How to remove this module selection error?

Question

[gupta00aritra]

norm_count[]<- sapply(norm_count, as.numeric)
soft_power <- 3
temp_cor <- cor
cor <- WGCNA::cor

#Memory estimate with  respect to blockwise
bwnet<- blockwiseModules(norm_count, maxBlockSize = 18000,
              TOMType = "signed",
                 power = soft_power,
                 mergeCutHeight = 0.25,
                 numericLabels = F,
                 randomSeed = 04321,
              minModuleSize = 30,
                 verbose=3)Warning message:
In blockwiseModules(norm_count, maxBlockSize = 18000, TOMType = "signed",  :
  blockwiseModules: mergeCloseModules failed with the following error message:
     Error in mergeCloseModules(datExpr, colors[gsg$goodGenes], cutHeight = mergeCutHeight,  : 
  Error in moduleEigengenes(expr = exprData[[set]]$data, colors = setColors,  : 
  Color levels are empty. Possible reason: the only color is grey and grey module is excluded from the calculation.

--> returning unmerged colors

[andres.firrincieli]

Soft-power=3?

How did you come up with this number?

[gupta00aritra]

by plotting power vs signed R^2

[gupta00aritra]

# choose a set of soft thresholding powers
 power<- c(c(1:10), seq(from = 12, to = 50 , by = 2))

 # call the network topology analysis function
 sft<- pickSoftThreshold(norm_count,powerVector = power, networkType = 'signed', verbose = 5)
 sft.data<- sft$fitIndices

# visualisation to pick power
 a1<-ggplot(sft.data, aes(Power, SFT.R.sq, label = Power))+
   geom_point()+
   geom_text(nudge_y = 0.2)+
   geom_hline(yintercept = 0.1, color = 'Red')+
   labs(x = 'Power', y = 'Scale free topology model fit, signed R^2')+
   theme_classic()
a2<- ggplot(sft.data, aes(Power, mean.k., label = Power))+
  geom_point()+
  geom_text(nudge_y = 0.1)+
  labs(x = 'Power', y = 'Mean connctivity')+
  theme_classic()

[andres.firrincieli]

1. check that in norm_count columns correspond to genes and rows to samples.
2. how many samples do you have?
3. show me the the two plots

Thank you!

ps. soft power of 3 is too low. For a signed network you should have soft-power threshold > 10