chooseNodeStats: Specify Node-level Network Properties to Compute

Description

Create a vector specifying node-level network properties to compute. Intended for use with buildRepSeqNetwork() or addNodeNetworkStats.

node_stat_settings() is a deprecated equivalent of chooseNodeStats().

Usage

chooseNodeStats(
  degree = TRUE,
  cluster_id = FALSE,
  transitivity = TRUE,
  closeness = FALSE,
  centrality_by_closeness = FALSE,
  eigen_centrality = TRUE,
  centrality_by_eigen = TRUE,
  betweenness = TRUE,
  centrality_by_betweenness = TRUE,
  authority_score = TRUE,
  coreness = TRUE,
  page_rank = TRUE,
  all_stats = FALSE
)
exclusiveNodeStats(
  degree = FALSE,
  cluster_id = FALSE,
  transitivity = FALSE,
  closeness = FALSE,
  centrality_by_closeness = FALSE,
  eigen_centrality = FALSE,
  centrality_by_eigen = FALSE,
  betweenness = FALSE,
  centrality_by_betweenness = FALSE,
  authority_score = FALSE,
  coreness = FALSE,
  page_rank = FALSE
)

Value

A named logical vector with one entry for each of the function's arguments (except for all_stats). Each entry has the same name as the corresponding argument, and its value matches the argument's value.

Arguments

degree: Logical. Whether to compute network degree.
cluster_id: Logical. Whether to perform cluster analysis and record the cluster membership of each node. See addClusterMembership().
transitivity: Logical. Whether to compute node-level network transitivity using transitivity() with type = "local". The local transitivity of a node is the the number of triangles connected to the node relative to the number of triples centered on that node.
closeness: Logical. Whether to compute network closeness using closeness().
centrality_by_closeness: Logical. Whether to compute network centrality by closeness. The values are the entries of the res element of the list returned by centr_clo().
eigen_centrality: Logical. Whether to compute the eigenvector centrality scores of node network positions. The scores are the entries of the vector element of the list returned by eigen_centrality() with weights = NA. The centrality scores correspond to the values of the first eigenvector of the adjacency matrix for the cluster graph.
centrality_by_eigen: Logical. Whether to compute node-level network centrality scores based on eigenvector centrality scores. The scores are the entries of the vector element of the list returned by centr_eigen().
betweenness: Logical. Whether to compute network betweenness using betweenness().
centrality_by_betweenness: Logical. Whether to compute network centrality scores by betweenness. The scores are the entires of the res element of the list returned by centr_betw().
authority_score: Logical. Whether to compute the authority score using authority_score().
coreness: Logical. Whether to compute network coreness using coreness().
page_rank: Logical. Whether to compute page rank. The page rank values are the entries of the vector element of the list returned by page_rank().
all_stats: Logical. If TRUE, all other argument values are overridden and set to TRUE.

Author

Brian Neal (Brian.Neal@ucsf.edu)

Details

These functions return a vector that can be passed to the stats_to_include argument of addNodeStats() (or buildRepSeqNetwork(), if node_stats = TRUE) in order to specify which node-level network properties to compute.

chooseNodeStats and exclusiveNodeStats each have default argument values suited to a different use case, in order to reduce the number of argument values that must be set manually.

chooseNodeStats has most arguments TRUE by default. It is best suited for including a majority of the available properties. It can be called with all_stats = TRUE to set all values to TRUE.

exclusiveNodeStats has all of its arguments set to FALSE by default. It is best suited for including only a few properties.

References

Hai Yang, Jason Cham, Brian Neal, Zenghua Fan, Tao He and Li Zhang. (2023). NAIR: Network Analysis of Immune Repertoire. Frontiers in Immunology, vol. 14. doi: 10.3389/fimmu.2023.1181825

Webpage for the NAIR package

Examples

Run this code

set.seed(42)
toy_data <- simulateToyData()

net <- generateNetworkObjects(
  toy_data, "CloneSeq"
)

# Add default set of node properties
net <- addNodeStats(net)

# Modify default set of node properties
net <- addNodeStats(
  net,
  stats_to_include =
    chooseNodeStats(
      closeness = TRUE,
      page_rank = FALSE
    )
)

# Add only the spepcified node properties
net <- addNodeStats(
  net,
  stats_to_include =
    exclusiveNodeStats(
      degree = TRUE,
      transitivity = TRUE
    )
)

# Add all node-level network properties
net <- addNodeStats(
  net,
  stats_to_include = "all"
)

Run the code above in your browser using DataLab