Hi Everyone,
To preface this, I have little programming experience and have mainly been using web tools.
Undergrad student here; I'm working in a cancer bio lab over the Summer. The project concerns comparing LNCaP (a prostate cancer cell line) to LNCaP BoM (a bone-metastatic prostate cancer cell line). Essentially, we want to find out key drivers of bone metastasis. Previously, the lab has RNA-seq completed for one sample of LNCaP and one sample of LNCap BoM.
One of my jobs is to find significant genes that can be targeted by effective drugs. One lab member recommended trying OCTAD and uploading my RNA-seq data. However, the software requires p-values (meaning multiple sample RNA-seq, which we don't have). I'm stuck at this point and am not sure how to proceed. I've seen software such as LINCS, ssGSEA, etc, but I'm not sure how to use any of them.
I would greatly appreciate it if someone would be able to help me!
Attached is a screen shot of the RNA-seq:
Thank you!
It's probably as you eluded to, having only a single sample from each condition and trying to find differential expressed genes is not gonna bring up anything useful. You would need multiple samples to be able to estimate the gene's variances etc to be able know whether it is statistically significant or e.g part of the genes normal variation. If you are willing to do some work in
R, you could try and possibly use edgeR to see if you would be able to something there (I am unsure if it works for just 1 sample in each condition). The user guide is really helpful when going through the various steps, and giving explanations; it also cautions about what to do if there are no replicates. There is the F1000Research article from edgeR as well which contains some additional information (here) which could maybe prove helpful (iirc the answer to reviewer report 1 mentioned that 2 replicates v 1 replicates /could/ still work, don't know about 1v1).