Transcriptome sequencing

[edwardwong1070584]

Dear all,

I have a naive question. I would like to do transcriptome sequencing using human total rna as sample. I need to calculate the read depth with this formula: Coverage= (read length*number of reads)/genome size. We all know the human genome size is approximately 3 billions. But do we know the size of the total rna? Am I even asking the right question? Your kind advice would be very much appreciated.

Thank you.
Ed

[dpryan]

That's not a meaningful metric in RNAseq, for the reason you mentioned and the fact that even if the exact transcriptome size of a particular cell type were known you'd want enough reads to properly handle the dynamic range in expression levels, rather than caring about an average level.

[edwardwong1070584]

That's not a meaningful metric in RNAseq, for the reason you mentioned and the fact that even if the exact transcriptome size of a particular cell type were known you'd want enough reads to properly handle the dynamic range in expression levels, rather than caring about an average level.

Thanks Dpryan for your reply. So let me to get it right, in your opinions, you would rather to look at the number of reads/ sample from a run than the coverage depth as we commonly look at when dealing with transcriptome sequencing? Actually, I had a similar thought as you. I was taught that the number of reads is important when handling the RNAse, but, I was confused when my PI wanted me to do a transcriptome sequencing on human total RNA to generate a 30-50millions reads and read depth of 50x. I have no issue with the number of reads but I was confused with the read depth. How should I calculate the depth? Can I use the normal human genome size (~3billions) for it?

Thanks again for the advice
Ed

[dpryan]

Just ignore the "50x" part, there's no worthwhile way to calculate it (if your PI complains, just send him/her here to ask).