Hi everyone, I see many questions about strandedness in the rnaseq pipeline, but not here. I am analyzing an RNA-seq dataset from GEO which has no direct mention of the strandedness of the reads. Since they used the Illumina TruSeq Stranded mRNA kit, I inferred that the _R1 fastqs have to have reverse strandedness and _R2 fastqs forward strandedness. Yet, on the Usage/parameters page of the rnasplice, I see that the sample sheet only takes "forward", "reverse", or "unstranded" as strandedness indicators. How would I have to correctly indicate the strandedness in the sample sheet this case? Thank you for your help!

The

strandedness

specification in the sample sheet goes by where the R1 is located, so

reverse

in your case (if the kit works that way; if you are in doubt, you can grab e.g. 100,000 reads and map them, then look at the BAM with IGV (inside exons) and see how the direction of the R1 compares to the direction of the gene/trnscript; you could even run a subsample through the #CE8SSJV3N pipeline with

auto

and minimal parameters (skip everything unnecessary). I hope the

auto

option will finds its way into this pipeline as well.

Thank you very much for your elaborate answer, Thomas! I really appreciate the different solution ideas you provided me with 🙂

I have the same question as well!. Before, when running

rnaseq

inside

nf-core

, or

nf-core/rnaseq

, for testing purposes, they look like this format/example: sample fastq_1 fastq_2 strandedness condition NORMAL_polyA_RNA_SRX1020493 /home/tud03125/pipeline/fq.gz_human/Resource_of_rRNA_depleted_RNA_seq/fastq/SRR2014237_pass_1.fastq.gz /home/tud03125/pipeline/fq.gz_human/Resource_of_rRNA_depleted_RNA_seq/fastq/SRR2014237_pass_2.fastq.gz auto NORMAL NORMAL_polyA_RNA_SRX1020495 /home/tud03125/pipeline/fq.gz_human/Resource_of_rRNA_depleted_RNA_seq/fastq/SRR2014238_pass_1.fastq.gz /home/tud03125/pipeline/fq.gz_human/Resource_of_rRNA_depleted_RNA_seq/fastq/SRR2014238_pass_2.fastq.gz auto NORMAL NORMAL_Total_RNA_seq_SRX12134672 /home/tud03125/pipeline/fq.gz_human/Total_RNA_seq_with_rRNA_depletion/fastq/SRR15843320_pass_1.fastq.gz /home/tud03125/pipeline/fq.gz_human/Total_RNA_seq_with_rRNA_depletion/fastq/SRR15843320_pass_2.fastq.gz auto NORMAL For

nf-core/rnaseq

, having

auto

as strandedness was alright. It was acceptable for

nf-core/rnaseq

pipeline: https://nf-co.re/rnaseq/3.14.0/docs/usage/. But, for

rnasplice

or

nf-core/rnasplice

, it wouldn’t accept it and requests me to identify the strandedness. I was looking into this website for it: https://nf-co.re/rnasplice/1.0.4/docs/usage/. In there, their description is this: "`strandedness` Sample strand-specificity. Must be one of

unstranded

,

forward

or

reverse

." I’m currently having difficulty determining strandedness for this dataset here. But also, the other problem is that some of my datasets are 1) single-ended or 2) paired-ended (like this table above) types. So, like this is one of the datasets: https://www.ncbi.nlm.nih.gov/sra/?term=SRR2014237. Because it's paired end, they have both

forward

and "`reverse`, and I think I was told to just keep

auto

for that case since

forward, reverse

wouldn't work. So, how to determine the strandedness, single-end or paired-end?