Interconnection motif forms

Description

The dictionary of interconnection motif forms without intra-guild interactions.

Usage

Multi_motif(
  ID = c("M111", "M112", "M113", "M114", "M211", "M212", "M213", "M311", "M312", "M411",
    "M121", "M122-1", "M122-2", "M122-3", "M123-1", "M123-2", "M123-3", "M123-4",
    "M123-5", "M221-1", "M221-2", "M221-3", "M222-1", "M222-2", "M222-3", "M222-4",
    "M222-5", "M222-6", "M222-7", "M222-8", "M222-9", "M321-1", "M321-2", "M321-3",
    "M321-4", "M321-5", "M131", "M132-1", "M132-2", "M132-3", "M132-4", "M132-5",
    "M231-1", "M231-2", "M231-3", "M231-4", "M231-5", "M141", "all")
)

Arguments

Details

An interconnection motif is defined to comprise three sets of connected nodes: the connector nodes (belonging to b-nodes), the nodes in one subnetwork (belonging to a-nodes in the P subnetwork), and the nodes in the other subnetwork (belonging to c-nodes in the Q subnetwork). Each motif has maximumly 6 nodes, resulting in a total of 48 distinct motif forms.

Value

If ID is one ID, it returns a corresponding motif "igraph" object.

If ID is "all", it returns a list containing all 48 motifs of "igraph" objects.

Examples

ID<-c("M111","M112","M113","M114","M211","M212","M213","M311",
"M312","M411","M121","M122-1","M122-2","M122-3","M123-1",
"M123-2","M123-3","M123-4","M123-5","M221-1","M221-2",
"M221-3","M222-1","M222-2","M222-3","M222-4","M222-5",
"M222-6","M222-7","M222-8","M222-9","M321-1","M321-2",
"M321-3","M321-4","M321-5","M131","M132-1","M132-2",
"M132-3","M132-4","M132-5","M231-1","M231-2","M231-3",
"M231-4","M231-5","M141")

m<-Multi_motif("M222-4")
plot(m,main="M222-4")

mr <- par(mfrow=c(4,6),mar=c(1,1,3,1))
for(i in 1:24){
    plot(Multi_motif("all")[[i]],
    vertex.size=30, vertex.label=NA,
    vertex.color="blue",main=ID[i])
}
par(mr)

An empirical tripartite pollinator-plant-herbivore network

Description

A pollinator-plant-herbivore tripartite network from Villa-Galaviz et. al. 2020. Journal of Animal Ecology

Usage

data(PPH_Coltparkmeadow)

Details

An 'igraph' object. This is an empirical tripartite network provided in Villa-Galaviz et. al. 2020. This network has three guilds of species: pollinators, plants and herbivores. Pollinators and plants form the subnetwork with mutualistic interactions, and plants and herbivores form the subnetwork with antagonistic interactions. No intra-guild interactions. Plants are the shared set of species.

References

Villa-Galaviz, E., S. M. Smart, E. L. Clare, S. E. Ward, and J. Memmott. 2021. Differential effects of fertilisers on pollination and parasitoid interaction networks. Journal of animal ecology 90:404-414.

Examples

data(PPH_Coltparkmeadow)

Generating an example tripartite network randomly

Description

Generate a binary tripartite network with three groups of nodes (a-, b- and c-nodes) and two subnetworks (P and Q). Subnetwork P contains links between a- and b-nodes; Subnetwork Q contains links between b- and c-nodes;b-nodes (some of which are connector nodes) are shared nodes between two subnetworks.

Usage

build_toy_net(N_a, N_b, N_c, Co, output_matrices = FALSE)

Arguments

Value

Return a random binary tripartite network.

References

Pilosof, S., Porter, M., Pascual, M. et al. The multilayer nature of ecological networks. Nat Ecol Evol 1, 0101 (2017). https://doi.org/10.1038/s41559-017-0101

Examples

set.seed(12)
Net <- build_toy_net(11,15,16,0.2)
plot(Net)

set.seed(12)
Net <- build_toy_net(11,15,16,0.2,output_matrices=TRUE)
Net

Correlation of degree of connector nodes

Description

Calculating correlation of degree of connector nodes.

Usage

coid(
  network.or.subnet_mat1,
  subnet_mat2 = NULL,
  weighted = FALSE,
  weight_type = "shannon",
  method = "kendall"
)

Arguments

Details

In this package, a tripartite network contains three groups of nodes (a-nodes,b-nodes,c-nodes) and two subnetworks (P includes the links between a-nodes and b-nodes, Q includes the links between b-nodes and c-nodes). Connector nodes belong to b-nodes.

This function follows Sauve et al.(2016) to calculate the correlation of interaction degree (or weighted degree ) of connector nodes. For the binary network, connector nodes' degree is calculated in each subnetwork. For the quantitative network, Shannon diversity or sum of link strength for each connector node is calculated. Three correlation methods are supported. Kendall correlation is recommended following Sauve et al.(2016).

Two types of inputs network.or.subnet_mat1 can be processed:

• An "igraph" object with node attribute 'level' (0 for a-nodes, 1 for b-nodes,2 for c-nodes). If the input is a weighted network, the edge should have a 'weight' attribute.

• Or a matrix representing subnetwork P, and must be input with subnet_mat2 representing subnetwork Q.

If the inputs are two matrices, please make sure the rows of network.or.subnet_mat1 and subnet_mat2 both represent the groups of connector species,i.e, the b-group species. If both matrices have row names, then the function matches row names to produce connector nodes. Otherwise, row numbers are assigned to row names and matched. Within the two matrices (P and Q), columns represents a-group nodes and c-group nodes respectively. Elements in matrices are non-zero values if two nodes are linked with or without weights, and 0 otherwise.

Value

Return a numeric value representing correlation of interaction degree for connector nodes.

References

Sauve, A. M., Thebault, E., Pocock, M. J., & Fontaine, C. (2016). How plants connect pollination and herbivory networks and their contribution to community stability. Ecology, 97(4), 908-917.

Examples

## generate a random binary tripartite network
set.seed(12)
Net <- build_toy_net(11,15,16,0.2)
coid(Net)

## empirical network
data(PPH_Coltparkmeadow)
Net <- PPH_Coltparkmeadow
coid(Net)
set.seed(13)
library(igraph)
E(Net)$weight<-runif(length(E(Net)),0.1,1)#random weights assigned
coid(Net,weighted=TRUE)

##input as binary matrices, with row names.
set.seed(12)
md1 <- matrix(sample(c(0,1),8*11,replace=TRUE),8,11,
       dimnames = list(paste0("b",1:8),paste0("c",1:11)))
md2 <- matrix(sample(c(0,1),10*12,replace=TRUE),10,12,
       dimnames = list(paste0("b",1:10),paste0("a",1:12)))
coid(md1,md2)

##input as weighted matrices,with row numbers as row names.
set.seed(12)
mdw1 <- matrix(sample(c(rep(0,40),runif(48,0,1))),8,11)
mdw2 <- matrix(sample(c(rep(0,40),runif(80,0,1))),10,12)
coid(mdw1,mdw2)
coid(mdw1,mdw2,weighted=TRUE)
coid(mdw1,mdw2,weighted=TRUE, weight_type="sum")

Correlation of Interaction Similarity for connector nodes

Description

Calculating correlation of interaction similarity for connector nodes.

Usage

cois(
  network.or.subnet_mat1,
  subnet_mat2 = NULL,
  weighted = FALSE,
  method = "kendall"
)

Arguments

Details

In this package, a tripartite network contains three groups of nodes (a-nodes,b-nodes,c-nodes) and two subnetworks (P includes the links between a-nodes and b-nodes, Q includes the links between b-nodes and c-nodes). Connector nodes belong to b-nodes.

This function follows Sauve et al.(2016) to calculate the correlation of interaction similarity of connector nodes. When weighted=FALSE, the Jaccard similarity coefficients between all pairs of connector nodes in each subnetwork, defined as the number of shared interaction partners divided by the total number of interaction partners involved (i.e. the maximum number of interactions they could share). When weighted=TRUE, the quantitative similarities of all pairs of connector nodes in each subnetwork were measured using the generalised Jaccard coefficient. Three correlation methods are provided. Kendall correlation is recommended following Sauve et al.(2016).

Two types of inputs network.or.subnet_mat1 can be processed:

• An "igraph" object with node attribute 'level' (0 for a-nodes, 1 for b-nodes,2 for c-nodes). If the input is a weighted network, the edge should have a 'weight' attribute.

• Or a matrix representing subnetwork P, and must be input with subnet_mat2 representing subnetwork Q.

If the inputs are two matrices, please make sure the rows of network.or.subnet_mat1 and subnet_mat2 both represent the groups of connector species,i.e, the b-group species. If both matrices have row names, then the function matches row names to produce connector nodes. Otherwise, row numbers are assigned to row names and matched. Within the two matrices (P and Q), columns represents a-group nodes and c-group nodes respectively. Elements in matrices are non-zero values if two nodes are linked with or without weights, and 0 otherwise.

Value

Return a numeric value representing correlation of interaction similarity for connector species.

References

Sauve, A. M., Thebault, E., Pocock, M. J., & Fontaine, C. (2016). How plants connect pollination and herbivory networks and their contribution to community stability. Ecology, 97(4), 908-917.

Examples

## generate a random binary tripartite network
set.seed(12)
Net <- build_toy_net(11,15,16,0.2)
cois(Net)
## empirical network
data(PPH_Coltparkmeadow)
Net <- PPH_Coltparkmeadow
cois(Net)
set.seed(13)
library(igraph)
E(Net)$weight<-runif(length(E(Net)),0.1,1.5)#random weights assigned
cois(Net,weighted=TRUE)

##input as binary matrices,with row names.
md1 <- matrix(sample(c(0,1),8*11,replace=TRUE),8,11,
       dimnames = list(paste0("b",1:8),paste0("c",1:11)))
md2 <- matrix(sample(c(0,1),10*12,replace=TRUE),10,12,
       dimnames = list(paste0("b",1:10),paste0("a",1:12)))
cois(md1,md2)

##input as weighted matrices,with row numbers as row names.
mdw1 <- matrix(sample(c(rep(0,40),runif(48,0,1))),8,11)
mdw2 <- matrix(sample(c(rep(0,40),runif(80,0,1))),10,12)
cois(mdw1,mdw2)
cois(mdw1,mdw2,weighted=TRUE)

The proportion of hub connectors in the shared sets of nodes

Description

This function calculates the proportion of shared node hubs that are connectors (HC), i.e. the top x% of the nodes in the shared set of nodes with the highest degree that are connector nodes.

Usage

hc(network.or.subnet_mat1, subnet_mat2 = NULL, hubs = 0.2, weighted = FALSE)

Arguments

Details

In this package, a tripartite network contains three groups of nodes (a-nodes,b-nodes,c-nodes) and two subnetworks (P includes the links between a-nodes and b-nodes, Q includes the links between b-nodes and c-nodes). Connector nodes belong to b-nodes.

The function counts the proportion of connector nodes in connector node hubs (HC). The connector node hubs are the top x% of shared nodes with the highest degree (Dominguez-Garcia and Kefi 2024). The default x% is 20%. It always equals 1 if all b-nodes are connector nodes.

Two types of inputs network.or.subnet_mat1 can be processed:

• An "igraph" object with node attribute 'level' (0 for a-nodes, 1 for b-nodes, 2 for c-nodes). If the input is a weighted network, the edge should have a 'weight' attribute.

• Or a matrix representing subnetwork P, and must be input with subnet_mat2 representing subnetwork Q.

If the inputs are two matrices, please make sure the rows of network.or.subnet_mat1 and subnet_mat2 both represent the groups of connector species,i.e, the b-group species. If both matrices have row names, then the function matches row names to produce connector nodes. Otherwise, row numbers are assigned to row names and matched. Within the two matrices (P and Q), columns represents a-group nodes and c-group nodes respectively. Elements in matrices are non-zero values if two nodes are linked with or without weights, and 0 otherwise.

Value

Return a numeric value representing the proportion of connector nodes in node hubs.

References

Battiston, F., Nicosia, V. & Latora, V. (2014) Structural measures for multiplex networks. Physical Review E, 89, 032804.

Dominguez-Garcia, V. and Kefi, S. (2024). The structure and robustness of ecological networks with two interaction types. PLOS Computational Biology, 20(1), e1011770.

Guimera, R. & Amaral, L.A.N. (2005) Cartography of complex networks: modules and universal roles. Journal of Statistical Mechanics: Theory and Experiment, 2005, P02001.

Examples

## generate a random binary tripartite network
set.seed(12)
Net <- build_toy_net(11,15,16,0.2)
hc(Net)

#empirical network
data(PPH_Coltparkmeadow)
Net <- PPH_Coltparkmeadow
hc(Net)
set.seed(13)
library(igraph)
E(Net)$weight<-runif(length(E(Net)),0.1,1)#random weights assigned
pc(Net,weighted=TRUE)

##input as binary matrices,with row names.
set.seed(12)
md1 <- matrix(sample(c(0,1),5*8,replace=TRUE),5,8)
dimnames(md1) = list(paste0("b",1:5),paste0("c",1:8))
md2 <- matrix(sample(c(0,1),20*30,replace=TRUE),20,30)
dimnames(md2) = list(paste0("b",1:20),paste0("a",1:30))
hc(md1,md2)

##input as weighted matrices,with row numbers as row names.
set.seed(17)
mdw1 <- matrix(sample(c(rep(0,20),runif(20,0,1))),5,8)
mdw2 <- matrix(sample(c(rep(0,500),runif(100,0,1))),20,30)
hc(mdw1,mdw2)
hc(mdw1,mdw2,weighted=TRUE)

Count interconnection motifs

Description

Counting the frequencies of interconnection motifs for a tripartite interaction network.

Usage

icmotif_count(network.or.subnet_mat1, subnet_mat2 = NULL, weighted = FALSE)

Arguments

Details

In this package, a tripartite network contains three groups of nodes (a-nodes,b-nodes,c-nodes) and two subnetworks (P includes the links between a-nodes and b-nodes, Q includes the links between b-nodes and c-nodes). Connector nodes belong to b-nodes.
An interconnection motif is defined to comprise three sets of connected nodes: the connector nodes (belonging to b-nodes), the nodes in one subnetwork (belonging to a-nodes in the P subnetwork), and the nodes in the other subnetwork (belonging to c-nodes in the Q subnetwork). Each motif has maximumly 6 nodes, resulting in a total of 48 distinct motif forms. The algorithm for counting interconnection motifs is designed by extending the fast approach from Simmons et al.(2019), which uses mathematical operations directly on the bi-adjacency matrix. For interconnection motifs in tripartite networks with intra-guild interactions, please see ig_icmotif_count and ig_icmotif_role.

Two types of inputs network.or.subnet_mat1 can be processed:

• An "igraph" object with node attribute 'level' (0 for a-nodes, 1 for b-nodes,2 for c-nodes). If the input is a weighted network, the edge should have a 'weight' attribute.

• Or a matrix representing subnetwork P, and must be input with subnet_mat2 representing subnetwork Q.

If the inputs are two matrices, please make sure the rows of network.or.subnet_mat1 and subnet_mat2 both represent the groups of connector species,i.e, the b-group species. If both matrices have row names, then the function matches row names to produce connector nodes. Otherwise, row numbers are assigned to row names and matched. Within the two matrices (P and Q), columns represents a-group nodes and c-group nodes respectively. Elements in matrices are non-zero values if two nodes are linked with or without weights, and 0 otherwise.

Weighted networks
For weighted tripartite networks, the mean weight of a given motif is provided by averaging the weights of all the cases of a particular motif. The weight of a motif case is the arithmetic mean of the weights of its links, following Mora et al. (2018) and Simmons et al. (2019).

Value

Return a data.fame of the frequencies (and mean weight) of 48 interconnection motifs. See 'Multi_motif' for the forms.

References

Pilosof, S., Porter, M. A., Pascual, M., & KC)fi, S. (2017). The multilayer nature of ecological networks. Nature Ecology & Evolution, 1(4), 0101.

Mora, B.B., Cirtwill, A.R. and Stouffer, D.B. (2018). pymfinder: a tool for the motif analysis of binary and quantitative complex networks. bioRxiv, 364703.

Simmons, B. I., Sweering, M. J., Schillinger, M., Dicks, L. V., Sutherland, W. J., & Di Clemente, R. (2019). bmotif: A package for motif analyses of bipartite networks. Methods in Ecology and Evolution, 10(5), 695-701.

Examples

## generate a random tripartite network
set.seed(12)
Net <- build_toy_net(11,15,16,0.2)
icmotif_count(Net)

## empirical network
data(PPH_Coltparkmeadow)
Net <- PPH_Coltparkmeadow
icmotif_count(Net)
set.seed(13)
library(igraph)
E(Net)$weight<-runif(length(E(Net)),0.1,1)#random weights assigned
icmotif_count(Net, weighted=TRUE)


##input as binary matrices, with row names.
set.seed(12)
md1 <- matrix(sample(c(0,1),8*11,replace=TRUE),8,11)
dimnames(md1) = list(paste0("b",1:8),paste0("c",1:11))
md2 <- matrix(sample(c(0,1),10*12,replace=TRUE),10,12)
dimnames(md2) = list(paste0("b",1:10),paste0("a",1:12))
icmotif_count(md1,md2)

##input as weighted matrices,with row numbers as row names.
set.seed(12)
mdw1 <- matrix(sample(c(rep(0,40),runif(48,0,1))),8,11)
mdw2 <- matrix(sample(c(rep(0,40),runif(80,0,1))),10,12)
icmotif_count(mdw1,mdw2,weighted=TRUE)

Count the roles of connector nodes defined by interconnection motifs

Description

Count the roles of connector nodes defined by interconnection motifs in a tripartite network.

Usage

icmotif_role(network.or.subnet_mat1, subnet_mat2 = NULL, weighted = FALSE)

Arguments

Details

In this package, a tripartite network contains three groups of nodes (a-nodes,b-nodes,c-nodes) and two subnetworks (P includes the links between a-nodes and b-nodes, Q includes the links between b-nodes and c-nodes). Connector nodes belong to b-nodes.
An interconnection motif is defined to comprise three sets of connected nodes: the connector nodes (belonging to b-nodes), the nodes in one subnetwork (belonging to a-nodes in the P subnetwork), and the nodes in the other subnetwork (belonging to c-nodes in the Q subnetwork). Each motif has maximumly 6 nodes, resulting in a total of 48 distinct motif forms. The algorithm for counting interconnection motifs is designed by extending the fast approach from Simmons et al.(2019), which uses mathematical operations directly on the bi-adjacency matrix. For interconnection motifs in tripartite networks with intra-guild interactions, please see ig_icmotif_count and ig_icmotif_role.

Two types of inputs network.or.subnet_mat1 can be processed:

• An "igraph" object with node attribute 'level' (0 for a-nodes, 1 for b-nodes,2 for c-nodes). If the input is a weighted network, the edge should have a 'weight' attribute.

• Or a matrix representing subnetwork P, and must be input with subnet_mat2 representing subnetwork Q.

If the inputs are two matrices, please make sure the rows of network.or.subnet_mat1 and subnet_mat2 both represent the groups of connector species,i.e, the b-group species. If both matrices have row names, then the function matches row names to produce connector nodes. Otherwise, row numbers are assigned to row names and matched. Within the two matrices (P and Q), columns represents a-group nodes and c-group nodes respectively. Elements in matrices are non-zero values if two nodes are linked with or without weights, and 0 otherwise.

Weighted networks
For weighted tripartite networks, the mean weight of the motif occurrence (i.e., a motif occurrence isomorphic to a particular motif form) is provided for a given node with a given role, following Mora et al. (2018) and Simmons et al. (2019).

Value

For binary networks, return a matrix with elements representing the number of times each connector node plays for each unique role within interconnection motifs; for weighted networks, the matrix element represents the mean weight of the motif occurrences where the node exists.

References

#' Mora, B.B., Cirtwill, A.R. and Stouffer, D.B. (2018). pymfinder: a tool for the motif analysis of binary and quantitative complex networks. bioRxiv, 364703.

Simmons, B. I., Sweering, M. J., Schillinger, M., Dicks, L. V., Sutherland, W. J., & Di Clemente, R. (2019). bmotif: A package for motif analyses of bipartite networks. Methods in Ecology and Evolution, 10(5), 695-701.

Examples

## generate a random tripartite network
set.seed(12)
Net <- build_toy_net(11,15,16,0.2)
icmotif_role(Net)

## empirical network
data(PPH_Coltparkmeadow)
Net <- PPH_Coltparkmeadow
icmotif_role(Net)
set.seed(13)
library(igraph)
E(Net)$weight<-runif(length(E(Net)),0.1,1)#random weights assigned
icmotif_role(Net, weighted=TRUE)


##input as binary matrices, with row names.
set.seed(12)
md1 <- matrix(sample(c(0,1),8*11,replace=TRUE),8,11)
dimnames(md1) = list(paste0("b",1:8),paste0("c",1:11))
md2 <- matrix(sample(c(0,1),10*12,replace=TRUE),10,12)
dimnames(md2) = list(paste0("b",1:10),paste0("a",1:12))
icmotif_role(md1,md2)


##input as weighted matrices,with row numbers as row names.
set.seed(12)
mdw1 <- matrix(sample(c(rep(0,40),runif(48,0,1))),8,11)
mdw2 <- matrix(sample(c(rep(0,40),runif(80,0,1))),10,12)
icmotif_role(mdw1,mdw2,weighted=TRUE)

Degree of diagonal dominance

Description

Calculating the degree of diagonal dominance for a tripartite network with intra-guild interactions.

Usage

ig_ddom(mat)

Arguments

Details

To calculate degree of diagonal dominance, the input should be a block matrix (M) to represent three groups of nodes (a-nodes, b-nodes and c-nodes): three intra-guild interaction matrices (m_{aa},m_{bb},m_{cc}), two inter-guild matrices of a and b-nodes (m_{ab},m_{ba}), and two inter-guild matrices of b- and c-nodes(m_{bc},m_{cb}).

\left( \begin{array}{ccc} m_{aa} & m_{ab} & 0 \\ m_{ba} & m_{bb} & m_{bc} \\ 0 & m_{cb} & m_{cc} \end{array} \right)

This function follows the definition by Garcia-Callejas et al (2023). Diagonal dominance is a matrix property indicating diagonal elements are larger than the sum of non-diagonal elements. In a tripartite network with intra-guild interactions, Garcia-Callejas et al. propose a continuous diagonal dominance index by calculating the difference between the diagonal and the sum of non-diagonal elements for each node. The average degree of diagonal dominance (d) is calculated by averaging all the nodes' values.

d = \frac{1}{S} \sum_{i \in S} \left( M_{ii} - \sum_{\substack{j \in S \\ j \neq i}} M_{ij} \right)

Where S is the number of species in the tripartite network M.

Value

Return the average degree of diagonal dominance and a vector of diagonal dominance indices for all nodes.

References

Garcia-Callejas, D., Godoy, O., Buche, L., Hurtado, M., Lanuza, J.B., Allen-Perkins, A. et al. (2023) Non-random interactions within and across guilds shape the potential to coexist in multi-trophic ecological communities. Ecology Letters, 26, 831-842.

Examples

##A toy tripartite network with intra-guild negative interactions:
##Inter-guild mutualistic interactions and inter-guild antagonistic interactions.
set.seed(12)
## 4 a-nodes, 5 b-nodes, and 3 c-nodes.

##Intra-guild interaction matrices
mat_aa<-matrix(runif(16,-0.8,-0.2),4,4)
mat_bb<-matrix(runif(25,-0.8,-0.2),5,5)
mat_cc<-matrix(runif(9,-0.8,-0.2),3,3)

##Inter-guild interaction matrices between a- and b-nodes.
mat_ab<-mat_ba<-matrix(sample(c(rep(0,8),runif(12,0,0.5))),4,5,byrow=TRUE)
mat_ba[mat_ba>0]<-runif(12,0,0.5);mat_ba<-t(mat_ba)

##inter-guild interaction matrices between b- and c-nodes.
mat_cb<-mat_bc<-matrix(sample(c(rep(0,8),runif(7,0,0.5))),3,5,byrow=TRUE)
mat_bc[mat_bc>0]<-runif(7,0,0.5);mat_bc<--t(mat_bc)
mat<-rbind(cbind(mat_aa,mat_ab,matrix(0,4,3)),cbind(mat_ba,mat_bb,mat_bc))
mat<-rbind(mat,cbind(matrix(0,3,4),mat_cb,mat_cc))

##set the node names
rownames(mat)<-c(paste0("a",1:4),paste0("b",1:5),paste0("c",1:3))
colnames(mat)<-c(paste0("a",1:4),paste0("b",1:5),paste0("c",1:3))
diag(mat)<--1 #assume -1 for diagonal elements

##Visualization of the block matrix.
library(plot.matrix)
pal <- colorRampPalette(c("darkblue", "lightblue", "white", "pink", "darkred"))(100)
par(mar=c(5,5,5,5));plot(mat,col = pal,
    breaks = seq(-max(abs(mat)), max(abs(mat)), length.out = 101),
    main = "Matrix visualization")
    clip(x1 = 0.5,# Left boundary
    x2 = ncol(mat) + 0.5, # Right boundary
    y1 = 0.5,            # Top boundary
    y2 = nrow(mat) + 0.5  )
    abline(v = c(4.5,9.5), h = c(3.5,8.5), lwd = 3, col = "black")

ig_ddom(mat)

Count interconnection motifs for tripartite networks with intra-guild interactions

Description

Counting the frequencies of interconnection motifs for a tripartite interaction network with intra-guild interactions.

Usage

ig_icmotif_count(mat, guilds, weighted = FALSE)

Arguments

Details

This function is designed for tripartite networks with intra-guild interactions. The input network should be nput as a block matrix (M) to represent three groups of nodes (a-nodes, b-nodes and c-nodes): three intra-guild interaction matrices (m_{aa},m_{bb},m_{cc}), two inter-guild matrices of a and b-nodes (m_{ab},m_{ba}), and two inter-guild matrices of b- and c-nodes(m_{bc},m_{cb}).

\left( \begin{array}{ccc} m_{aa} & m_{ab} & 0 \\ m_{ba} & m_{bb} & m_{bc} \\ 0 & m_{cb} & m_{cc} \end{array} \right)

guilds should be a vector of the same length as the row of mat like c("a","a"..."b","b"..."c","c"..)

Interconnection motifs in tripartite networks with intra-guild interactions
An interconnection motif is defined to comprise three sets of connected nodes: the connector nodes (belonging to b-nodes), the nodes in one subnetwork (belonging to a-nodes in the P subnetwork), and the nodes in the other subnetwork (belonging to c-nodes in the Q subnetwork). Each guild has two nodes at most, resulting in a total of 107 distinct motif forms. The algorithm for counting interconnection motifs is designed by extending the fast approach from Simmons et al.(2019). For interconnection motifs in tripartite networks without intra-guild interactions, please see icmotif_count and icmotif_role.

Weighted networks
For weighted tripartite networks, the mean weight of a given motif is provided by averaging the weights of all motif occurrences isomorphic to the motif. The weight of a motif occurrence is the arithmetic mean of the weights of its links, following Mora et al. (2018) and Simmons et al. (2019).

Value

Return a matrix of counts (and mean weight) of 107 interconnection motifs.

References

Garcia-Callejas, D., Godoy, O., Buche, L., Hurtado, M., Lanuza, J.B., Allen-Perkins, A. et al. (2023) Non-random interactions within and across guilds shape the potential to coexist in multi-trophic ecological communities. Ecology Letters, 26, 831-842.

Pilosof, S., Porter, M. A., Pascual, M., & KC)fi, S. (2017). The multilayer nature of ecological networks. Nature Ecology & Evolution, 1(4), 0101.

Mora, B.B., Cirtwill, A.R. and Stouffer, D.B. (2018). pymfinder: a tool for the motif analysis of binary and quantitative complex networks. bioRxiv, 364703.

Simmons, B. I., Sweering, M. J., Schillinger, M., Dicks, L. V., Sutherland, W. J., & Di Clemente, R. (2019). bmotif: A package for motif analyses of bipartite networks. Methods in Ecology and Evolution, 10(5), 695-701.

Examples

##A toy tripartite network with intra-guild negative interactions:
##Inter-guild mutualistic interactions and inter-guild antagonistic interactions.
set.seed(12)
##4 a-nodes, 5 b-nodes, and 3 c-nodes

##Intra-guild interaction matrices
mat_aa<-matrix(runif(16,-0.8,-0.2),4,4)
mat_aa <- mat_aa + t(mat_aa); diag(mat_aa) <- 0
mat_bb<-matrix(runif(25,-0.8,-0.2),5,5)
mat_bb <- mat_bb + t(mat_bb); diag(mat_bb) <- 0
mat_cc<-matrix(runif(9,-0.8,-0.2),3,3)
mat_cc <- mat_cc + t(mat_cc); diag(mat_cc) <- 0

##Inter-guild interaction matrices between a- and b-nodes.
mat_ab<-mat_ba<-matrix(sample(c(rep(0,8),runif(12,0,0.5))),4,5,byrow=TRUE)
mat_ba<-t(mat_ba)

##Inter-guild interaction matrices between b- and c-nodes.
mat_cb<-mat_bc<-matrix(sample(c(rep(0,8),runif(7,0,0.5))),3,5,byrow=TRUE)
mat_bc<--t(mat_bc)
mat<-rbind(cbind(mat_aa,mat_ab,matrix(0,4,3)),cbind(mat_ba,mat_bb,mat_bc))
mat<-rbind(mat,cbind(matrix(0,3,4),mat_cb,mat_cc))

##set the node names
rownames(mat)<-c(paste0("a",1:4),paste0("b",1:5),paste0("c",1:3))
colnames(mat)<-c(paste0("a",1:4),paste0("b",1:5),paste0("c",1:3))

#mat[mat!=0] <- 1
myguilds=c(rep("a",4),rep("b",5),rep("c",3))
ig_icmotif_count(mat,guilds=myguilds,TRUE)

Count the roles of connector nodes defined by interconnection motifs with intra-guild interactions

Description

Count the roles of connector nodes defined by interconnection motifs in a tripartite network with intra-guild interactions.

Usage

ig_icmotif_role(mat, guilds, weighted = FALSE)

Arguments

Details

This function is designed for tripartite networks with intra-guild interactions. The input network should be nput as a block matrix (M) to represent three groups of nodes (a-nodes, b-nodes and c-nodes): three intra-guild interaction matrices (m_{aa},m_{bb},m_{cc}), two inter-guild matrices of a and b-nodes (m_{ab},m_{ba}), and two inter-guild matrices of b- and c-nodes(m_{bc},m_{cb}).

\left( \begin{array}{ccc} m_{aa} & m_{ab} & 0 \\ m_{ba} & m_{bb} & m_{bc} \\ 0 & m_{cb} & m_{cc} \end{array} \right)

guilds should be a vector of the same length as the row of mat like c("a","a"..."b","b"..."c","c"..)

Interconnection motifs in tripartite networks with intra-guild interactions
An interconnection motif is defined to comprise three sets of connected nodes: the connector nodes (belonging to b-nodes), the nodes in one subnetwork (belonging to a-nodes in the P subnetwork), and the nodes in the other subnetwork (belonging to c-nodes in the Q subnetwork). Each guild has two nodes at most, resulting in a total of 107 distinct motif forms. The algorithm for counting interconnection motifs is designed by extending the fast approach from Simmons et al.(2019). Currently, only the roles in first 35 motifs are provided (see ). For interconnection motifs in tripartite networks without intra-guild interactions, please see icmotif_count and icmotif_role.

Weighted networks
For weighted tripartite networks, the mean weight of the motif occurrence (i.e., a motif occurrence isomorphic to a particular motif form) is provided for a given node with a given role, following Mora et al. (2018) and Simmons et al. (2019).

Value

For binary networks, return a matrix with elements representing the number of times each connector node plays for each unique role within interconnection motifs; for weighted networks, the matrix element represents the mean weight of the motif occurrences where the node exists.

References

Garcia-Callejas, D., Godoy, O., Buche, L., Hurtado, M., Lanuza, J.B., Allen-Perkins, A. et al. (2023) Non-random interactions within and across guilds shape the potential to coexist in multi-trophic ecological communities. Ecology Letters, 26, 831-842.

Pilosof, S., Porter, M. A., Pascual, M., & KC)fi, S. (2017). The multilayer nature of ecological networks. Nature Ecology & Evolution, 1(4), 0101.

Mora, B.B., Cirtwill, A.R. and Stouffer, D.B. (2018). pymfinder: a tool for the motif analysis of binary and quantitative complex networks. bioRxiv, 364703.

Simmons, B. I., Sweering, M. J., Schillinger, M., Dicks, L. V., Sutherland, W. J., & Di Clemente, R. (2019). bmotif: A package for motif analyses of bipartite networks. Methods in Ecology and Evolution, 10(5), 695-701.

Examples

##A toy tripartite network with intra-guild negative interactions:
##Inter-guild mutualistic interactions and inter-guild antagonistic interactions.
set.seed(12)
##4 a-nodes, 5 b-nodes, and 3 c-nodes

##Intra-guild interaction matrices
mat_aa<-matrix(runif(16,-0.8,-0.2),4,4)
mat_aa <- mat_aa + t(mat_aa); diag(mat_aa) <- 0
mat_bb<-matrix(runif(25,-0.8,-0.2),5,5)
mat_bb <- mat_bb + t(mat_bb); diag(mat_bb) <- 0
mat_cc<-matrix(runif(9,-0.8,-0.2),3,3)
mat_cc <- mat_cc + t(mat_cc); diag(mat_cc) <- 0

##Inter-guild interaction matrices between a- and b-nodes.
mat_ab<-mat_ba<-matrix(sample(c(rep(0,8),runif(12,0,0.5))),4,5,byrow=TRUE)
mat_ba<-t(mat_ba)

##Inter-guild interaction matrices between b- and c-nodes.
mat_cb<-mat_bc<-matrix(sample(c(rep(0,8),runif(7,0,0.5))),3,5,byrow=TRUE)
mat_bc<--t(mat_bc)
mat<-rbind(cbind(mat_aa,mat_ab,matrix(0,4,3)),cbind(mat_ba,mat_bb,mat_bc))
mat<-rbind(mat,cbind(matrix(0,3,4),mat_cb,mat_cc))

##Set the node names
rownames(mat)<-c(paste0("a",1:4),paste0("b",1:5),paste0("c",1:3))
colnames(mat)<-c(paste0("a",1:4),paste0("b",1:5),paste0("c",1:3))

myguilds=c(rep("a",4),rep("b",5),rep("c",3))
ig_icmotif_role(mat,guilds=myguilds)
ig_icmotif_role(mat,guilds=myguilds,TRUE)

Intra-guild and inter_guild interaction overlap

Description

Calculating species-level intra-guild and inter_guild interaction overlap for a tripartite network with intra-guild interactions.

Usage

ig_overlap_guild(mat, guilds, method = "horn")

Arguments

Details

The input is a block matrix (M) to represent interactions among three groups of species (a-nodes, b-nodes and c-nodes): three intra-guild interaction matrices (m_{aa},m_{bb},m_{cc}), two inter-guild matrices of a and b-nodes (m_{ab},m_{ba} with symmetric links), and two inter-guild matrices of b- and c-nodes(m_{bc},m_{cb} with symmetric links). Connector species belong to b-nodes.

\left( \begin{array}{ccc} m_{aa} & m_{ab} & 0 \\ m_{ba} & m_{bb} & m_{bc} \\ 0 & m_{cb} & m_{cc} \end{array} \right)

guilds should be a vector of the same length as the row of mat like c("a","a"..."b","b"..."c","c"..)

This function follows the definition by Garcia-Callejas et al (2023). Species-level interaction overlap is derived from the overlap between each pair of species, calculated using the dissimilarity index (d_{ij}, default to Morisita-Horn index) as in the R package vegan. The net overlap of species (o_{i})is represented by the sum of pairwise overlaps with every other species:

o_i = \sum_{j \in S} (1 - d_{ij})

Value

Return a list including three species-level intra-guild overlap vectors for a-, b- and c-nodes (a_intra_overlap,b_intra_overlap,c_intra_overlap), two vectors of inter-guild interaction overlap for a-nodes (a_inter_b_overlap) and b-nodes (b_inter_a_overlap), and two vectors of inter-guild interaction overlap for b-nodes (b_inter_c_overlap) and c-nodes(c_inter_b_overlap).

References

Garcia-Callejas, D., Godoy, O., Buche, L., Hurtado, M., Lanuza, J.B., Allen-Perkins, A. et al. (2023) Non-random interactions within and across guilds shape the potential to coexist in multi-trophic ecological communities. Ecology Letters, 26, 831-842.

Examples

##A toy tripartite network with intra-guild negative interactions,
##Inter-guild mutualistic interactions and inter-guild antagonistic interactions.
set.seed(12)
##4 a-nodes, 5 b-nodes, and 3 c-nodes

##Intra-guild interaction matrices
mat_aa<-matrix(runif(16,-0.8,-0.2),4,4)
mat_bb<-matrix(runif(25,-0.8,-0.2),5,5)
mat_cc<-matrix(runif(9,-0.8,-0.2),3,3)

##Inter-guild interaction matrices between a- and b-nodes.
mat_ab<-mat_ba<-matrix(sample(c(rep(0,8),runif(12,0,0.5))),4,5,byrow=TRUE)
mat_ba[mat_ba>0]<-runif(12,0,0.5);mat_ba<-t(mat_ba)

##Inter-guild interaction matrices between b- and c-nodes.
mat_cb<-mat_bc<-matrix(sample(c(rep(0,8),runif(7,0,0.5))),3,5,byrow=TRUE)
mat_bc[mat_bc>0]<-runif(7,0,0.5);mat_bc<--t(mat_bc)
mat<-rbind(cbind(mat_aa,mat_ab,matrix(0,4,3)),cbind(mat_ba,mat_bb,mat_bc))
mat<-rbind(mat,cbind(matrix(0,3,4),mat_cb,mat_cc))

##Set the node names
rownames(mat)<-c(paste0("a",1:4),paste0("b",1:5),paste0("c",1:3))
colnames(mat)<-c(paste0("a",1:4),paste0("b",1:5),paste0("c",1:3))
diag(mat)<--1 #assume -1 for diagonal elements

##Visualization of this block matrix.
library(plot.matrix)
pal <- colorRampPalette(c("darkblue", "lightblue", "white", "pink", "darkred"))(100)
par(mar=c(5,5,5,5));plot(mat,col = pal,
    breaks = seq(-max(abs(mat)), max(abs(mat)), length.out = 101),
    main = "Matrix visualization")
    clip(x1 = 0.5,# Left boundary
    x2 = ncol(mat) + 0.5, # Right boundary
    y1 = 0.5,            # Top boundary
    y2 = nrow(mat) + 0.5  )
    abline(v = c(4.5,9.5), h = c(3.5,8.5), lwd = 3, col = "black")

myguilds=c(rep("a",4),rep("b",5),rep("c",3))
ig_overlap_guild(mat,guilds=myguilds)

Interconnection centrality for connector nodes in a tripartite network

Description

Calculating interconnection centrality (degree, betweenness and closeness) specified for connector nodes in a tripartite network.

Usage

node_icc(network.or.subnet_mat1, subnet_mat2 = NULL, weighted = FALSE)

Arguments

Details

In this package, a tripartite network contains three groups of nodes (a-nodes,b-nodes,c-nodes) and two subnetworks (P includes the links between a-nodes and b-nodes, Q includes the links between b-nodes and c-nodes). Connector nodes belong to b-nodes. This function calculates interconnection degree, betweenness and closeness centrality for connector nodes.

• For binary networks, Interconnection degree of each connector species is defined as the product of its degree values from two subnetworks.

• Interconnection betweenness is defined by the number of shortest paths from a-nodes to c-nodes going through connector species, \sum_{\substack{i \in a,\, j \in c \\ i \neq j}} \frac{g_{ivj}}{g_{ij}}, where g_{ij} is the total number of shortest paths between node i from a-nodes and j from c-nodes while g_{ivj} is the number of those shortest paths which pass through connector species v.

• Interconnection closeness is defined by the inverse of the sum of distances from connector species to both a-nodes and c-nodes, 1/(\sum_{v \neq i, i \in a \cup c} d_{vi}), where d_{vi} is the distance between connector species v and species i from a-nodes and c-nodes. For weighted networks, interaction strengths are used in the calculation of weighted degree, shorest path, and distance.

network.or.subnet_mat1 and subnet_mat2

Two types of inputs network.or.subnet_mat1 can be processed:

• An "igraph" object with node attribute 'level' (0 for a-nodes, 1 for b-nodes,2 for c-nodes). If the input is a weighted network, the edge should have a 'weight' attribute.

• Or a matrix representing subnetwork P, and must be input with subnet_mat2 representing subnetwork Q.

If the inputs are two matrices, please make sure the rows of network.or.subnet_mat1 and subnet_mat2 both represent the groups of connector species,i.e, the b-group species. If both matrices have row names, then the function matches row names to produce connector nodes. Otherwise, row numbers are assigned as row names.

Value

Return a data frame with interconnection degree, betweenness and closeness for connector nodes.

Examples

## generate a random tripartite network
set.seed(12)
Net <- build_toy_net(11,15,16,0.2)

data(PPH_Coltparkmeadow)
Net <- PPH_Coltparkmeadow
node_icc(Net)
set.seed(13)
library(igraph)
E(Net)$weight<-runif(length(E(Net)),0.1,1)#random weights assigned
node_icc(Net,weighted=TRUE)

##input as binary matrices,with row names.
set.seed(12)
md1 <- matrix(sample(c(0,1),5*8,replace=TRUE),5,8,
       dimnames = list(paste0("b",1:5),paste0("c",1:8)))
md2 <- matrix(sample(c(0,1),20*30,replace=TRUE),20,30,
       dimnames = list(paste0("b",1:20),paste0("a",1:30)))
node_icc(md1,md2)

##input as weighted matrices,with row numbers as row names.
set.seed(17)
mdw1 <- matrix(sample(c(rep(0,20),runif(20,0,1))),5,8)
mdw2 <- matrix(sample(c(rep(0,500),runif(100,0,1))),20,30)
node_icc(mdw1,mdw2)
node_icc(mdw1,mdw2,weighted=TRUE)

Participation coefficient

Description

Calculating participation coefficient for connector nodes.

Usage

pc(network.or.subnet_mat1, subnet_mat2 = NULL, weighted = FALSE)

Arguments

Details

In this package, a tripartite network contains three groups of nodes (a-nodes,b-nodes,c-nodes) and two subnetworks (P includes the links between a-nodes and b-nodes, Q includes the links between b-nodes and c-nodes). Connector nodes belong to b-nodes.

The pc function calculates the participation coefficient following Dominguez-Garcia and Kefi (2024). For each connector node i, pc_{i} is calculated as two times the ratio between the lowest degree in both interaction subnetworks divided by the total degree of the node (2\frac{d_{lowest}}{d_{total}}). Hence, the participation coefficient for all connector nodes (PC_{c}) is represented by the average value of all pc_{i}.

Two types of inputs network.or.subnet_mat1 can be processed:

• An "igraph" object with node attribute 'level' (0 for a-nodes, 1 for b-nodes,2 for c-nodes). If the input is a weighted network, the edge should have a 'weight' attribute.

• Or a matrix representing subnetwork P, and must be input with subnet_mat2 representing subnetwork Q.

If the inputs are two matrices, please make sure the rows of network.or.subnet_mat1 and subnet_mat2 both represent the groups of connector species,i.e, the b-group species. If both matrices have row names, then the function matches row names to produce connector nodes. Otherwise, row numbers are assigned to row names and matched. Within the two matrices (P and Q), columns represents a-group nodes and c-group nodes respectively. Elements in matrices are non-zero values if two nodes are linked with or without weights, and 0 otherwise.

Value

Return a numeric value representing participation coefficient for connector species.

References

Battiston, F., Nicosia, V. & Latora, V. (2014) Structural measures for multiplex networks. Physical Review E, 89, 032804.

Dominguez-Garcia, V. and Kefi, S. (2024). The structure and robustness of ecological networks with two interaction types. PLOS Computational Biology, 20(1), e1011770.

Guimera, R. & Amaral, L.A.N. (2005) Cartography of complex networks: modules and universal roles. Journal of Statistical Mechanics: Theory and Experiment, 2005, P02001.

Examples

## generate a random binary tripartite network
set.seed(12)
Net <- build_toy_net(11,15,16,0.2)
pc(Net)

##empirical network
data(PPH_Coltparkmeadow)
Net <- PPH_Coltparkmeadow
pc(Net)
set.seed(13)
library(igraph)
E(Net)$weight<-runif(length(E(Net)),0.1,1)#random weights assigned
pc(Net,weighted=TRUE)

##input as binary matrices,with row names.
md1 <- matrix(sample(c(0,1),8*11,replace=TRUE),8,11)
dimnames(md1) = list(paste0("b",1:8),paste0("c",1:11))
md2 <- matrix(sample(c(0,1),10*12,replace=TRUE),10,12)
dimnames(md2) = list(paste0("b",1:10),paste0("a",1:12))
pc(md1,md2)

##input as weighted matrices,with row numbers as row names.
mdw1 <- matrix(sample(c(rep(0,40),runif(48,0,1))),8,11)
mdw2 <- matrix(sample(c(rep(0,40),runif(80,0,1))),10,12)
pc(mdw1,mdw2)
pc(mdw1,mdw2,weighted=TRUE)

Proportion of connector nodes

Description

Calculating the proportion of connector nodes in the shared set of nodes.

Usage

poc(network.or.subnet_mat1, subnet_mat2 = NULL)

Arguments

Details

In this package, a tripartite network contains three groups of nodes (a-nodes,b-nodes,c-nodes) and two subnetworks (P includes the links between a-nodes and b-nodes, Q includes the links between b-nodes and c-nodes). Connector nodes belong to b-nodes.

Two types of inputs network.or.subnet_mat1 can be processed:

• An "igraph" object with node attribute 'level' (0 for a-nodes, 1 for b-nodes,2 for c-nodes).

• Or a matrix representing subnetwork P, and must be input with subnet_mat2 representing subnetwork Q.

If the inputs are two matrices, please make sure the rows of network.or.subnet_mat1 and subnet_mat2 both represent the groups of connector species,i.e, the b-group species. If both matrices have row names, then the function matches row names to produce connector nodes. Otherwise, row numbers are assigned to row names and matched. Within the two matrices (P and Q), columns represents a-group nodes and c-group nodes respectively. Elements in matrices are non-zero values if two nodes are linked with or without weights, and 0 otherwise.

Value

Return a vector of POC (proportion of connector nodes), number of connector nodes and number of shared species set.

References

Battiston, F., Nicosia, V. & Latora, V. (2014) Structural measures for multiplex networks. Physical Review E, 89, 032804.

Dominguez-Garcia, V. and Kefi, S. (2024). The structure and robustness of ecological networks with two interaction types. PLOS Computational Biology, 20(1), e1011770.

Guimera, R. & Amaral, L.A.N. (2005) Cartography of complex networks: modules and universal roles. Journal of Statistical Mechanics: Theory and Experiment, 2005, P02001.

Examples

## generate a random binary tripartite network
set.seed(12)
Net <- build_toy_net(11,15,16,0.2)
poc(Net)

## empirical network
data(PPH_Coltparkmeadow)
Net <- PPH_Coltparkmeadow
poc(Net)

##input as binary matrices,with row names.
md1 <- matrix(sample(c(0,1),8*11,replace=TRUE),8,11)
dimnames(md1) = list(paste0("b",1:8),paste0("c",1:11))
md2 <- matrix(sample(c(0,1),10*12,replace=TRUE),10,12)
dimnames(md2) = list(paste0("b",1:10),paste0("a",1:12))
poc(md1,md2)

##input as weighted matrices,with row numbers as row names.
mdw1 <- matrix(sample(c(rep(0,40),runif(48,0,1))),8,11)
mdw2 <- matrix(sample(c(rep(0,40),runif(80,0,1))),10,12)
poc(mdw1,mdw2)

Similarity of roles of interconnecting species

Description

The average of Similarity of 65 roles of interconnecting species of tripartite network.

Usage

role_sim(spe_role_mat)

Arguments

Details

spe_role_mat

Should acquire from function motif_role.

Value

Return a numeric value.

Examples

## generate a random tripartite network
set.seed(12)
Net <- build_toy_net(11,15,16,0.2)
role_sim(icmotif_role(Net))

data(PPH_Coltparkmeadow)
Net <- PPH_Coltparkmeadow
role_sim(icmotif_role(Net))

Generates null models for tripartite network

Description

A wrapper function to generate different null models for binary and quantitative tripartite networks

Usage

tri_null(
  trinet,
  null_N = 100,
  null_type = c("sauve", "sub_P", "sub_Q", "both_sub"),
  sub_method
)

Arguments

Details

In this package, a tripartite network contains three groups of nodes (a-nodes, b-nodes, c-nodes) and two subnetworks (P includes the links between a-nodes and b-nodes, Q includes the links between b-nodes and c-nodes). Connector nodes belong to b-nodes.

null_type

• "sub_P", "sub_Q" and "both_sub" use the null model algorithms from vegan::commsim to shuffle single subnetwork or both of them.

• "sauve" rearranges connector species without changing subnetworks, following sauve et al.(2016).

Value

Return a list of null models of a tripartite network.

References

Sauve, A. M., Thebault, E., Pocock, M. J., & Fontaine, C. (2016). How plants connect pollination and herbivory networks and their contribution to community stability. Ecology, 97(4), 908-917.

Examples

## generate a random tripartite network
set.seed(12)
Net <- build_toy_net(11,15,16,0.2)

data(PPH_Coltparkmeadow)
Net <- PPH_Coltparkmeadow

set.seed(123)
tri_null_list<-tri_null(Net,null_type="both_sub",sub_method="r00")
set.seed(123)
tri_null_list<-tri_null(Net,null_type="sauve")

Transforming two matrices into an igraph object.

Description

Transform two matrices into an igraph object.

Usage

trigraph_from_mat(mat1, mat2, weighted = FALSE)

Arguments

Details

In this package, a tripartite network contains three groups of nodes (a-nodes,b-nodes,c-nodes) and two subnetworks (P includes the links between a-nodes and b-nodes, Q includes the links between b-nodes and c-nodes). Connector nodes belong to b-nodes. Please make sure the rows of mat1 and mat2 both represent the groups of connector species,i.e, the b-group species. If both matrices have row names, then the function matches row names to define connector nodes. Otherwise, row numbers are assigned to row names and matched, which might produce an incorrected network. Within the two matrices (P and Q), columns represents a-group nodes and c-group nodes respectively. Elements in matrices are non-zero values if two nodes are linked with or without weights, and 0 otherwise.

Value

Return a network of type "igraph".

Examples

set.seed(12)
MAT <- build_toy_net(11,22,21,0.2,output_matrices=TRUE)
M <- trigraph_from_mat((MAT[[3]]),MAT[[4]])
M