Metadata.txt configuration

[Johannes]

I am trying to run a metatranscriptome analysis. I have five different samples of the same habitat,

sample1: t0
sample2: after 1 hour
sample3: after 2 hours
sample4: after 3 hours
sample5: after 4 hours

Can someone explain me how I have to set the last two columns in the Metadata.txt configuration files (F1-Paired and F2-Diet)?

Thank you for our help.

Kind regards

Johannes

[Johannes]

I tried to simplify my config file by removing the last column (F2-Diet) but I still have the same problem.

Short clarification: This is a test experiment for testing the experimental setup. Therefore, there are no replicates for each sample.

Can I run the analysis anyway? And if so, how?

Thank you.

[metatr...@gmail.com]

Hi Johannes,

In the doc you will find a description of how to fill in the metadata table: http://metatrans.org/#metadata

Basically, "F1" to "Fn" are the factors of interest of your samples, i.e. the columns that can be used to
test for the differential expression analysis. These factors if they are not "F0" are included in the GLM model
of the DESeq2 package to test for DE. This is a brief description of the metadata columns:

#SampleName: name of your sample
SampleFiles: name of the files of your samples, 2 files for each of the pair-ends
Shortname: brief name of the sample used for the analysis
Order: order in which samples will be processed (0 to avoid processing for that sample)
Type: description of the type of molecule you are analyzing (mRNa, totalRNA,...)
Description: alternative names of your sample or other info
F1-Fn: The factor of interest must be the last and only can have a contrast of two levels or conditions, i.e.
           if you have column names:  F0-XX,F2-XX,F0-XX,F0-XX,F5-XX, then F2-XX and F5-XX will be included
           in the GLM model, but only the conditions of F5-XX are used for DE.

If you would add replicates you would include a column like: F4-Replicates, identifying with the same label the samples
of the same replicate.

Cheers