Bookmark with More
Category 'Data-Analysis'
Status: Unanswered
Views: 1928

Divide or Subtract?

Graham Pockley (Author)

Monday, November 19, 2012 6:35 AM

When trying to relate positive statining to that of the isotype control across a series of experiments, which is the best approach? Subtract the isotype or divide the positive value by it?

Michael Daws

Monday, November 19, 2012 8:29 AM

If you want a relative value, then you should divide, but I am not sure that it is entirely linear. You could test this by adjusting the gain in the relevant fluorescent channel to set your isotype at different values and then read off the value for your positive staining. The problem will be that you will need to adjust the compensation at each gain setting, and this may affect the linearity. It's easy enough to test, and if you do, post the answer :)