Hello everyone. I have RNAseq data of bulk transcriptome corresponding to different tissues in different infection conditions.
I have: a) Infected spleen (3 replicates) b) Non-infected spleen (3 replicates) c) Infected muscle (3 replicates) d) Non-infected muscle (3 replicates)
I have a table with the individual number of reads for each gene for each replicate of each condition (a-d). I also have two more tables (one per tissue) with the fold change of each gene expression comparing the infected vs non-infected conditions. Those 2 tables have also the p-values for every comparison.
I would like to write a script to do a Fold change vs fold change plot like this one:
The idea would be to compare differentially expressed genes between tissues to see which ones behave similar upon infection irrespective of the tissue studied.
Can anyone tell me which package would be useful to do this kind of plots with the data tables that I currently have?
I have found the package "Vidger" wich allows to do a "four way plot", that looks quite simmilar to the one I want to create. However, Vidger compares 2 conditions with a control one, and that is not the case for my experimental design.
Do you know any R? It would be fairly easy to make a plot like this in R. However, unlike the plot above, the fold-changes should be in log2 scale, so the distance on the plot is the same whether something is enriched in muscle or spleen (i.e. in the numerator or denominator of any given fraction). If you need to do the statistics you could use the edgeR package, which has a fairly easy way to compare genes as follows: (infected spleen - spleen) - (infected muscle - muscle). Which would be the equivalent of (infected spleen/spleen)/(infected muscle/muscle), in other words it would show you genes responding differently to infection in spleen versus muscle. The edgeR userguide is a great resource, with examples. You could also examine the changes with a traditional MA plot. Or, similar to the xy plot above just plot logFC vs logFC.
Thank you very much for the information! I will give it a try with edgeR package.