QC with RNA-SeQC output weird

Question

[koaeb]

Hi, I have a set of RNA-seq data, I used STAR for mapping then used RNA-SeQC to check the quality. After STAR mapping I checked the Log.final.out file, there are 81% reads uniquely mapped. I then used RNA-SeQC to produce quality metrics. I used collapsed gtf as input. However in the metrics.tsv file only 18% reads uniquely mapped, and the gene_tpm.gct.gz file and other files contain nan or 0 reads for each gene value, like below: ENSG00000227232.5 WASH7P 0 or ENSG00000223972.5 DDX11L1 -nan. It seems something wrong with RNA-SeQC step. I downloaded rnaseqc.v2.3.6.linux, and used following command: rnaseqc ${genes_gtf} ${bam_file} . -s ${sample_id} --stranded rf -vv. No error message showed up, so I'm not sure where went wrong. Any thoughts? Thank you.

[raplayer]

Hi, did you ever find a solution for this?

Similar trouble here via the GTEx pipeline (v10, so using RNA-SeQC v2).

From metrics file:

• Mapping Rate 0.98066
• Unique Rate of Mapped 0.598804

But still all zeros in the count data...

Answer 1

[Jeremy Leipzig]

You probably have some name mismatch in your reference genome vs gtf. If you like you can sign up for a free trial at Truwl and run GTEx v10 there: https://truwl.com/workflows/library/GTEx/v10