I often use IF to develop a-tubulin as my housekeeping gene/loading control and I use ultrasensitive ECL to develop weakly expressed proteins on the same blot. Does mixing these two development techniques create any problems for data analysis?
As long as the assumption of relative linearity holds, I believe you should be fine. Perhaps better than fine since your loading control likely benefits from using the less saturating IF signal (thus exposing more subtlety in intensity to accomplish its purpose as a bonafide loading control) while your weakly-expressed target signal is brought nicely out of hiding to be assessed with more precision using the ultra sensitive ECL reagent. I think you are getting the best of both worlds with your approach. Interesting way to do this - I like the idea. Densitometry, as I understand it, boils down to just a "black and white" evaluation for the most part, so whatever you can do to make your signals fall into the range of "not too dark," and "not too light," is key -- and is what I believe you are actually attaining here if I understand this correctly.
As long as the assumption of relative linearity holds, I believe you should be fine. Perhaps better than fine since your loading control likely benefits from using the less saturating IF signal (thus exposing more subtlety in intensity to accomplish its purpose as a bonafide loading control) while your weakly-expressed target signal is brought nicely out of hiding to be assessed with more precision using the ultra sensitive ECL reagent. I think you are getting the best of both worlds with your approach. Interesting way to do this - I like the idea. Densitometry, as I understand it, boils down to just a "black and white" evaluation for the most part, so whatever you can do to make your signals fall into the range of "not too dark," and "not too light," is key -- and is what I believe you are actually attaining here if I understand this correctly.