Why is the scaling factor always million while dealing with RNA-Seq expression?

Question

[Sandeep]

I have a basic conceptual question on RNA-Seq expression units. While quantifying expression using different methods, why is that we always scale by a million and not any other number? I have seen the same in RPKM, FPKM, TPM and CPM.

Can anyone explain the rationale behind it?? Why not hundred thousand?

Thanks

[GouthamAtla]

Might be because the scale we use in NGS data is million ( 'n' million reads), the scaling factor used is million ( per million reads).

Answer 1

[Devon Ryan]

1 million is a nice round number and the range of resulting values is more or less convenient.

[Sandeep]

Is that it? Isnt there any logical explanation as to why 1 million was chosen to be most convenient and nice round number? We could probably use 10^5.

I guess there is something more to it. Probably we need a statistician to answer the same.

[Devon Ryan]

You can ask a statistician and you'll get the same answer. There are many arbitrary values that we use every day.

[Alternative]

Yes you could by how would you call it then? "Reads Per one hundred thousand ?". RPM was used because it make the number looks nicer, easy to define/pronounce.

[matted]

There's nothing more to it. Why do we give distances in kilometers and not megameters? A mixture of convenience for the quantities we typically report and historical inertia.