MD plot for every sample (?!)

Question

[Barista]

Dear all,

I am conducting a QuantSeq analysis and in the manuals I find online one of the steps include making an MD-plot for every sample. This is considered data exploration as well as a quality control check, this of course sounds very informative, but as I have 600 samples, making a plot for every sample will be a lot of work.

Is this a necessary step and if so, are there any other methods that I could try?

Kind regards,

Barista

[colindaven]

You can use Rscript to automate analysis like this, and save each to a file.

This might help get you started:

https://stackoverflow.com/questions/3301694/running-r-scripts-with-plots

Else, if you don't like that, then you can try a for loop in R saving each to a PDF.

Answer 1

[Gordon Smyth]

No, it is not a necessary step. I am the author of the plotMD function (and of the plotMA function before that) but I definitely wouldn't use it as a data exploration tool with 600 samples. The edgeR workflows and edgeR User's Guide illustrate more streamlined data exploration tools.

[Barista]

Dear Gordon,

Thanks for you reply!

I have done a MDS plots and BCV plot already (according to the manual).

Will this suffice?