The expression difference test

[primary-11]

The expression difference test between the IncLevel1 column and the IncLevel2 column should use what statistical test method?

[EricKutschera]

http://dx.doi.org/10.1073/pnas.1419161111 describes the statistical test used in rMATS-turbo and the supporting information PDF from that article provides more details

[primary-11]

http://dx.doi.org/10.1073/pnas.1419161111描述了 rMATS-turbo 中使用的统计测试，该文章的支持信息 PDF 提供了更多详细信息

I may have expressed myself incorrectly.IJC_SAMPLE_1 and IJC_SAMPLE_2 are count numbers, and I would like to perform differential expression analysis on these two columns. Can I use DEXseq for differential analysis, or do I need to convert the counts to FPKM for this analysis? If conversion is required, how should I perform this conversion?

[EricKutschera]

I'm not familiar with DEXseq but it seems like it needs counts for a specific exon and also total counts for the gene. What reads contribute to IJC counts are shown in the diagram from the README: https://github.com/Xinglab/rmats-turbo/tree/v4.3.0?tab=readme-ov-file#output

For SE events it's counting the junctions around the exon and the JCEC file also counts the exon reads without a splice junction. Maybe you could use those counts with DEXseq, but I'm not sure of the details. Getting the gene totals from the rMATS output could be an issue since not all exons in a gene will be detected as being skipped and also some exons may be reported as multiple SE events if there are different possible upstream or downstream exons for the skipped exon