CCC on Mitya's Data

YP & & VL & DS & MT & NL & CEP & …
Department of Applied Mathematics and Statistics
Johns Hopkins University

Fri Mar 6 08:18:19 2015

Cleaning the data

#fname <- url("http://www.cis.jhu.edu/~parky/Data/drosophila_retina_1.graphml")
#g <- read.graph(fname, format="graphml")
fname <- "drosophila_retina.Rbin"
load(fname); g <- upgrade_graph(g)
summary(g)
# IGRAPH DN-- 1781 33641 -- 
# + attr: source (g/c), info (g/c), name (v/c), x (v/n), y (v/n), z
# | (v/n), M (v/n), pre.x (e/n), post.x (e/n), pre.y (e/n), post.y
# | (e/n), pre.z (e/n), post.z (e/n), Proofreading.Details (e/c)

plot of chunk plot-orig-g-deg

Fig 1.

g <- simplify(g)      # make g unweighted & hollow
g <- as.undirected(g) # make g undirected (symmetric)
# IGRAPH UN-- 1781 8911 -- 
# + attr: source (g/c), info (g/c), name (v/c), x (v/n), y (v/n), z
# | (v/n), M (v/n)

plot of chunk reorder

Fig 2.

## remove the argmax(deg)
maxdv <- which.max(degree(g))
g <- delete.vertices(g,maxdv)

## find the largest connected component (lcc)
cc <- clusters(g)
table(cc$memb)
# 
#    1    2    3    4    5    6    7    8    9   10   11   12   13   14   15 
# 1748    1    1    1    1    1    1    1    3    2    1    1    1    1    1 
#   16   17   18   19   20   21   22   23   24   25   26 
#    2    2    2    1    1    1    1    2    1    1    1
g <- induced.subgraph(g, which(cc$memb == which.max(cc$csize)))
A <- g[]
summary(g)
# IGRAPH UN-- 1748 7977 -- 
# + attr: source (g/c), info (g/c), name (v/c), x (v/n), y (v/n), z
# | (v/n), M (v/n)