when to use paired stats model

[Llywelyn Griffith]

Hi,

Should I use the paired stats model if I have paired-end RNA-seq data? Or does this refer to pairs of replicates between sample groups? I'm comparing three replicates of condition A vs three replicates of condition B (the replicates within each condition are arbitrarily labelled and are not linked to a corresponding replicate in the other condition), my data is paired-end, so I am unsure as to whether the paired stats model is appropriate for my analysis.

Many thanks for any advice!

Kind regards,

Llywelyn

[kutsc...@gmail.com]

`--paired-stats` is for when each replicate in the first condition has a corresponding replicate in the other condition. The paired stats model is not related to paired-end RNA-seq data. The rMATS argument for paired-end data is `-t paired`. The default stats model makes sense for the situation you described

Eric

[Llywelyn Griffith]

Hi Eric, 

Great, thank you for your reply.

Llywelyn