testPackage.r
In lsaravia/EcoNetwork: Ecological network analyses including multiplex networks
# Test
require(igraph)
fileName <- c(system.file("extdata", package = "multiweb"))
fileName <- system.file("extdata", "BarentsBoreal_FW.csv", package = "multiweb")
g <- readNetwork(fileName)
V(g)$weigth <- runif(vcount(g))
plotTrophLevel(g,vertexLabel = FALSE,vertexSizeFactor = 3,modules = TRUE,maxTL=4,weights=NULL, frame.color="black")
tk <- plotTrophLevel(g,vertexLabel = FALSE,vertexSizeFactor = 3,modules = TRUE, tk=TRUE)
plotTrophLevel(g,vertexLabel = FALSE,vertexSizeFactor = 3,modules = TRUE, lMat=tk)
dn <- list.files("inst/extdata",pattern = "^.*\\.csv$")
g<- readNetwork(dn,"inst/extdata")
names(g)
E(g[[1]])[which_loop(g[[1]])]
plotTrophLevel(g[[1]])
require(igraph)
require(NetIndices)
g <- graph_from_literal( 1 -+ 4 -+ 7,2 -+ 5 -+7, 3-+6-+7, 7-+7, 4+-3, 2-+7, simplify = FALSE)
calc_topological_indices(g)
plotTrophLevel(g,vertexLabel = TRUE,vertexSizeFactor = 20,modules = TRUE,maxTL=4,weights=NULL)
plotTrophLevel(g,vertexLabel = TRUE,vertexSizeFactor = 20,modules = TRUE,maxTL=4,weights=NULL, bpal=viridisLite::viridis(11))
# Test tk and use of lMat
tk <- plotTrophLevel(g,vertexLabel = TRUE,vertexSizeFactor = 20,modules = TRUE, tk=TRUE,vertex.label.color="white")
tk [,2] <- seq(0,100, length.out = 7)
plotTrophLevel(g,vertexLabel = TRUE,vertexSizeFactor = 20,modules = TRUE, lMat=tk,maxTL=2)
# add weights
#
E(g)$weight <- c(2,2.5,0.1,0.2,1,5,0.1,2)
plotTrophLevel(g,vertexLabel = TRUE,vertexSizeFactor = 20,modules = TRUE,maxTL=4,weights=NA,edge.width=5)
plotTrophLevel(g,vertexLabel = TRUE,vertexSizeFactor = 20,modules = TRUE,maxTL=4,weights=NULL,edge.width=5)
# Test incoherence
#
calcIncoherence(g)
g <- graph_from_literal( 1 -+ 4 -+ 7,2 -+ 5 -+7, 3-+6-+7, 4+-5, simplify = FALSE)
calcTopologicalIndices(g)
plotTrophLevel(g,vertexLabel = TRUE,vertexSizeFactor = 20, modules = TRUE)
nullg <- generateERbasal(g,10)
calcIncoherence(nullg,ncores=0)
calcModularitySWnessZScore(g,nullg,0.1)
calc_swness_zscore(g,nullg,0.1,ncores = 0)
# test reading .csv files
#
dn <- list.files("~/Academicos/GitProjects/MarineFoodWebsSmallWorld/Data/FoodWebs",pattern = "^.*\\.csv$")
#
# BEWARE different files have different structure
#
nets <- readNetwork(dn[11],"~/Academicos/GitProjects/MarineFoodWebsSmallWorld/Data/FoodWebs")
nets <- readNetwork(dn[16],"~/Academicos/GitProjects/MarineFoodWebsSmallWorld/Data/FoodWebs")
nets <- readNetwork(dn[17],"~/Academicos/GitProjects/MarineFoodWebsSmallWorld/Data/FoodWebs")
nets <- readNetwork(dn[18],"~/Academicos/GitProjects/MarineFoodWebsSmallWorld/Data/FoodWebs")
nets <- readNetwork(dn[19],"~/Academicos/GitProjects/MarineFoodWebsSmallWorld/Data/FoodWebs")
nets <- readNetwork(dn[20],"~/Academicos/GitProjects/MarineFoodWebsSmallWorld/Data/FoodWebs")
nets <- readNetwork(dn[21],"~/Academicos/GitProjects/MarineFoodWebsSmallWorld/Data/FoodWebs")
V(nets)[[]]
nets <- readNetwork(dn[22],"~/Academicos/GitProjects/MarineFoodWebsSmallWorld/Data/FoodWebs")
nets <- readNetwork(dn[23],"~/Academicos/GitProjects/MarineFoodWebsSmallWorld/Data/FoodWebs")
nets <- readNetwork(dn,"~/Academicos/GitProjects/MarineFoodWebsSmallWorld/Data/FoodWebs")
nets$CaymanIs_FW <- readNetwork(dn[8],"~/Academicos/GitProjects/MarineFoodWebsSmallWorld/Data/FoodWebs",fhead = FALSE,skipColumn = 0)
nets$Jamaica_FW <- readNetwork(dn[15],"~/Academicos/GitProjects/MarineFoodWebsSmallWorld/Data/FoodWebs",fhead = FALSE,skipColumn = 0)
nets$Cuba_FW <- readNetwork(dn[11],"~/Academicos/GitProjects/MarineFoodWebsSmallWorld/Data/FoodWebs",fhead = FALSE,skipColumn = 0)
calcTopologicalIndices(nets)
calcIncoherence(nets)
#netData <- nets
#devtools::use_data(netData,overwrite = TRUE)
# test reading .dat files from Johnson 2017
#
dn <- list.files("~/Dropbox/Projects/NetworksAsUnifyingPrinciple/Data",pattern = "^.*\\.dat$")
nets <- readNetwork(dn,"~/Dropbox/Projects/NetworksAsUnifyingPrinciple/Data",fhead=FALSE)
calcTopologicalIndices(list(netData$CaymanIs_FW,nets$cayman_islands))
require(RColorBrewer)
plotTrophLevel(nets$cayman_islands)
plotTrophLevel(netData$CaymanIs_FW)
#
nets <- readNetwork("Meta.dat","~/Dropbox/Projects/StabilityConstraints/Data",edgeListFormat = 2)
# The incoherence parameter seems not well calculated compared to Johnson 2017 Appendix Table S1
#
require(igraph)
calcTopologicalIndices(nets)
calcIncoherence(nets)
mean(degree(nets$benguela,mode="out"))
calcIncoherence(nets$bridge)
mean(degree(nets$bridge))
degree()
# Test modularity & incoherence Q
# 1 =Omnivory link 5+-4
#
g <- graph_from_literal( 1 -+ 4 -+ 7,2 -+ 5 -+ 9, 4+-5,
3 -+ 6 -+ 8,5 -+8, simplify = FALSE)
plotTrophLevel(g,vertexLabel = TRUE,vertexSizeFactor = 20,modules = TRUE)
title(main="9 species")
calcTopologicalIndices(g)
calcIncoherence(g)
calcModularitySWnessZScore(g,generateERbasal(g,10),0.1)
# Completely coherent Q=0 network
#
g <- graph_from_literal( 1 -+ 4 -+ 7,2 -+ 5 , 2-+4, 5-+7, 3-+6 -+8,
5 -+8, simplify = FALSE)
plotTrophLevel(g,vertexLabel = TRUE,vertexSizeFactor = 20,modules = TRUE)
calcTopologicalIndices(g)
calcIncoherence(g)
calcModularitySWnessZScore(g,generateERbasal(g,100),0.1,ncores = 0)
dg <- components(g)
m<-cluster_spinglass(g)
m$membership
#
# Testing Curve Ball algorithm
#
g <- graph_from_literal( 2 -+ 1 +-3,4 -+ 1, 3+-3, 4-+6-+2, 2+-5-+3, simplify = FALSE)
calcTopologicalIndices(g)
plotTrophLevel(g,vertexLabel = TRUE,vertexSizeFactor = 20,modules = TRUE)
gg <- curveBall(g,10)
gg <- curve_ball(g,1000)
calcTopologicalIndices(gg,ncores=48)
plotTrophLevel(gg[[1]],vertexLabel = TRUE,vertexSizeFactor = 20,modules = TRUE)
plotTrophLevel(gg[[2]],vertexLabel = TRUE,vertexSizeFactor = 10,modules=TRUE)
plotTrophLevel(gg[[3]],vertexLabel = TRUE,vertexSizeFactor = 10,modules = TRUE)
plotTrophLevel(gg[[10]],vertexLabel = TRUE,vertexSizeFactor = 20,modules = TRUE)
#
# Add weight
#
E(g)$weight <- sample(c(.1,.2,.8,.9),gsize(g),replace=TRUE)
m <- get.adjacency(g,sparse=FALSE,attr="weight")
gg <- curveBall(g,2,istrength =TRUE)
m1 <- get.adjacency(gg[[1]],sparse=FALSE,attr="weight")
m
m1
# Test SW zscores
#
any(sapply(gg, function(g) components(g)$no>1))
calcModularitySWnessZScore(g,gg,0.01)
# Test Parallel
#
p <- calcTopologicalIndices(gg,4)
p <- calcIncoherence(gg,4)
# Calculate 1000 randomizations of a real network
#
gg <- curveBall(netData[[1]],100)
sapply(gg, function(g) components(g)$no)
lapply(gg,count_components)!=1
tp <- calcTopologicalIndices(gg)
tp1<- calcTopologicalIndices(gg,4)
any(tp!=tp1)
tp <- calcIncoherence(gg)
tp1 <- calcIncoherence(gg,ncores=4)
any(tp!=tp1)
#
# Duplicate the same network to test the function
#
gg <- rep(list(netData[[1]]),100)
p <- calcIncoherence(gg)
p1 <- calcIncoherence(gg,ncores=4)
any(p!=p1)
require(ggplot2)
ggplot(tp, aes(Omnivory)) + geom_density() + theme_bw()
require(ggplot2)
ggplot(tp, aes(Q)) + geom_density() + theme_bw()
ggplot(tp, aes(mTI)) + geom_density() + theme_bw()
tp1<- calcIncoherence(netData[[1]])
tp1$y <- 0
ggplot(tp, aes(Q)) + geom_density() + theme_bw() + geom_point(data=tp1,aes(Q,y))
# generalization = in degree
# vulnerability = out degree
dn <- list.files("~/Academicos/InconclusedProjects/NetworksAsUnifyingPrinciple/Data",pattern = "^PotterCove.*\\.txt$")
gt <- readMultiplex("~/Academicos/InconclusedProjects/NetworksAsUnifyingPrinciple/Data/PotterCove_Multiplex.txt",format="GLV")
# Number of interaction by type
sapply(gt,ecount)
sapply(gt,vcount)
# Proportion of interactions by type
sapply(gt,ecount)/sum(sapply(gt,ecount))
# Connectivity
sapply(gt,ecount)/(sapply(gt,vcount)*sapply(gt,vcount))
# Multiplex QSS
calc_QSS(gt)
# QSS MEing
# 0 25.87422
calc_QSS(gt[[3]])
# QSS MEing
# 0 2.099394
glv <- toGLVadjMat(gt)
sum(glv==-1) # Competitive + trophic
# Meta-web assembly models package
#
require(meweasmo)
pin <- calcPropInteractionsGLVadjMat(glv, rep(1,times=nrow(glv)))
# Read the file directly
mtm <- as.matrix(mtm[,2:92])
#
mtm <- read.delim("~/Dropbox/Projects/NetworksAsUnifyingPrinciple/Data/PotterCove_Multiplex.txt")
pin1 <- calcPropInteractionsGLVadjMat(mtm, rep(1,times=nrow(glv)))
# The proportion of types of interactions must be equal
#
pin == pin1
plotTrophLevel(gt[[3]],modules = T)
# Read multiple interaction network in different layers
#
fileName <- c(system.file("extdata", package = "multiweb"))
dn <- list.files(fileName,pattern = "^Kefi2015.*\\.txt$")
g <- readNetwork(dn,fileName, skipColumn = 2)
gt <- fromIgraphToMgraph(g,c("Negative","Positive","Antagonistic"))
types <- c("Competitive","Mutualistic","Trophic")
gt <- readMultiplex(dn,types,fileName, skipColumn = 2)
tp <- calc_topological_indices(gt)
# how many interactions from each type
#
sapply(gt,ecount)/sum(sapply(gt,ecount))
# Number of species
sapply(gt,vcount)
# Total Connectivity
sum(sapply(gt,ecount))/(106*106)
# Calc multi type interaction QSS
#
calc_QSS(gt)
# QSS MEing
# 1 0 26.04826
calc_QSS(g[[3]]) # QSS with trophic interactions only should be more stable
# QSS MEing
# 1 0 3.434581
#
# Select the first type of interaction
#
nm <- names(gt)
gt[[nm[1]]]
glv <- toGLVadjMat(gt)
sum(glv==-1) # Competitive + trophic
#
# Simple test
#
g <- graph_from_literal( 2 -+ 1 +-3,4 -+ 1, 4-+4, 3+-3, 5-+5, 4-+6-+2, 2+-5-+3, simplify = FALSE)
E(g)$type <- "Trophic"
c <- cluster_infomap(as.undirected(g))
plot_troph_level(g,vertexLabel = TRUE,vertexSizeFactor = 20,vertexSizeMin = 12, modules = TRUE, community_obj = c)
plot_troph_level(g,tk=FALSE,vertexLabel = TRUE,vertexSizeFactor = 20,vertexSizeMin = 12, modules = TRUE,vertex.label.color="white",edge.width = 0.3)
plot_troph_level(g,modules =FALSE ,tk=TRUE, vertexLabel = TRUE,vertexSizeFactor = 20)
g1 <- graph_from_literal(1,2,3, 4 -+ 5, 5-+4,6, simplify = FALSE)
E(g1)$type <- "Competitive"
g2 <- graph_from_literal(1 -+ 2,2,3,4,5, 6 -+ 3, 3-+6, simplify = FALSE)
E(g2)$type <- "Mutualistic"
g2
#
# Builds an mgraph object
#
mg <- fromIgraphToMgraph(list(g1,g2,g),c("Competitive", "Mutualistic", "Trophic"))
#
# Automatically uses the function toGLVadjMat to calculate the Generalized Lotka-Volterra interaction matrix
# and then calculates the QSS
#
set.seed(423)
calc_QSS(list(g,g1,g2))
set.seed(423)
calc_QSS(list(g,g1,g2),returnRaw = TRUE)
calc_QSS(mg)
#
# Add weights
#
set.seed(1231)
E(g)$weight <- sample(c(.1,.2,.3,.9),gsize(g),replace=TRUE)
E(g1)$weight <- sample(c(.1,.2,.8,.9),gsize(g1),replace=TRUE)
E(g2)$weight <- sample(c(.1,.2,.8,.9),gsize(g2),replace=TRUE)
mg <- fromIgraphToMgraph(list(g1,g2,g),c("Competitive", "Mutualistic", "Trophic"))
toGLVadjMat(mg,istrength=TRUE)
calc_QSS(list(g,g1,g2))calc_QSS(list(g,g1,g2),istrength = TRUE)
calc_QSS(mg)
calc_QSS(mg,istrength = TRUE)
calc_QSS(mg,istrength = TRUE,nsim=1)
calc_QSS(mg,istrength = FALSE,nsim=1)
# Calc QSS with mean & maximum weigth
#
set.seed(3231)
calc_QSS(g,istrength = FALSE)
calc_QSS(g,istrength = TRUE)
calc_QSS(g,istrength = TRUE,nsim = 1)
calc_QSS(g,istrength = FALSE,nsim = 1)
g1 <- g
E(g1)$weight <- max(E(g)$weight)
calc_QSS(g1,istrength = TRUE)
calc_QSS(g1,istrength = TRUE,nsim=1)
E(g1)$weight <- mean(E(g)$weight)
calc_QSS(g1,istrength = TRUE)
#
require(meweasmo)
calcPropInteractionsGLVadjMat(glv, rep(1,times=nrow(glv)))
#
# Test network with weights
#
fileName <- "../../Collaborations/NetworkGolfoSanJorge/Data/TTSinPescaInteraccion.dat"
g <- readNetwork(fileName,edgeListFormat=2 )
V(g)[nei(V(g)[name=="POM"],"out")]
V(g)[nei(V(g)[name=="POM"],"in")]
g
g[]
# Get the weights of a predator
g[,"Acanthistius patachonicus"]
# Get the adjacency matrix with weights
as_adjacency_matrix(g,attr="weight",sparse = FALSE)
edge_attr(g,"weight")
#
# Test calc_QSS
#
set.seed(342)
calc_QSS(g,istrength = TRUE, nsim=10)
set.seed(342)
calc_QSS(g,istrength = TRUE, nsim=10,returnRaw = TRUE)
glvI <- fromIgraphToMgraph(list(make_empty_graph(n=vcount(g)),make_empty_graph(n=vcount(g)),g) ,c("empty","empty","Antagonistic"))
glv <- toGLVadjMat(glvI,c("empty","empty","Antagonistic"),istrength = TRUE) #
mean(glv[glv>0])
glv[glv>0] <- glv[glv>0]*0.1
plotTrophLevel(g,vertexLabel = TRUE,vertexSizeFactor = 5,modules = TRUE)
null <- curveBall(g,100)
calc_weighted_topological_indices(null[[1]])
calc_topological_indices(null[[2]])
null <- curveBall(g,100,istrength = TRUE)
calc_weighted_topological_indices(null[[1]])
calc_weighted_topological_indices(null[[2]])
calc_modularity_swness_zscore(g,null)
# Plot a network with two different components as modules
#
g <- graph_from_literal( 2 -+ 1 +-3,4 -+ 1, 3+-3, 5-+5, 4-+6-+2, 2+-5-+3, simplify = FALSE)
g1 <- graph_from_literal( 7 -+ 8 +-9,10 -+ 8, 10-+10, 9+-9, 11-+11, 10-+13-+12, 12+-11-+9, simplify = FALSE)
g2 <- g+g1
plotTrophLevel(g2,vertexLabel = TRUE,vertexSizeFactor = 7,modules = TRUE)
#
# Test calc_weighted_topological_indices
#
E(g)$weight <- sample(c(0.001,.01,.1,.5,.9),gsize(g),replace=TRUE)
calc_weighted_topological_indices(g)
E(g)$weight <- rep(.01,times=gsize(g))
calc_weighted_topological_indices(g)
calc_topological_indices(g)
# Test quantitative connectance
#
#
# Equal fluxes
m <- matrix(0,nrow=4,ncol=4)
m[1,2] <- m[1,3] <- m[3,4]<- 2
m[2,1] <- m[3,1] <- m[4,3] <- -2
calc_quantitative_connectance(m, c(1,1,1,1))
calcQuantitativeConnectance(m, c(1,1,1,1))
g <- fromGLVadjToIgraph(m,c(1,1,1,1))
print(calcTopologicalIndices(g))
print(calc_topological_indices(g))
plotTrophLevel(g,vertexLabel = T)
# Unequal fluxes
#
m <- matrix(0,nrow=5,ncol=5)
m[1,2] <- m[1,3] <- m[3,4]<- .2
m[2,1] <- m[3,1] <- m[4,3] <- -2
m[5,4] <- m[4,5] <- 0.1 # Mutualistic
m[1,1] <- -0.01 # Cannibalistic
g <- fromGLVadjToIgraph(m,c(1,1,1,1,0))
g <- fromGLVadjToIgraph(m,c(0,1,1,1,1))
calc_quantitative_connectance(m, c(1,1,1,1))
# Examples from Ulanowicz, R.E. & Wolff, W.F. (1991). Ecosystem flow networks: Loaded dice? Math. Biosci., 103, 45–68
#
m <- matrix(0,nrow=4,ncol=4)
m[1,2] <- m[2,3] <- m[3,4] <- m[4,1] <- 75
m[1,3] <- m[3,1] <- m[2,4] <- m[4,2] <- 75
calc_quantitative_connectance(m, c(1,1,1,1)) # 0.5, 2
sum(m>0)/(nrow(m)^2)
m <- matrix(0,nrow=4,ncol=4)
m[1,2] <- m[2,3] <- m[3,4] <- m[4,1] <- 50
m[1,3] <- m[3,1] <- m[2,4] <- m[4,2] <- 100
print(calc_quantitative_connectance(m, c(1,1,1,1))) # =0.4724704, 1.889882
sum(m>0)/(nrow(m)^2)
m <- matrix(0,nrow=4,ncol=4)
m[1,2] <- 560
m[1,3] <- m[2,3] <- m[3,4] <- m[4,1] <- m[3,1] <- m[2,4] <- m[4,2] <- 5.66
print(calc_quantitative_connectance(m, c(1,1,1,1))) # =0.4411018, 1.096489
sum(m>0)/(nrow(m)^2)
#
# Test Topological roles
#
sapply(netData,igraph::vcount)
g <- netData[[23]]
tp <- calc_topological_roles(g,nsim=10,ncores=4)
classify_topological_roles(tp,g,plt=TRUE)
m <- igraph::cluster_infomap(g)
length(m)
m <- igraph::cluster_walktrap(g)
length(m)
igraph::membership(m)
tp1 <- calc_topological_roles(g,nsim=10,ncores=4,community=m)
tp1
classify_topological_roles(tp1,g,community=m,plt=TRUE)
#
# Test QSS
#
sapply(netData,vcount)
set.seed(342)
g <- netData[[2]]
g1 <- netData[[18]]
tp <- calc_QSS(list(g,g1),nsim=100,ncores=8)
# with default values QSS is almost always 0
tp
# QSS MEing
#1 0 4.591426
#2 0 2.052344
set.seed(342)
tp <- calc_QSS(list(g,g1),nsim=100,ncores=8,negative=-1,positive=1)
tp
# QSS MEing
# 1 0.0005 0.24024671
# 2 0.3224 0.07553226
# Test for differences
#
prop.test(x=tp$QSS*10000, n=c(10000,10000))
#
# test calc_modularity
#
nullg <- generateERbasal(netData[[1]],10)
calc_modularity(nullg)
# with weigths
g <- netData[[1]]
names(netData)[1]
calc_modularity(g)
V(g)$weight <- runif(vcount(g))
calc_modularity(g,weights = NULL)
#
# test QSS with 1 simulation vs mean of QSS with 1000 simulations
#
set.seed(5431)
g <- netData[[1]]
E(g)$weight <- runif(ecount(g))
calc_QSS(g,istrength = FALSE) # 1 0 3.106538
calc_QSS(g,istrength = TRUE,ncores =4 ) # 1 0 1.575575
calc_QSS(g,istrength = FALSE,nsim=1) # 1 0 2.461278
calc_QSS(g,istrength = TRUE,nsim=1) # 1 0 1.421724
raw <- calc_QSS(g,istrength = TRUE,ncores =4 ,returnRaw = TRUE) # 1 0 1.637529
require(ggplot2)
ggplot(raw, aes(maxre)) + geom_density() + theme_bw() + geom_vline(xintercept = calc_QSS(g,istrength = TRUE,nsim=1)$MEing, linetype="dashed")
#
# using lognormal interaction strengths
#
E(g)$weight <- rlnorm(ecount(g))
calc_QSS(g,istrength = FALSE) # 1 0 3.093221
calc_QSS(g,istrength = TRUE,ncores =4 ) # 1 0 4.577607
calc_QSS(g,istrength = FALSE,nsim=1) # 1 0 2.461278
calc_QSS(g,istrength = TRUE,nsim=1) # 1 0 4.01689
raw <- calc_QSS(g,istrength = TRUE,ncores =4 ,returnRaw = TRUE)
ggplot(raw, aes(maxre)) + geom_density() + theme_bw() + geom_vline(xintercept = calc_QSS(g,istrength = TRUE,nsim=1)$MEing, linetype="dashed")
#
# Test calc_QSS_extinction_dif
#
g <- netData[[1]]
E(g)$weight <- runif(ecount(g))
E(g)$w <- runif(ecount(g))
calc_QSS(g,istrength = FALSE)
calc_QSS(g,istrength = TRUE)
edge_attr_names(g)
tic()
dif <- calc_QSS_extinction_dif(g,V(g)$name[1:3],nsim=20,istrength = FALSE)
toc()
ggplot(dif, aes(difQSS,fill=Deleted)) + geom_density( alpha=0.3 ) + theme_bw() + geom_vline(xintercept = 0, linetype="dashed") +
facet_wrap(~Deleted, ncol = 1)
dif1 <- calc_QSS_extinction_dif(g,V(g)$name[1:3],nsim=10,istrength = TRUE)
calc_QSS_extinction_dif_grp(g,V(g)$name[1:3],nsim=10,istrength = TRUE)
#
# Test calc_QSS_extinction_dif
#
eseq <- calc_QSS_extinctions_seq(g,V(g)$name[1:5],nsim=10,istrength = FALSE)
eseq
eseq <- calc_QSS_extinctions_seq(g,V(g)$name,nsim=100,istrength = TRUE)
eseq
g_del <- calc_QSS_extinction_dif_grp(g, bygroup,nsim,ncores,istrength)
# Test calc_interaction_intensity
#
# get the data frame of interactions
#
g <- netData[[1]]
require(dplyr)
require(igraph)
set.seed(7815)
da <- as_long_data_frame(g) %>% dplyr::select(from:to) %>% mutate(con_mm=rlnorm(n(),5,2),res_mm=con_mm - 30 ,int_dim=sample(c("2D","3D"),n(),replace=TRUE), res_den = -999)
calc_interaction_intensity(da,res_mm,res_den,con_mm,int_dim,10)
#
# Test the nsims parameter
#
i1 <- calc_interaction_intensity(da[1,],res_mm,res_den,con_mm,int_dim,1000)
i <- calc_interaction_intensity(da[1,],res_mm,res_den,con_mm,int_dim,1)
require(ggplot2)
ggplot(i1,aes(qRC)) + geom_histogram()
ggplot(i1,aes(qRC)) + geom_density() + scale_x_log10() + geom_vline(xintercept = i$qRC)
# Using values for res_den
#
da <- as_long_data_frame(g) %>% dplyr::select(from:to) %>%
mutate(con_mm=rlnorm(n(),5,2),res_mm=con_mm - 30 ,int_dim=sample(c("2D","3D"),n(),replace=TRUE), res_den = runif(n(),1e-5,1))
calc_interaction_intensity(da,res_mm,res_den,con_mm,int_dim)
lsaravia/EcoNetwork documentation built on March 20, 2024, 3:27 p.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
# Test
require(igraph)
fileName <- c(system.file("extdata", package = "multiweb"))
fileName <- system.file("extdata", "BarentsBoreal_FW.csv", package = "multiweb")
g <- readNetwork(fileName)
V(g)$weigth <- runif(vcount(g))
plotTrophLevel(g,vertexLabel = FALSE,vertexSizeFactor = 3,modules = TRUE,maxTL=4,weights=NULL, frame.color="black")
tk <- plotTrophLevel(g,vertexLabel = FALSE,vertexSizeFactor = 3,modules = TRUE, tk=TRUE)
plotTrophLevel(g,vertexLabel = FALSE,vertexSizeFactor = 3,modules = TRUE, lMat=tk)
dn <- list.files("inst/extdata",pattern = "^.*\\.csv$")
g<- readNetwork(dn,"inst/extdata")
names(g)
E(g[[1]])[which_loop(g[[1]])]
plotTrophLevel(g[[1]])
require(igraph)
require(NetIndices)
g <- graph_from_literal( 1 -+ 4 -+ 7,2 -+ 5 -+7, 3-+6-+7, 7-+7, 4+-3, 2-+7, simplify = FALSE)
calc_topological_indices(g)
plotTrophLevel(g,vertexLabel = TRUE,vertexSizeFactor = 20,modules = TRUE,maxTL=4,weights=NULL)
plotTrophLevel(g,vertexLabel = TRUE,vertexSizeFactor = 20,modules = TRUE,maxTL=4,weights=NULL, bpal=viridisLite::viridis(11))
# Test tk and use of lMat
tk <- plotTrophLevel(g,vertexLabel = TRUE,vertexSizeFactor = 20,modules = TRUE, tk=TRUE,vertex.label.color="white")
tk [,2] <- seq(0,100, length.out = 7)
plotTrophLevel(g,vertexLabel = TRUE,vertexSizeFactor = 20,modules = TRUE, lMat=tk,maxTL=2)
# add weights
#
E(g)$weight <- c(2,2.5,0.1,0.2,1,5,0.1,2)
plotTrophLevel(g,vertexLabel = TRUE,vertexSizeFactor = 20,modules = TRUE,maxTL=4,weights=NA,edge.width=5)
plotTrophLevel(g,vertexLabel = TRUE,vertexSizeFactor = 20,modules = TRUE,maxTL=4,weights=NULL,edge.width=5)
# Test incoherence
#
calcIncoherence(g)
g <- graph_from_literal( 1 -+ 4 -+ 7,2 -+ 5 -+7, 3-+6-+7, 4+-5, simplify = FALSE)
calcTopologicalIndices(g)
plotTrophLevel(g,vertexLabel = TRUE,vertexSizeFactor = 20, modules = TRUE)
nullg <- generateERbasal(g,10)
calcIncoherence(nullg,ncores=0)
calcModularitySWnessZScore(g,nullg,0.1)
calc_swness_zscore(g,nullg,0.1,ncores = 0)
# test reading .csv files
#
dn <- list.files("~/Academicos/GitProjects/MarineFoodWebsSmallWorld/Data/FoodWebs",pattern = "^.*\\.csv$")
#
# BEWARE different files have different structure
#
nets <- readNetwork(dn[11],"~/Academicos/GitProjects/MarineFoodWebsSmallWorld/Data/FoodWebs")
nets <- readNetwork(dn[16],"~/Academicos/GitProjects/MarineFoodWebsSmallWorld/Data/FoodWebs")
nets <- readNetwork(dn[17],"~/Academicos/GitProjects/MarineFoodWebsSmallWorld/Data/FoodWebs")
nets <- readNetwork(dn[18],"~/Academicos/GitProjects/MarineFoodWebsSmallWorld/Data/FoodWebs")
nets <- readNetwork(dn[19],"~/Academicos/GitProjects/MarineFoodWebsSmallWorld/Data/FoodWebs")
nets <- readNetwork(dn[20],"~/Academicos/GitProjects/MarineFoodWebsSmallWorld/Data/FoodWebs")
nets <- readNetwork(dn[21],"~/Academicos/GitProjects/MarineFoodWebsSmallWorld/Data/FoodWebs")
V(nets)[[]]
nets <- readNetwork(dn[22],"~/Academicos/GitProjects/MarineFoodWebsSmallWorld/Data/FoodWebs")
nets <- readNetwork(dn[23],"~/Academicos/GitProjects/MarineFoodWebsSmallWorld/Data/FoodWebs")
nets <- readNetwork(dn,"~/Academicos/GitProjects/MarineFoodWebsSmallWorld/Data/FoodWebs")
nets$CaymanIs_FW <- readNetwork(dn[8],"~/Academicos/GitProjects/MarineFoodWebsSmallWorld/Data/FoodWebs",fhead = FALSE,skipColumn = 0)
nets$Jamaica_FW <- readNetwork(dn[15],"~/Academicos/GitProjects/MarineFoodWebsSmallWorld/Data/FoodWebs",fhead = FALSE,skipColumn = 0)
nets$Cuba_FW <- readNetwork(dn[11],"~/Academicos/GitProjects/MarineFoodWebsSmallWorld/Data/FoodWebs",fhead = FALSE,skipColumn = 0)
calcTopologicalIndices(nets)
calcIncoherence(nets)
#netData <- nets
#devtools::use_data(netData,overwrite = TRUE)
# test reading .dat files from Johnson 2017
#
dn <- list.files("~/Dropbox/Projects/NetworksAsUnifyingPrinciple/Data",pattern = "^.*\\.dat$")
nets <- readNetwork(dn,"~/Dropbox/Projects/NetworksAsUnifyingPrinciple/Data",fhead=FALSE)
calcTopologicalIndices(list(netData$CaymanIs_FW,nets$cayman_islands))
require(RColorBrewer)
plotTrophLevel(nets$cayman_islands)
plotTrophLevel(netData$CaymanIs_FW)
#
nets <- readNetwork("Meta.dat","~/Dropbox/Projects/StabilityConstraints/Data",edgeListFormat = 2)
# The incoherence parameter seems not well calculated compared to Johnson 2017 Appendix Table S1
#
require(igraph)
calcTopologicalIndices(nets)
calcIncoherence(nets)
mean(degree(nets$benguela,mode="out"))
calcIncoherence(nets$bridge)
mean(degree(nets$bridge))
degree()
# Test modularity & incoherence Q
# 1 =Omnivory link 5+-4
#
g <- graph_from_literal( 1 -+ 4 -+ 7,2 -+ 5 -+ 9, 4+-5,
3 -+ 6 -+ 8,5 -+8, simplify = FALSE)
plotTrophLevel(g,vertexLabel = TRUE,vertexSizeFactor = 20,modules = TRUE)
title(main="9 species")
calcTopologicalIndices(g)
calcIncoherence(g)
calcModularitySWnessZScore(g,generateERbasal(g,10),0.1)
# Completely coherent Q=0 network
#
g <- graph_from_literal( 1 -+ 4 -+ 7,2 -+ 5 , 2-+4, 5-+7, 3-+6 -+8,
5 -+8, simplify = FALSE)
plotTrophLevel(g,vertexLabel = TRUE,vertexSizeFactor = 20,modules = TRUE)
calcTopologicalIndices(g)
calcIncoherence(g)
calcModularitySWnessZScore(g,generateERbasal(g,100),0.1,ncores = 0)
dg <- components(g)
m<-cluster_spinglass(g)
m$membership
#
# Testing Curve Ball algorithm
#
g <- graph_from_literal( 2 -+ 1 +-3,4 -+ 1, 3+-3, 4-+6-+2, 2+-5-+3, simplify = FALSE)
calcTopologicalIndices(g)
plotTrophLevel(g,vertexLabel = TRUE,vertexSizeFactor = 20,modules = TRUE)
gg <- curveBall(g,10)
gg <- curve_ball(g,1000)
calcTopologicalIndices(gg,ncores=48)
plotTrophLevel(gg[[1]],vertexLabel = TRUE,vertexSizeFactor = 20,modules = TRUE)
plotTrophLevel(gg[[2]],vertexLabel = TRUE,vertexSizeFactor = 10,modules=TRUE)
plotTrophLevel(gg[[3]],vertexLabel = TRUE,vertexSizeFactor = 10,modules = TRUE)
plotTrophLevel(gg[[10]],vertexLabel = TRUE,vertexSizeFactor = 20,modules = TRUE)
#
# Add weight
#
E(g)$weight <- sample(c(.1,.2,.8,.9),gsize(g),replace=TRUE)
m <- get.adjacency(g,sparse=FALSE,attr="weight")
gg <- curveBall(g,2,istrength =TRUE)
m1 <- get.adjacency(gg[[1]],sparse=FALSE,attr="weight")
m
m1
# Test SW zscores
#
any(sapply(gg, function(g) components(g)$no>1))
calcModularitySWnessZScore(g,gg,0.01)
# Test Parallel
#
p <- calcTopologicalIndices(gg,4)
p <- calcIncoherence(gg,4)
# Calculate 1000 randomizations of a real network
#
gg <- curveBall(netData[[1]],100)
sapply(gg, function(g) components(g)$no)
lapply(gg,count_components)!=1
tp <- calcTopologicalIndices(gg)
tp1<- calcTopologicalIndices(gg,4)
any(tp!=tp1)
tp <- calcIncoherence(gg)
tp1 <- calcIncoherence(gg,ncores=4)
any(tp!=tp1)
#
# Duplicate the same network to test the function
#
gg <- rep(list(netData[[1]]),100)
p <- calcIncoherence(gg)
p1 <- calcIncoherence(gg,ncores=4)
any(p!=p1)
require(ggplot2)
ggplot(tp, aes(Omnivory)) + geom_density() + theme_bw()
require(ggplot2)
ggplot(tp, aes(Q)) + geom_density() + theme_bw()
ggplot(tp, aes(mTI)) + geom_density() + theme_bw()
tp1<- calcIncoherence(netData[[1]])
tp1$y <- 0
ggplot(tp, aes(Q)) + geom_density() + theme_bw() + geom_point(data=tp1,aes(Q,y))
# generalization = in degree
# vulnerability = out degree
dn <- list.files("~/Academicos/InconclusedProjects/NetworksAsUnifyingPrinciple/Data",pattern = "^PotterCove.*\\.txt$")
gt <- readMultiplex("~/Academicos/InconclusedProjects/NetworksAsUnifyingPrinciple/Data/PotterCove_Multiplex.txt",format="GLV")
# Number of interaction by type
sapply(gt,ecount)
sapply(gt,vcount)
# Proportion of interactions by type
sapply(gt,ecount)/sum(sapply(gt,ecount))
# Connectivity
sapply(gt,ecount)/(sapply(gt,vcount)*sapply(gt,vcount))
# Multiplex QSS
calc_QSS(gt)
# QSS MEing
# 0 25.87422
calc_QSS(gt[[3]])
# QSS MEing
# 0 2.099394
glv <- toGLVadjMat(gt)
sum(glv==-1) # Competitive + trophic
# Meta-web assembly models package
#
require(meweasmo)
pin <- calcPropInteractionsGLVadjMat(glv, rep(1,times=nrow(glv)))
# Read the file directly
mtm <- as.matrix(mtm[,2:92])
#
mtm <- read.delim("~/Dropbox/Projects/NetworksAsUnifyingPrinciple/Data/PotterCove_Multiplex.txt")
pin1 <- calcPropInteractionsGLVadjMat(mtm, rep(1,times=nrow(glv)))
# The proportion of types of interactions must be equal
#
pin == pin1
plotTrophLevel(gt[[3]],modules = T)
# Read multiple interaction network in different layers
#
fileName <- c(system.file("extdata", package = "multiweb"))
dn <- list.files(fileName,pattern = "^Kefi2015.*\\.txt$")
g <- readNetwork(dn,fileName, skipColumn = 2)
gt <- fromIgraphToMgraph(g,c("Negative","Positive","Antagonistic"))
types <- c("Competitive","Mutualistic","Trophic")
gt <- readMultiplex(dn,types,fileName, skipColumn = 2)
tp <- calc_topological_indices(gt)
# how many interactions from each type
#
sapply(gt,ecount)/sum(sapply(gt,ecount))
# Number of species
sapply(gt,vcount)
# Total Connectivity
sum(sapply(gt,ecount))/(106*106)
# Calc multi type interaction QSS
#
calc_QSS(gt)
# QSS MEing
# 1 0 26.04826
calc_QSS(g[[3]]) # QSS with trophic interactions only should be more stable
# QSS MEing
# 1 0 3.434581
#
# Select the first type of interaction
#
nm <- names(gt)
gt[[nm[1]]]
glv <- toGLVadjMat(gt)
sum(glv==-1) # Competitive + trophic
#
# Simple test
#
g <- graph_from_literal( 2 -+ 1 +-3,4 -+ 1, 4-+4, 3+-3, 5-+5, 4-+6-+2, 2+-5-+3, simplify = FALSE)
E(g)$type <- "Trophic"
c <- cluster_infomap(as.undirected(g))
plot_troph_level(g,vertexLabel = TRUE,vertexSizeFactor = 20,vertexSizeMin = 12, modules = TRUE, community_obj = c)
plot_troph_level(g,tk=FALSE,vertexLabel = TRUE,vertexSizeFactor = 20,vertexSizeMin = 12, modules = TRUE,vertex.label.color="white",edge.width = 0.3)
plot_troph_level(g,modules =FALSE ,tk=TRUE, vertexLabel = TRUE,vertexSizeFactor = 20)
g1 <- graph_from_literal(1,2,3, 4 -+ 5, 5-+4,6, simplify = FALSE)
E(g1)$type <- "Competitive"
g2 <- graph_from_literal(1 -+ 2,2,3,4,5, 6 -+ 3, 3-+6, simplify = FALSE)
E(g2)$type <- "Mutualistic"
g2
#
# Builds an mgraph object
#
mg <- fromIgraphToMgraph(list(g1,g2,g),c("Competitive", "Mutualistic", "Trophic"))
#
# Automatically uses the function toGLVadjMat to calculate the Generalized Lotka-Volterra interaction matrix
# and then calculates the QSS
#
set.seed(423)
calc_QSS(list(g,g1,g2))
set.seed(423)
calc_QSS(list(g,g1,g2),returnRaw = TRUE)
calc_QSS(mg)
#
# Add weights
#
set.seed(1231)
E(g)$weight <- sample(c(.1,.2,.3,.9),gsize(g),replace=TRUE)
E(g1)$weight <- sample(c(.1,.2,.8,.9),gsize(g1),replace=TRUE)
E(g2)$weight <- sample(c(.1,.2,.8,.9),gsize(g2),replace=TRUE)
mg <- fromIgraphToMgraph(list(g1,g2,g),c("Competitive", "Mutualistic", "Trophic"))
toGLVadjMat(mg,istrength=TRUE)
calc_QSS(list(g,g1,g2))calc_QSS(list(g,g1,g2),istrength = TRUE)
calc_QSS(mg)
calc_QSS(mg,istrength = TRUE)
calc_QSS(mg,istrength = TRUE,nsim=1)
calc_QSS(mg,istrength = FALSE,nsim=1)
# Calc QSS with mean & maximum weigth
#
set.seed(3231)
calc_QSS(g,istrength = FALSE)
calc_QSS(g,istrength = TRUE)
calc_QSS(g,istrength = TRUE,nsim = 1)
calc_QSS(g,istrength = FALSE,nsim = 1)
g1 <- g
E(g1)$weight <- max(E(g)$weight)
calc_QSS(g1,istrength = TRUE)
calc_QSS(g1,istrength = TRUE,nsim=1)
E(g1)$weight <- mean(E(g)$weight)
calc_QSS(g1,istrength = TRUE)
#
require(meweasmo)
calcPropInteractionsGLVadjMat(glv, rep(1,times=nrow(glv)))
#
# Test network with weights
#
fileName <- "../../Collaborations/NetworkGolfoSanJorge/Data/TTSinPescaInteraccion.dat"
g <- readNetwork(fileName,edgeListFormat=2 )
V(g)[nei(V(g)[name=="POM"],"out")]
V(g)[nei(V(g)[name=="POM"],"in")]
g
g[]
# Get the weights of a predator
g[,"Acanthistius patachonicus"]
# Get the adjacency matrix with weights
as_adjacency_matrix(g,attr="weight",sparse = FALSE)
edge_attr(g,"weight")
#
# Test calc_QSS
#
set.seed(342)
calc_QSS(g,istrength = TRUE, nsim=10)
set.seed(342)
calc_QSS(g,istrength = TRUE, nsim=10,returnRaw = TRUE)
glvI <- fromIgraphToMgraph(list(make_empty_graph(n=vcount(g)),make_empty_graph(n=vcount(g)),g) ,c("empty","empty","Antagonistic"))
glv <- toGLVadjMat(glvI,c("empty","empty","Antagonistic"),istrength = TRUE) #
mean(glv[glv>0])
glv[glv>0] <- glv[glv>0]*0.1
plotTrophLevel(g,vertexLabel = TRUE,vertexSizeFactor = 5,modules = TRUE)
null <- curveBall(g,100)
calc_weighted_topological_indices(null[[1]])
calc_topological_indices(null[[2]])
null <- curveBall(g,100,istrength = TRUE)
calc_weighted_topological_indices(null[[1]])
calc_weighted_topological_indices(null[[2]])
calc_modularity_swness_zscore(g,null)
# Plot a network with two different components as modules
#
g <- graph_from_literal( 2 -+ 1 +-3,4 -+ 1, 3+-3, 5-+5, 4-+6-+2, 2+-5-+3, simplify = FALSE)
g1 <- graph_from_literal( 7 -+ 8 +-9,10 -+ 8, 10-+10, 9+-9, 11-+11, 10-+13-+12, 12+-11-+9, simplify = FALSE)
g2 <- g+g1
plotTrophLevel(g2,vertexLabel = TRUE,vertexSizeFactor = 7,modules = TRUE)
#
# Test calc_weighted_topological_indices
#
E(g)$weight <- sample(c(0.001,.01,.1,.5,.9),gsize(g),replace=TRUE)
calc_weighted_topological_indices(g)
E(g)$weight <- rep(.01,times=gsize(g))
calc_weighted_topological_indices(g)
calc_topological_indices(g)
# Test quantitative connectance
#
#
# Equal fluxes
m <- matrix(0,nrow=4,ncol=4)
m[1,2] <- m[1,3] <- m[3,4]<- 2
m[2,1] <- m[3,1] <- m[4,3] <- -2
calc_quantitative_connectance(m, c(1,1,1,1))
calcQuantitativeConnectance(m, c(1,1,1,1))
g <- fromGLVadjToIgraph(m,c(1,1,1,1))
print(calcTopologicalIndices(g))
print(calc_topological_indices(g))
plotTrophLevel(g,vertexLabel = T)
# Unequal fluxes
#
m <- matrix(0,nrow=5,ncol=5)
m[1,2] <- m[1,3] <- m[3,4]<- .2
m[2,1] <- m[3,1] <- m[4,3] <- -2
m[5,4] <- m[4,5] <- 0.1 # Mutualistic
m[1,1] <- -0.01 # Cannibalistic
g <- fromGLVadjToIgraph(m,c(1,1,1,1,0))
g <- fromGLVadjToIgraph(m,c(0,1,1,1,1))
calc_quantitative_connectance(m, c(1,1,1,1))
# Examples from Ulanowicz, R.E. & Wolff, W.F. (1991). Ecosystem flow networks: Loaded dice? Math. Biosci., 103, 45–68
#
m <- matrix(0,nrow=4,ncol=4)
m[1,2] <- m[2,3] <- m[3,4] <- m[4,1] <- 75
m[1,3] <- m[3,1] <- m[2,4] <- m[4,2] <- 75
calc_quantitative_connectance(m, c(1,1,1,1)) # 0.5, 2
sum(m>0)/(nrow(m)^2)
m <- matrix(0,nrow=4,ncol=4)
m[1,2] <- m[2,3] <- m[3,4] <- m[4,1] <- 50
m[1,3] <- m[3,1] <- m[2,4] <- m[4,2] <- 100
print(calc_quantitative_connectance(m, c(1,1,1,1))) # =0.4724704, 1.889882
sum(m>0)/(nrow(m)^2)
m <- matrix(0,nrow=4,ncol=4)
m[1,2] <- 560
m[1,3] <- m[2,3] <- m[3,4] <- m[4,1] <- m[3,1] <- m[2,4] <- m[4,2] <- 5.66
print(calc_quantitative_connectance(m, c(1,1,1,1))) # =0.4411018, 1.096489
sum(m>0)/(nrow(m)^2)
#
# Test Topological roles
#
sapply(netData,igraph::vcount)
g <- netData[[23]]
tp <- calc_topological_roles(g,nsim=10,ncores=4)
classify_topological_roles(tp,g,plt=TRUE)
m <- igraph::cluster_infomap(g)
length(m)
m <- igraph::cluster_walktrap(g)
length(m)
igraph::membership(m)
tp1 <- calc_topological_roles(g,nsim=10,ncores=4,community=m)
tp1
classify_topological_roles(tp1,g,community=m,plt=TRUE)
#
# Test QSS
#
sapply(netData,vcount)
set.seed(342)
g <- netData[[2]]
g1 <- netData[[18]]
tp <- calc_QSS(list(g,g1),nsim=100,ncores=8)
# with default values QSS is almost always 0
tp
# QSS MEing
#1 0 4.591426
#2 0 2.052344
set.seed(342)
tp <- calc_QSS(list(g,g1),nsim=100,ncores=8,negative=-1,positive=1)
tp
# QSS MEing
# 1 0.0005 0.24024671
# 2 0.3224 0.07553226
# Test for differences
#
prop.test(x=tp$QSS*10000, n=c(10000,10000))
#
# test calc_modularity
#
nullg <- generateERbasal(netData[[1]],10)
calc_modularity(nullg)
# with weigths
g <- netData[[1]]
names(netData)[1]
calc_modularity(g)
V(g)$weight <- runif(vcount(g))
calc_modularity(g,weights = NULL)
#
# test QSS with 1 simulation vs mean of QSS with 1000 simulations
#
set.seed(5431)
g <- netData[[1]]
E(g)$weight <- runif(ecount(g))
calc_QSS(g,istrength = FALSE) # 1 0 3.106538
calc_QSS(g,istrength = TRUE,ncores =4 ) # 1 0 1.575575
calc_QSS(g,istrength = FALSE,nsim=1) # 1 0 2.461278
calc_QSS(g,istrength = TRUE,nsim=1) # 1 0 1.421724
raw <- calc_QSS(g,istrength = TRUE,ncores =4 ,returnRaw = TRUE) # 1 0 1.637529
require(ggplot2)
ggplot(raw, aes(maxre)) + geom_density() + theme_bw() + geom_vline(xintercept = calc_QSS(g,istrength = TRUE,nsim=1)$MEing, linetype="dashed")
#
# using lognormal interaction strengths
#
E(g)$weight <- rlnorm(ecount(g))
calc_QSS(g,istrength = FALSE) # 1 0 3.093221
calc_QSS(g,istrength = TRUE,ncores =4 ) # 1 0 4.577607
calc_QSS(g,istrength = FALSE,nsim=1) # 1 0 2.461278
calc_QSS(g,istrength = TRUE,nsim=1) # 1 0 4.01689
raw <- calc_QSS(g,istrength = TRUE,ncores =4 ,returnRaw = TRUE)
ggplot(raw, aes(maxre)) + geom_density() + theme_bw() + geom_vline(xintercept = calc_QSS(g,istrength = TRUE,nsim=1)$MEing, linetype="dashed")
#
# Test calc_QSS_extinction_dif
#
g <- netData[[1]]
E(g)$weight <- runif(ecount(g))
E(g)$w <- runif(ecount(g))
calc_QSS(g,istrength = FALSE)
calc_QSS(g,istrength = TRUE)
edge_attr_names(g)
tic()
dif <- calc_QSS_extinction_dif(g,V(g)$name[1:3],nsim=20,istrength = FALSE)
toc()
ggplot(dif, aes(difQSS,fill=Deleted)) + geom_density( alpha=0.3 ) + theme_bw() + geom_vline(xintercept = 0, linetype="dashed") +
facet_wrap(~Deleted, ncol = 1)
dif1 <- calc_QSS_extinction_dif(g,V(g)$name[1:3],nsim=10,istrength = TRUE)
calc_QSS_extinction_dif_grp(g,V(g)$name[1:3],nsim=10,istrength = TRUE)
#
# Test calc_QSS_extinction_dif
#
eseq <- calc_QSS_extinctions_seq(g,V(g)$name[1:5],nsim=10,istrength = FALSE)
eseq
eseq <- calc_QSS_extinctions_seq(g,V(g)$name,nsim=100,istrength = TRUE)
eseq
g_del <- calc_QSS_extinction_dif_grp(g, bygroup,nsim,ncores,istrength)
# Test calc_interaction_intensity
#
# get the data frame of interactions
#
g <- netData[[1]]
require(dplyr)
require(igraph)
set.seed(7815)
da <- as_long_data_frame(g) %>% dplyr::select(from:to) %>% mutate(con_mm=rlnorm(n(),5,2),res_mm=con_mm - 30 ,int_dim=sample(c("2D","3D"),n(),replace=TRUE), res_den = -999)
calc_interaction_intensity(da,res_mm,res_den,con_mm,int_dim,10)
#
# Test the nsims parameter
#
i1 <- calc_interaction_intensity(da[1,],res_mm,res_den,con_mm,int_dim,1000)
i <- calc_interaction_intensity(da[1,],res_mm,res_den,con_mm,int_dim,1)
require(ggplot2)
ggplot(i1,aes(qRC)) + geom_histogram()
ggplot(i1,aes(qRC)) + geom_density() + scale_x_log10() + geom_vline(xintercept = i$qRC)
# Using values for res_den
#
da <- as_long_data_frame(g) %>% dplyr::select(from:to) %>%
mutate(con_mm=rlnorm(n(),5,2),res_mm=con_mm - 30 ,int_dim=sample(c("2D","3D"),n(),replace=TRUE), res_den = runif(n(),1e-5,1))
calc_interaction_intensity(da,res_mm,res_den,con_mm,int_dim)
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.