Hi! When I try to calculate module preservation statistics from adjacency data, modulePreservation function in WGCNA 1.67 package throws an error as follows:

> mp = modulePreservation(multiExpr,
+                           multiColor,
+                           dataIsExpr = F,
+                          nPermutations = 2)
  ..checking adjacencies for excessive amounts of missing data
  ..calculating observed preservation values
Error in make.unique(rownames(datRef)) : 
  'names' must be a character vector

Here is a reproducible example with a fake dataset:

library(WGCNA)
setLabels = c("one", "two");
multiExpr = list(one = list(data = c()),
                 two = list(data = c()))
n <- 1000 # dimensionality
genes <- paste0("Gene", 1:n)

# Fill both datasets with fake adjacencies
for (i in 1:2) {
  x <- matrix(runif(n*n, min = 0, max = 1), n)
  diag(x) <- 1
  x[lower.tri(x)] = t(x)[lower.tri(x)] # make matrix symmetric
  row.names(x) <- genes
  colnames(x) <- genes
  print(x[1:3,1:3])
  multiExpr[[i]]$data <- x
}

multiColor = list(one = sample(standardColors(50), n, replace = T))

mp = modulePreservation(multiExpr,
                          multiColor,
                          dataIsExpr = F,
                         nPermutations = 2)

I see this error only in WGCNA 1.67 when dataIsExpr = F. The same code works in both R 3.5.2/3.4.2 1) when dataIsExpr = T or 2) on WGCNA 1.64. I calculated preservation statistics directly from expression already, but I'm still curious whether this is just a WGCNA bug or some problem with my data which WGCNA 1.67 is unable to tolerate.