In cds-group/GraphDistances: Distribution based distance computation between graphs

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)

GraphDistances

Compute distribution based distance measures between directed and undirected graphs/networks.

Usage

Load package

library(GraphDistances)

Distance computation: Pairwise

Distance computation between the NDD (node distance distribution) and Transition Probability (TM-one walk) matrices of two example graphs.

library(igraph)
num_nodes <- 60
g1 <- make_tree(num_nodes, children =2, mode = "out")
E(g1)$weight <- seq_len(ecount(g1))
g2 <- make_tree(num_nodes, children = 3, mode = "out")
E(g2)$weight <- seq_len(ecount(g2))

# Compute the distances between the node distance distribution of 2 graphs
binList <- getBins(list(g1, g2))
ndd1 <- getNodeDistanceDistr(g1, binList)
ndd2 <- getNodeDistanceDistr(g2, binList)
nddDistancepair <- getGraphpairdistance(ndd1, ndd2)

# Compute the distances between the transition probability matrix (one walk) of 2 graphs
# For n walk matrix change the 'walk' parameter to n
trans1 <- getTransitionmatrix(g1, walk=1)
trans2 <- getTransitionmatrix(g2, walk=1)
transDistancepair <- getGraphpairdistance(trans1, trans2) 

Distance computation: Multiple graphs

nddDistanceMulti and transDistanceMulti are the Gram matrices produced after distance computation between the NDD and TM-one walk matrices of 40 graphs provided with the package. The annotation of these graphs can be loaded by running data(annoKidney) in R for classification, clustering and visualization tasks.

data("KidneyGraphs")

binList <- getBins(KidneyGraphs)
nddList <- lapply(KidneyGraphs, function(x) getNodeDistanceDistr(x, binList))
nddDistanceMulti <- getGraphlistdistance(nddList) #Takes sometime for computation

transList <- lapply(KidneyGraphs, function(x) getTransitionmatrix(x, walk=1))
transDistanceMulti <- getGraphlistdistance(transList) #Takes sometime for computation

Distance computation: Multiple graphs: Parallel

getGraphdistance4PartsParallel can be used to run the distance computation in parallel.

library(future.apply)
library(future)
plan(multiprocess)

data("KidneyGraphs")
binList <- getBins(KidneyGraphs)
nddListPar <- future_lapply(KidneyGraphs, function(x) getNodeDistanceDistr(x, binList))
nddDistanceMultiPar <- getGraphdistance4PartsParallel(nddListPar)

transListPar <- future_lapply(KidneyGraphs, function(x) getTransitionmatrix(x, walk=1))
transDistanceMultiPar <- getGraphdistance4PartsParallel(transListPar)

Cite

1. Granata, I., Guarracino, M.R., Kalyagin, V.A., Maddalena, L., Manipur, I. and Pardalos, P.M., 2018, December. Supervised classification of metabolic networks. In 2018 IEEE International Conference on Bioinformatics and Biomedicine (BIBM) (pp. 2688-2693). IEEE. https://ieeexplore.ieee.org/abstract/document/8621500
2. Granata, I., Guarracino, M.R., Kalyagin, V.A., Maddalena, L., Manipur, I. and Pardalos, P.M., 2020. Model simplification for supervised classification of metabolic networks. Annals of Mathematics and Artificial Intelligence, 88(1), pp.91-104. https://link.springer.com/article/10.1007/s10472-019-09640-y

cds-group/GraphDistances documentation built on Dec. 14, 2020, 8:26 a.m.