Skip to contents

Calculates the QSS difference between the full network and the network minus a group of species

Source: R/calcExtinctionIndices.r
calc_QSS_extinction_dif_grp.Rd

The Quasi-sign stability is estimated with the maximum eigenvalue of the community matrix (Jacobian) and characterizes the linear stability of the network. This uses the function calc_QSS() so it can take into account the interaction strength if weights are present. The comparison is made using the Anderson-Darling test with the function kSamples::ad.test() and the Kolmogorov-Smirnov test stats::ks.test(), both the p-values are reported as a measure of strength of the difference.

Usage

calc_QSS_extinction_dif_grp(
  g,
  sp_list,
  nsim = 1000,
  ncores = 4,
  istrength = FALSE
)

Arguments

g

igraph network

sp_list

list with the group pf species/nodes we will delete for the comparison

nsim

number of simulations to calculate QSS

ncores

number of cores used to perform the operation

istrength

if TRUE takes the weight attribute of the network as interaction strength to calculate QSS.

Value

a data.frame with:

  • Size of the network with deleted nodes

  • Connectance of the network with deleted nodes

  • Number of components of the network with deleted nodes

    • QSS of the complete network

    • QSS of the network with the deleted nodes

    • difference between the two previous QSS

Examples

if (FALSE) { # \dontrun{
g <- netData[[1]]

# Generate random weights
#
V(g)$weight <-  runif(vcount(g))

# Without interaction strength
#
calc_QSS_extinction_dif_grp(g,V(g)$name[1:3],nsim=10,istrength = FALSE)

# With interaction strength
#
calc_QSS_extinction_dif_grp(g,V(g)$name[1:3],nsim=10,istrength = TRUE)
} # }