Rows of read count files

Question

[Abhishek]

Hi all,

I had previously worked with transcriptomic data containing microRNA info. My understanding was that the read count file I obtained from a public dataset contained miRNA names as rownames.

Now, while actually trying to create a read count file from fastq file, I realise that the final read count file should contain the names of corresponding DNAs, and not the original rna transcripts.

So my understanding is : for rna sequencing of coding RNAs, the read count file would contain as rownames the corresponding DNAs. However, if the read count file is for non-coding RNA, then the rownames will be the RNA transcripts themselves. Both these types of read count files are considered to be transcriptomic data

Is this understanding correct ?

Answer 1

[swbarnes2]

I'm not sure what you mean. The count file for RNASeq has gene names as rows and sample names as columns:

http://bioconductor.org/packages/release/bioc/vignettes/DESeq2/inst/doc/DESeq2.html#count-matrix-input

[Abhishek]

Thank you for your response. Apologies for the lack of clarity.

My study aims to identify differentially expressed sncRNA between groups.

So I was hoping to create a read count file with rownames as sncRNAs (similar to the normalized count file in https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE122488 )

In this case is it still necessary to map it to genes ? I'm not really sure how to create a read count file similar to the link I added

I am using Salmon to quantify the reads from fastq files.