DEseq2 input from nf-core/rnaseq

Question

[وفاء]

I used the nf-core/rnaseq pipeline in the cluster human genome, and I got some files as the output of Star salmon.

My question is which file I use as input to DESeq2? and where is the full file that contains 58,000 genes?

salmon.merged.gene_counts_length.tsv (29744 genes)
salmon.merged.gene_counts_length_scaled.tsv (29744 genes)
salmon.merged.gene_counts_scaled.tsv (29744 genes)
salmon.merged.gene_counts.tsv (29744 genes)
salmon.merged.gene_temp.tsv (29744 genes)

[GenoMax]

Looks like the salmon reference you used contained 29744 transcripts. There are currently 19396 protein coding genes in human genome as of this month.

[وفاء]

Thanks GenoMax. So, what is the correct salmon reference that I should use?

[GenoMax]

How did you run the pipeline? Which reference did you use?

[وفاء]

I ran the pipeline using the nextflow and the reference I used for GRCh38.

From where I can download the latest version of the human reference genome (I need to use wget in linux but I need the link). I need to use it for RNA-seq analysis

[GenoMax]

There is probably nothing wrong with the reference you used. It may simply not have every transcript in there but it may be sufficient.

You can find transctipt/genome sequence files at:

GENCODE: https://www.gencodegenes.org/human/

Ensembl (only transcripts): http://ftp.ensembl.org/pub/current_fasta/homo_sapiens/cdna/Homo_sapiens.GRCh38.cdna.all.fa.gz

Answer 1

[bioinfo8]

You should use salmon.merged.gene_counts_length_scaled.tsv and here is the reference.

While feeding to DESeqDataSetFromMatrix() use round() for inputing counts.

[وفاء]

Thank you bioinfo8. Also, I am not sure if I use the correct reference genome. So, from where I can download the latest version of the human reference genome (I need to use wget in linux but I need the link). I need to use it for RNA-seq analysis