Question

Problem with WGCNA - Module-trait relationship

0

Entering edit mode

amoltej ▴ 10

@amoltej-7192

Last seen 7.2 years ago

Australia

Hello everyone,

I am trying to run the WGCNA on my data set of 87 samples.

First 2 steps executed successfully. In the third step when running following commands I am getting all the "NA"s in the output.

moduleTraitCor = cor(MEs, datTraits, use = "p")
moduleTraitPvalue = corPvalueStudent(moduleTraitCor, nSamples)

please find input files for these commands below

MEs0 - https://drive.google.com/open?id=0B42oFSm-XOAUbm9fRkx2Qk8wNDQ

MEs - https://drive.google.com/open?id=0B42oFSm-XOAULTEtcHdSWUMtQ0k

Can someone please help to understand whats happening.

thanks in advance

wgcna • 2.1k views

ADD COMMENT • link updated 7.2 years ago by Lluís Revilla Sancho ▴ 760 • written 7.2 years ago by amoltej ▴ 10

score 0 · Answer 1 · 2017-09-04

0

Entering edit mode

Lluís Revilla Sancho ▴ 760

@lluis-revilla-sancho

Last seen 15 days ago

European Union

The first line calculates the correlation of each trait with the egienvalue of each module, the second line gives you the p-value of said correlations for a given number of p-values. I advise you to use a function named somehow like corAndPvalue which gives both correlations and p-values takeing into account the number of samples with information for each trait

ADD COMMENT • link 7.2 years ago Lluís Revilla Sancho ▴ 760

0

Entering edit mode

Hi Lluis,

Thanks for the reply but the result is still same. I suspect for some reason correlation is not calculating or calculating it to 0 for all the values

ADD REPLY • link 7.2 years ago amoltej ▴ 10

0

Entering edit mode

What code do you use for "the third step"? Are there any NA in your input data? Did you use the "complete" observations option to calculate the correlations?

ADD REPLY • link 7.2 years ago Lluís Revilla Sancho ▴ 760

0

Entering edit mode

I am using this code for third step - https://labs.genetics.ucla.edu/horvath/CoexpressionNetwork/Rpackages/WGCNA/Tutorials/FemaleLiver-03-relateModsToExt.pdf

Yes, I am using complete observations to calculate correlations. There are no NA's in the input dataset for these commands. you can check input files from my question.

ADD REPLY • link 7.2 years ago amoltej ▴ 10