tests/testthat/test_netCompare_diffnet.R

In stefpeschel/NetCoMi: Network Construction and Comparison for Microbiome Data

set.seed(123456)
data("amgut1.filt")

groups_diss <- sample(0:1, ncol(amgut1.filt), replace = TRUE)
groups_asso <- sample(0:1, nrow(amgut1.filt), replace = TRUE)

context("test netCompare")

assonet2 <- netConstruct(amgut1.filt, group = groups_asso,
                         filtTax = "highestVar",
                         filtTaxPar = list(highestVar = 50),
                         filtSamp = "totalReads",
                         filtSampPar = list(totalReads = 1000),
                         zeroMethod = "none", normMethod = "none",
                         measure = "pearson",
                         sparsMethod = "threshold", thresh = 0.3,
                         dissFunc = "signed",
                         seed = 20190101)

dissnet2 <- netConstruct(amgut1.filt, group = groups_diss,
                         filtTax = "totalReads",
                         filtTaxPar = list(highestVar = 1000),
                         filtSamp = "highestFreq",
                         filtSampPar = list(totalReads = 50),
                         zeroMethod = "none", normMethod = "none",
                         measure = "bray",
                         sparsMethod = "threshold", thresh = 0.3,
                         dissFunc = "signed",
                         seed = 20190101)

assoprops2 <- netAnalyze(assonet2, clustMethod = "cluster_fast_greedy",
                            hubPar = "eigenvector")

dissprops2 <- netAnalyze(dissnet2, clustMethod = "cluster_fast_greedy",
                            hubPar = "eigenvector")


context("association network; without permutation test")
netcomp_asso <- netCompare(assoprops2, permTest = FALSE)

context("association network; with permutation test")
netcomp_asso <- netCompare(assoprops2, permTest = TRUE, nPerm = 100, cores = 1L)

context("dissimilarity network; without permutation test")
netcomp_diss <- netCompare(dissprops2, permTest = FALSE)

context("dissimilarity network; with permutation test")
netcomp_diss <- netCompare(dissprops2, permTest = TRUE, nPerm = 100, cores = 1L)

context("test summary method")

summary(netcomp_asso)
summary(netcomp_diss)

#-------------------------------------------------------------------------------
context("differential network")
set.seed(123456)

context("permutation test")
diff_perm <- diffnet(assonet2, diffMethod = "permute", nPerm = 20, cores = 1L,
                     adjust = "none")

context("Fisher test")
diff_fisher <- diffnet(assonet2, diffMethod = "fisherTest", adjust = "none")

context("Discordant method")
diff_discord <- diffnet(assonet2, diffMethod = "discordant", adjust = "none")

context("test plot.diffnet")
plot(diff_perm)
plot(diff_fisher)
plot(diff_discord)


assonet3 <- netConstruct(assonet2$assoEst1, 
                         data2 = assonet2$assoEst2,
                         dataType = "correlation",
                         filtTax = "highestVar",
                         filtTaxPar = list(highestVar = 50),
                         filtSamp = "totalReads",
                         filtSampPar = list(totalReads = 1000),
                         zeroMethod = "none", normMethod = "none",
                         measure = "pearson",
                         sparsMethod = "threshold", thresh = 0.3,
                         dissFunc = "signed",
                         seed = 20190101)

context("Fisher test")
diff_fisher <- diffnet(assonet3, diffMethod = "fisherTest", adjust = "none",
                       n1 = sum(groups_asso == 0), n2 = sum(groups_asso == 1))



#-------------------------------------------------------------------------------
context("Small sample size")

suppressWarnings(assonet2 <- netConstruct(amgut1.filt[c(1,2,3,6), ], 
                                          group = groups_asso[c(1,2,3,6)],
                                          filtTax = "highestVar",
                                          filtTaxPar = list(highestVar = 50),
                                          zeroMethod = "none", normMethod = "none",
                                          measure = "pearson",
                                          sparsMethod = "threshold", thresh = 0.3,
                                          dissFunc = "signed",
                                          seed = 20190101))

assoprops2 <- netAnalyze(assonet2, clustMethod = "cluster_fast_greedy",
                         hubPar = "eigenvector")

netcomp_asso <- netCompare(assoprops2, permTest = FALSE)

expect_error(netcomp_asso <- netCompare(assoprops2, 
                                        permTest = TRUE, 
                                        nPerm = 100, cores = 1L),
             "Possible number of permutations")

expect_error(diff_perm <- diffnet(assonet2, 
                                  diffMethod = "permute", 
                                  nPerm = 100, cores = 1L,
                                  adjust = "none"),
             "Possible number of permutations")