference_list,
  list_type = c("edge", "neighbor"),
  verbose = TRUE,
  w = 0.4
)

Arguments

x

ID of the origin node.

y

ID of the destination node.

paths_x

Paths from the origin node following the meta-path of interest as a data.table.

paths_y

Paths from the destination node following the meta-path of interest as a data.table.

reference_list

Either an edge list as a data.table which must contain the columns Origin, Destination, OriginType, DestinationType, and EdgeType, or a neighbor reference object constructed by get_neighbor_list().

list_type

If an edge list is provided, specify "edge". If a neighbor list is provided, specify "neighbor".

verbose

Should the intermediate calculations be printed to the console?

Value

A list with two elements:

Metric

The name of the similarity metric (i.e., "Degree-Weighted Path Count").

Similarity

The degree-weighted path count similarity score.

References

Himmelstein, D. S. & Baranzini, S. E. Heterogeneous Network Edge Prediction: A Data Integration Approach to Prioritize Disease-Associated Genes. PLOS Computational Biology 11, e1004259 (2015).

See also

get_neighbor_list() for neighbor reference object construction.

Other similarity: get_npc(), get_pathsim(), get_pc()