Calculating fold change...CONFUSION

Question

[lmmjrf]

Okay, feeling very dumb. Why do some people calculate fold change like this (B-A)/A and some others like this (B/A)?

I am trying to manually calculate the fold change for an RNA seq experiment and am reading conflicting ways of calculating this seemingly super basic measure.

HELP.

Answer 1

[Devon Ryan]

I have never seen anyone use "(B-A)/A". While it's technically a fold-change, it's a confusing one since a value of 1 indicates that B is twice A. The standard way to manually calculate a fold-change is B/A.

[jordi.planells]

You can calculate a value between -1 and 1 (-1 < x < 1) by using a similar formula (B-A)/max(A,B). Could this be what you have seen?

[Devon Ryan]

Presumably you meant to direct that toward OP.