Hi everyone, I have a question regarding the

samplesheet.csv

format for nf-core/rnasplice. The documentation indicates that “strandedness” is a required column, and the example header is:

sample,fastq_1,fastq_2,strandedness,condition

I have paired-end RNA-seq data for a single sample, where R1 is the forward read and R2 is the reverse read. Should I include one row for this sample (with both

fastq_1

and

fastq_2

specified), or should I provide two separate rows — one for each read file? Thank you very much.

You put each paired-read sample into one line (unless you have run it split over separate flow cells or lanes, or something, in which case you have a line per file pair). The R1 file goes under

fastq_1

, the R2 under

fastq_2

. If you know the strandedness of the kit, use that (if R1 is aligned to the coding strand of a transcript, use

forward

), otherwise try to figure it out, best by asking the people who made the library, or e.g. by mapping a smaller subset and looking at the direction of reads in exons. The #CE8SSJV3N pipeline has an

auto

option, but I think rnasplice doesn't have that yet.

Really appreciate @Thomas Danhorn I checked the RNA-seq report and found that it contains following message. Based on my limited biological know, should I conclude it as "forward"?

The way I read this is that the first strand is complementary to the mRNA, and the second strand a copy of it, and that one has sUTPs. But I'm not sure how exactly that determines the directionality of the adapter ligation. It does sound like the standard Illumina kit though, and that seams reverse stranded. I would take a few reads and map them, ideally with STAR, but even if you BLAST them you should get some insights. If the exon/transcript where the read maps is on the

+

strand (i.e. you can read the sequence left to right) and the mapped (R1) read also goes left to right, it is

forward

, if not, then reverse. Note that a small percentage (~2%) of reads typically map to the "wrong" strand, and some kits have an equal mixture of both and are unstranded (but yours does not sound like it), so check more than one read.

Thank you so much for the clear and thoughtful explanation, this really helped me understand the logic RNA sequencing. I’ll read through the Illumina documentation and try your suggestion to verify the strandedness using alignment tools. Thanks again for taking the time to walk through the reasoning so patiently! 👍❤️