Not covered in the splicekit manual: Questions about _altsplice.tab and _difffeature.tab result interpretation

[santataRU]

Hi Gregor,

I want to analyze the differential usage of junctions between two groups (treatment vs. control). I tried to read the manual for interpreting the results but could not find explanations for this part. In the results/edgeR/junctions folder, there are two files with suffixes _altsplice.tab and _difffeature.tab. I have the following questions:

1. There is a discrepancy in row counts: _difffeature.tab has 131,630 rows, while _altsplice.tab has only 130,432. Why is _difffeature.tab larger?

2. I see a column called exon.F in the _altsplice.tab but not in the _difffeature.tab. What does exon.F mean, and why does the _difffeature.tab lack this column?

3. I see a column called logCPM in the _difffeature.tab but not in the _altsplice.tab. Why does the _altsplice.tab lack this column?

4. For junctions with the same feature_id in both tables, their logFC values differ. Is this normal?

5. What do the pi_values mean in both tables? I couldn’t find an explanation in the manual.

6. According to the manual, the results should include sum_feature_test and sum_feature_control columns, but I don’t see them in either table. Where can I find the sum and average of counts for each junction across the treatment or control groups (or do I need to calculate these myself)?

I really appreciate your effort in developing this tool.

Best Regards,
Xiao

Table links shared: https://drive.google.com/drive/folders/1AUIsx52rEWnk4yDB8pMHhTAitnPVCk4W?usp=sharing