TopHat clarification for pair-end sequences

[gwilymh]

I am running paired-end RNASeq data through a Tuxedo pipeline. In the TopHat manual, it states that for the <reads1_1[,...,readsN_1]> input "this should be the *_1 ("left") set of files"; and for the <[reads1_2,...readsN_2]> input should refer to "the *_2 ("right") set of files".

Do these input files literally have to end in _1 and _2 suffixes? Or is this assumed from the file order?

[dpryan]

They don't literally have to end in _1 and _2, in fact you'll often see _R1 and _R2 or something similar. As you suspected, the order determines which file(s) are assumed to be the first or second read of a pair.

[gwilymh]

Thanks for clarifying dpryan, much appreciated.