I am trying to generate SJ.out.tab files to quantify splice junctions with my RNA-seq data.
I used values from column 7/8 as read counts. I compared the counts pulled from SJ.out.tab with outputs from another tool, I found contradicting results from them. While SJ.out.tab indicates that there is no read (i.e., count = 0) in particular loci, the other tool provides reads. I also used IGV and saw splice reads from both .bam and .bed files, which seems to indicate that the problem lies with the counts I pulled from SJ.out.tab.
I guess I might do something wrong with my SJ.out.tab file. Do you have any suggestions?