Help with DEseq2 design matrix and DE expression analysis

Question

[Naresh D J]

Hi,

I am having 6 samples (A,B,C,D,E,F) each with two replicates. I have the HT-seq count table with raw read counts. I would like to do the DE analysis

• A vs B
• A vs C
• A vs D
• A vs E
• A vs F
• B vs C
• B vs D
• ...etc

basically all the combinations of 1 sample vs other (in total 15 combinations). How can I do the this in DEseq2?

In addition, I would like to do the PCA and clustering on all 6 samples and would like to check whether replicates are similar in all samples and how similar the samples to each other.

Thank you very much.

Best Regards,
Naresh D J

Answer 1

[karl.stamm]

There's a bioconductor package called rgsepd which has a function to do exactly what you are asking. You provide it a table of read counts, (column names are samples, row names are human RefSeq NM_###), and a second table of sample names to condition. Then it can do all pairs DESeq2 for you, filter the tables, generate PCAs, clustering and heatmaps.

See the manuals or ask a question, we can build a tutorial.

[Naresh D J]

Thanks you.