RNA-SeQC result interpretation

Question

[gskbioinfo143]

This is regarding RNA-SeQC result clarification.

When gencode.v7.annotation.gtf was used with command

java \
  -jar RNA-SeQC_v1.1.8.jar \
  -n 84408 \
  -s "DGL1094|DGL-finalsort-104.bam|DGL1094" \
  -t gencode.v7.annotation.gtf -r hg19_.fasta \
  -o ./DGLfinalsort/ \
  -strat gc \
  -gc gencode.v7.gc.txt \
  -BWArRNA human_all_rRNA.fasta \
  -noDoC \
  -singleEnd \
  -transcriptDetails \
  -rRNAdSampleTarget 1 million

there was no rRNA count in my bam file but when used command

java \
  -jar RNA-SeQC_v1.1.8.jar \
  -n 84408 \
  -s "DGL1094|DGL-finalsort-104.bam|DGL1094" \
  -t gencode.v7.annotation.gtf \
  -r hg19_.fasta \
  -o ./DGLfinalsort/ \
  -strat gc \
  -gc gencode.v7.gc.txt \
  -noDoC \
  -singleEnd \
  -transcriptDetails \
  -rRNAdSampleTarget 1 million

rRNA count was 39462. I want to clarify is the problem with human_all_rRNA.fasta file(downloaded from webpage).

If the gencode.v19.annotation.gtf of hg19 version was used the rRNA count was different 4506. I just want to clarify what went wrong? Which gtf file to consider for calculation?

Hope to hear from you

Thank you

Answer 1

[GouthamAtla]

According to the manual using -BWArRNA is more robust method.

If you would like to know about rRNA reads, you could:

1. Pull out the rRNA coordinates from the GTF file in a bed format and see how many reads mapped to those regions? See their mapping quality stats etc.
2. Align the reads to human_all_rRNA.fasta and see how many reads mapped there ? This should not take long time.

This is what the tool is doing but if you do independently it will give more insights to the problem.