Hi, 

I am testing the JunctionSeq and QoRTs package via the example Walkthrough, but with my data. If I understood the protocol correctly, it supposes to be possible to analyse multiple conditions with one command. 

I my case I have four different conditions called ES, EB1-EB3.

when running the runJunctionSeqAnalysis I can give a condition column with my data, but if i want to analyse multiple conditions, i need to make sure, that the first level of this factor will be my try reference samples. 

This is not the case with my samples, in my vector it will be something like (EB1, EB2, EB3, ES), while ES is my reference. In the JunctionSeq vignette I have found a different approach for multiple conditions using the advanced analysis pipeline with the command readJunctionSeqCounts. 

For that I have two questions:

1. is it possible to use the relevel command to account for that problem like that:

design <- data.frame(condition = factor(decoder.bySample$group.ID))
design$condition <- relevel(x =  design$condition, ref = "ES")

Does this means that can understand that my reference is the ES samples?

2. What is happening in this kind of analysis? Are all the samples compared separately to the reference? Is it exactly like running the pair-wise comparisons of EB1 vs. ES, EB2 vs. Es and EB3 vs ES?

Do I get here an adjusted p-value for each of the comparisons?

thanks

Assa