Entering edit mode
Hello everyone,
I would like to use only "p- value" instead of "p-adj value" in DESeq2. but I could not find the proper command for doing so. While I used the below command but I think it is not correct and gives me error. is there any one to assist me in this case? Thank you
table(res$pvalue)
summary(results(dds, pvalue= 0.0037, lfcThreshold = 0.5))
Error: unexpected '=' in "summary(results(dds, res$pvalue="
# include your problematic code here with any corresponding output
# please also include the results of running the following in an R session
sessionInfo( )
Thank you for your response, sorry but I not got exactly your point, I am looking for command to have just raw p-value, because this command is rejected.
You can use standard R code to filter or work with a results table. See our workflow for some example code before going further.
resis a type of data frame (it's a Bioconductor DataFrame), so you can use whatever steps you like to work with it. You can even use "tidyverse" commands if you prefer, by converting it to a regular data.frame first:Then you could do e.g.:
Or in base R:
many thanks for your reply, I tried this command but again it gave me the counts based on padj-value not raw p-value, what is the solution? is it there some commands to do so?
We recommend to use adjusted p-value and not p-value. See the section of the workflow that describes why:
https://www.bioconductor.org/packages/devel/workflows/vignettes/rnaseqGene/inst/doc/rnaseqGene.html
I agree with you, but when I am using adjusted p-value, and trying different adjusted p-values, but it gives me nearly zero up and down regulation counts, In this case what is your suggestion to me?
That means you cannot reliably call DE genes in this dataset.
Thanks again for your reply. I followed different tutorial of DESeq2, there is not any problem, just the final results is wrong. could you please advice me what to do in such situation?
Final result is not wrong, it’s just saying something different than what you were expecting.
many thanks, so what is your suggestion to me in this case?
Could you update the link?
just look for the package here