Salmon quant with netflow nf-core/rnaseq -r 3.13.2

Question

[Hojn]

Dear community,

I am using this nextflow in order to produce count matrices from Smart-Seq data to analyze it further in Seurat. I went through documentations (Salmon/Nextflow) but couldn't really find what I wanted, so I give a try here.

First, I didn't really find a pipeline from SmartSeq to Seurat, if you know one with critical parameters to look at, I would appreciate. Second, I have hard time figuring out what are the outputs of traditionnal Star - Salmon, I mean how they are produced. I end with 3 files for Salmon quantifications (salmon.merged.gene_counts, salmon.merged.gene_counts_length_scaled and salmon.merged.gene_counts_scaled).

What are they, how were they produced? I currently use the first one since it does not seem like normalized, but I do have digits in gene_counts and I just want to know why (I guess it's about the quantification method?) and wonder if it's correct to use it that way.

Should I use the length_scaled one? What are differences between length_scaled and scaled?

Thanks in advance for your time

Answer 1

[Barry Digby]

I have hard time figuring out what are the outputs of traditional Star - Salmon, I mean how they are produced.

Excerpt from: https://nf-co.re/rnaseq/3.12.0/docs/output#star-and-salmon

All you need to run Salmon is a FASTA file containing your reference transcripts and a set of FASTA/FASTQ/BAM file(s) containing your reads. The transcriptome-level BAM files generated by STAR are provided to Salmon for downstream quantification.

Regarding the three files (https://nf-co.re/rnaseq/3.12.0/docs/output#salmon)

According to the txtimport documentation you can do one of the following:

• Use bias corrected counts with an offset: import all the salmon files with tximport and then use DESeq2 with dds <- DESeqDataSetFromTximport(txi, sampleTable, ~condition) to correct for changes to the average transcript length across samples.
• Use bias corrected counts without an offset: load and use salmon.merged.gene_counts_length_scaled.tsv or salmon.merged.gene_counts_scaled.tsv directly as you would with a regular counts matrix.
• Use bias uncorrected counts: load and use the txi$counts matrix (or salmon.merged.gene_counts.tsv) with DESeq2. This does not correct for potential differential isoform usage. Alternatively, if you have 3’ tagged RNA-seq data this is the most suitable method.

---

* When they say "import all of the salmon files", I assume this means the quant.sf files generated for each sample. See here: https://bioconductor.org/packages/devel/bioc/vignettes/tximport/inst/doc/tximport.html#Salmon