I have absolute copy number data for a diploid organism. I would like to distinguish between a Gain and an Amplification. I am currently using the term Gain to define any chromosomal copy number increase from the norm - in this case 2. I would like to know if anyone has an arbitrary absolute copy number cut-off to distinguish between a Gain and and Amplification. For example, would you consider it to be a copy number greater than say copy number 4?
So by your definition you can't distinguish between a gain and an amplification (in a diploid organism) - both refer to any copy number aberration greater than 2. Let's say then for those cases where copy number is 12 you would class in the same manner you would a copy number of 3, and call both an amplification despite any implications this may have biologically?
Part of the problem is that "gene amplification" is an ambiguous term and only has a particular meaning in the context of cancer biology (and even there only recently). There, I've seen suggestions that a copy number of 10 or more is needed to term something an amplification. Realistically speaking, you'd need to ensure that there's a concomitant expression change to really term something that, since even a 100x increase could result in no expression difference (though that'd be unlikely).
Outside of cancer biology, a copy number gain is an amplification (not that you'd normally term it as such).