All the values of IJC_SAMPLE_1 and SJC_SAMPLE_1 are zero

[yutaixian]

hi，I run rmats with python rmats.py --b1 ./d_1D.txt --tmp ./output/d_11/tmp --statoff --gtf 1D_43_stringtie.gtf -t single --readLength 75 --nthread 4 --od test_output/d_11 but all the values of IJC_SAMPLE_1 and SJC_SAMPLE_1 are zero. Why? I have checked my BAM files and there are no problems.

[EricKutschera]

Since you ran with only --b1 the statistical model is not run and there is no filtering so events will be included in the output files even if there are no supporting reads: #96 (comment)

If there are no supporting reads for any event then the read outcome section should show why reads were not USED: #328 (comment)

[yutaixian]

Since you ran with only --b1 the statistical model is not run and there is no filtering so events will be included in the output files even if there are no supporting reads: #96 (comment)

If there are no supporting reads for any event then the read outcome section should show why reads were not USED: #328 (comment)

This is my nohup output file. I'd like to ask you if this looks normal. Are there any data flaws? I noticed that the value of EXON_NOT_MATCHED_TO_ANNOTATION is 110,612,760, which accounts for nearly 50%. Any suggestion？

[EricKutschera]

The large number of reads filtered for EXON_NOT_MATCHED_TO_ANNOTATION looks like the main issue. Those reads were aligned without a splice junction and the region they aligned to is not annotated as an exon in your --gtf. One possible reason for EXON_NOT_MATCHED_TO_ANNOTATION is the bam files being aligned with a different gtf or version of the reference genome. If the aligner used different reference information then it makes sense for the alignments not to match the --gtf

Your previous post mentioned --gtf 1D_43_stringtie.gtf. It could be that the stringtie gtf just doesn't match up well with your reads