Should irrelevant RNAs be removed prior to putting the gene quantification data into INSPEcT? For example, samples often have varying levels of rRNA contamination or spike-in RNAs used for QC. I know that I can simply ignore the kinetic rates associated with these RNAs, but I'm not sure if having them in the dataset would affect the scaling between libraries and the downstream rate estimates. Should these contaminant RNAs be removed before running the analysis?

Thanks,
MK

Hi,

It could be worthed to remove RNAs or other possible contaminating RNAs before aligning the raw reads to the reference genome. For example, in many works, the raw FASTQ files are aligned to the ribosomal genome (for humans you can use https://www.ncbi.nlm.nih.gov/nuccore/555853) and only the unaligned reads to this genome are then aligned to the reference one. This helps to prevent spuriously aligned reads to the genome of interest.

In the case of already aligned reads, I honestly don't know the impact of ribosomal genes in the scaling of the libraries. Potentially they might have a role, being highly expressed. If you try and there is a difference in the scaling, please let me know. In that case, I would totally prefer to scale the libraries according to the transcriptome depleted from rRNAs, which I guess is more informative for the processes that are usually studied.

Best, Stefano