associate_phenotype error when phenotype matrix only has one column

[Kumquatum]

Minimal example :

eigengene_mat <- data.frame(mod1 = rnorm(20, 0.1, 0.2),
                            mod2 = rnorm(20, 0.2, 0.2))
phenotype_mat <- data.frame(phenA = sample(c("X", "Y", "Z"), 20,
                                           replace = TRUE))
association <- associate_phenotype(eigengene_mat, phenotype_mat)

>> Error in cor(x, y, use = use, ...) : 'y' has a zero dimension.

Supposed origin to check : the data.frame with one column is somewhere transformed into vector