DESeq2 workflow for RNAseq

Question

[AP]

Hi everyone,

I just made some graphics Pheatmap and MA plot using DESeq2 for my RNAseq differential expression analysis. However now I am really confused on what to do next. Can I make some table and calculate number of expressed genes across samples. If I could compare the data for my replications and treatments through barplots and venndiagrams .I tried to look on so many stuffs which is just making me more confused.

Looking forward for the suggestions and guidelines.

Thank you,

Ambika

Answer 1

[Devon Ryan]

You can plot the (normalized) counts across your samples for particular genes if you'd like. However, looking for "expressed" genes that change between samples is usually not useful (there's no really useful definition of "expressed"). Normally one takes the statistical results from DESeq2 and then goes from there, with things like GO/pathway enrichment or simply looking at what the various DE genes are.

[AP]

Devon Ryan thank you for the clarification. I think I will just be looking at what DE genes are. So in order to plot the counts across my samples how can I write a table for transformed values?

Thank you, Ambika

[Devon Ryan]

Just use the plotCounts() function.

[bioinfo456]

C: Using DESeq2 results for building a classifier

What are your thoughts? You reckon I should use an independent set of samples to check for accuracy?

[Devon Ryan]

That would be fine and it would be better to post questions like this as standalone posts.