Hi All,
I'm interested in analysing multiple RNA-seq studies (more than 10) for alternative splicing, each study has two conditions (diseased, control) and I want to assess alternative splicing patterns across these studies.
I've completed 1st pass mapping with STAR (v2.7.9a) with mostly default settings, and am currently doing 2nd pass. However, I'm confused as to the proper method to use. From previous posts, I am aware that SJ.out.tab files from 1st pass can be inputted directly into STAR without the need to concatante and filter them (Multi-sample 2-pass mapping-section 8.1, STAR manual). Older posts before 2020, refer to filtering SJ.out.tab files and combining them for index re-generation. I'm inclined to go for the filtering option using the parameters specifiied in this post, to minimise the chance of including false positive junctions but am unsure on how to assess the sutiability of filters for this purpose.
If anyone has any thoughts on this, that would be great, thanks in advance.
