Hi, I'm a computer engineer doing a thesis on bioinformatics (gene expression and regulation). I'm trying to infer regulation in a time-course analysis.

I have a set of 6 time-course samples (0h, 6h, 18h, 24h, 48h, and 84h), and put them through BasePairTech. From what I gathered, it runs a gene count with STAR and it feeds the results to DESeq2 for the following experiments:

• 6h vs 0h
• 18h vs 6h
• 24h vs 18h
• 48h vs 24h
• 84h vs 48h
• 1 vs. the rest for all samples.

I understand what the p-value represents, but I'm not sure it serves a purpose in my investigation. Because of this, I have the following questions:

• Is the p-value relevant for gene expression analysis?
• Will it help me identify relevant (not necessarily significant) genes?
• What's the purpose of the p-value in an experiment about regulation?
• Is the p-value specifically relevant in the DE experiments for two samples?

Also (very important): Which values from the gene count does DESeq2 uses from the gene count (Gene count, FPKM, TPM both by transcript and by gene)?

I'd really appreciate any leads on this.

Please take a look at the publication [1] and the workflow [2]. The publication discusses how the p-values are generated, and the workflow has an example of how we recommend to approach time course data.

[1] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4302049/

[2] https://bioconductor.org/packages/release/workflows/vignettes/rnaseqGene/inst/doc/rnaseqGene.html