Hi,
I have some data in which the library prep isn't known. We're doing QC on it, and trying to determine how contamination filtering should be done. I'm wondering if there is any way to infer if rRNA depletion or mRNA enrichment was done from artifacts within the sample data.
Thank you, Synanth
•
written
by
Synanth •
1 answer
It is basically impossible to truly know rRNA depletion vs mRNA enrichment computationally.
However, looking at exon-to-intron ratios may be a helpful indicator.
For example, see our dataset here comparing PolyA-enriched and Ribo-depleted libraries from the same cell line.
The ribo depletion leads to more intronic reads, while polyA-enriched regions should be almost all exon. Although this may vary by cell type and efficiency of the library prep.
rRNA depletion can be done shoddily with either library prep type though...
X-axis: log2 exon RPKMs
Y-axis: log2 intron RPKMs
Log in to answer this question.
-
FastQC duplicates questions
written by Beth •
-
Does sequencing depth influencing ATAC peak calling?
written by Char •
-
TMM on a targeted RNA-Seq dataset
written by oakhamwolf •
-
What sequence do I get when using paired-end stranded protocol
written by Assa Yeroslaviz
-
Bioanalyzer: fragment length and standard deviation for kallisto
written by piero •
-
trimming RNA-seq reads with adapter sequence?index? using trimmomatic
written by woongjaej •
-
Subtracting contamination from RNASeq data
written by Gareth Palidwor
-
proximal duplicate parameters - HiSeq X
written by Richard
-
Finding UTR regions in the RNA sequencing data from a non-model organism
written by seta
-
Rna Seq Mrna|Gene Count Data Filteration
written by Sudeep
If you have rRNA sequences of the organism you are working with then you could use
bbduk.shin filter mode (guide https://jgi.doe.gov/data-and-tools/software-tools/bbtools/bb-tools-user-guide/bbduk-guide/ ) to look for reads that match the sequences.You could also try to predict the sequences (if you are working with a unique organism) by using one of the programs here: Tools for rRNA gene annotations
It's human data, and I've been using fastq_screen to determine where the reads are from. Fastq_screen shows where reads potentially multimap against more than one genome, but the overlap amount seems to vary highly between samples. I want to try and infer the library prep to see if rRNA depletion was done shoddily on some samples, so that I know how to proceed with the filtering criteria.
Contamination with rRNA as opposed to some other organism are two different things. You can find the sequence of the complete human rDNA repeat in a prior thread (how can i download human ribosomal reference ? ). That should be good enough for the screen with
bbduk.sh. If the reads do not turn out to be rRNA then you will have to move on to the other possibilities.