[R] WGCNA: Combining block-wise dendrograms and modules into a single figure

[Peter Langfelder]

Hi Derek,

the simple answer is that the block-specific dendrograms cannot be
meaningfully combined into a single dendrogram. You have to plot them
separately. You can create a multi-panel figure that shows all block
dendrograms in one big figure, although with 10 blocks I would not
necessarily attempt it.

One comment that does not quite relate to your question but may still
be relevant is that you seem to have too many blocks. Unless you have
a really old computer with 1GB RAM or less, the block size you used is
likely too small. (The tutorial uses a small block size to illustrate
the approach.)  If you have 4GB RAM, try maxBlockSize = 8000. If you
have 8GB, you can try maxBlockSize about 12000.  You will get fewer
blocks and most likely a better clustering/modules.

Best,

Peter

On Thu, Oct 18, 2012 at 7:33 AM, Derek55 <derekgperez at utexas.edu> wrote:
> Hello fellow R users,
>
> I am currently learning to use R, so please forgive me if there is an
> obvious explanation for the following problem. My goal is to perform WGCNA
> on a dataset of 19776 genes, so I opted to follow the block-wise network
> construction (Section 2c) in the WGCNA R Tutorial by Peter Langfelder and
> Steve Horvath.
>
> I ended up with 10 blocks and am able to plot the dendrograms and module
> colors for each block, but I am having trouble grouping all of the blocks
> together for a single dendrogram/module color figure. On page 4 of section
> 2c in the WGCNA tutorial, the single-block network construction workspace is
> loaded and it contains a variable called 'geneTree'. geneTree references
> block 1 of the single-block network construction, which encompasses the
> entire dendrogram.
>
> geneTree = net$dendrograms[[1]]
>
> The problem here is that I cannot perform single-block construction because
> I do not have enough RAM, so I do not know how to reproduce the complete
> dendrogram using the blocks that I am currently working with. If I had
> simply reused the same definition for geneTree using the block-wise network,
> it would simply print block 1 out of 10 blocks (I have already tried this).
>
> Session Information:
>
> R version 2.15.1 (2012-06-22)
> Platform: i386-pc-mingw32/i386 (32-bit)
>
> locale:
> [1] LC_COLLATE=English_United States.1252
> [2] LC_CTYPE=English_United States.1252
> [3] LC_MONETARY=English_United States.1252
> [4] LC_NUMERIC=C
> [5] LC_TIME=English_United States.1252
>
> attached base packages:
> [1] splines   stats     graphics  grDevices utils     datasets  methods
> [8] base
>
> other attached packages:
>  [1] WGCNA_1.23-1        MASS_7.3-18         reshape_0.8.4
>  [4] plyr_1.7.1          cluster_1.14.2      Hmisc_3.9-3
>  [7] survival_2.36-14    flashClust_1.01-2   dynamicTreeCut_1.21
> [10] impute_1.32.0
>
> loaded via a namespace (and not attached):
> [1] grid_2.15.1    lattice_0.20-6
>
>
>
> I have tried several methods for compiling the separate dendrograms together
> for the complete figure:
>
> Input:
>
> plotDendroAndColors(bwnet$dendrograms[[1:10]],
> bwModuleColors[bwnet$blockGenes[[1:10]]],
>                   "Module colors",
>                   dendroLabels = FALSE, hang = 0.03,
>                   addGuide = TRUE, guideHang = 0.05)
>
> # Above, 'bwnew$dendrograms[[x]]' is equivalent to the parameter specified
> by 'geneTree', but must be changed each time to plot blocks 1-10
> # In addition, bwModuleColors[bwnet$blockGenes[[x]]] should be changed with
> the parameters for the geneTree equivalent
>
>  Output:
>
> "Error in bwnet$blockGenes[[1:10]] : recursive indexing failed at level 2"
>
> Input:
>
> Same as above, changing bwnet$dendrograms[[1:10]] to bwnet$dendrograms[1:10]
> and bwModuleColors[bwnet$blockGenes[[1:10]]] to
> bwModuleColors[bwnet$blockGenes[1:10]]
>
> Output:
>
> "Error in bwModuleColors[bwnet$blockGenes[1:10]] :
>     invalid subscript type 'list' "
>
> Input:
>
> Same as above, but changed bwModuleColors[bwnet$blockGenes[1:10]] back to
> bwModuleColors[bwnet$blockGenes[[1:10]]]
>
> Output:
>
> "Error in bwnet$blockGenes[[1:10]] : recursive indexing failed at level 2"
>
> Any of these examples could be replicated straight from the tutorial using
> the mouse data and trying to combine block 1 and block 2 without loading and
> referencing the auto-network construction workspace.
>
> Thank you very much for your time and input,
>
> ~Derek
>
>
>
> --
> View this message in context: http://r.789695.n4.nabble.com/WGCNA-Combining-block-wise-dendrograms-and-modules-into-a-single-figure-tp4646627.html
> Sent from the R help mailing list archive at Nabble.com.
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.