rRNA detection (for contamination) in RNA-seq

Question

[Ron]

Hi all,

I am working with human RNA-seq data.I want to know whats the best way to get information on rRNA reads in RNA-seq? Does rRNA gets passed on in the bam files after alignment(using human genome) or we have to check the raw fastq reads only for this?

Also I want to look for this irrespective of the library they came from (polyA or rRNA removal kits),as I just want to have an estimate of this rRNA reads as a QC measure.

let me know your thoughts.

Thanks,

Ron

Answer 1

[Ron]

I used RseQC for rRNA detection. They have a bed file too.

http://rseqc.sourceforge.net/

Answer 2

[michael.ante]

The rRNA-cluster are distributed in repeats over the genome. Therefore, reads origin from rRNA-molecules map too many times and get discarded by most RNA-SEQ aligners.

I'd download the rRNA fasta files from NCBI and map all reads against them (e.g. BBmap, bowtie2); the mapping rate is equal to your rRNA contamination. Additionally, you can use this as a first step to sort these reads before continuing your pipeline.

Answer 3

[igor]

I asked a related question recently that might be useful: RNA-seq rRNA contamination

Answer 4

[nick.a.rouse]

A very simple approach would be to download rRNA bed coordinates (from Ensembl) and then count the total number of reads that fall into this ROI (using bedtools suite) and divide this by total number of reads in the bam (samtools idxstats). This would give you a rough estimate of your background rRNA levels.

Answer 5

[Charles Plessy]

You can count or filter out rRNA reads from FASTQ files (single- or paired-end) using TagDust (and do many other things).