ImageJ vs. LI-COR Odyssey - Quantify western blot results (Aug/06/2009 )

Hi all, this is my very first project and I hadn't have any experience at all.

I am currently using IR-dye secondary antibodies purchased from LI-COR to detect some soluble proteins. Not sure if many people are familiar with the Odyssey software, but we use it to scan our membrane and to quantify the bands.

We are trying to compare between 2 sets of 8 samples, so I ran 2 western blots simultaneously all the time. However, no matter how similar the procedures were, there was still some huge difference in band expression. For example, I ran 40ug of proteins for all samples, but I would get an average intensity reading of 80 on one membrane and 120 on another for the housekeeping gene. If even the housekeeping gene readings are so different, how can we expect the soluble protein readings to be accurate? (this is a question)

Then I tried using ImageJ to re-do all measurements. This time the numbers are much similar to each other, but I still don't understand how the area and percent work and what they really mean. And my PI does not agree that ImageJ makes more sense than Odyssey.

Thus, I am now very confused of which program is more reliable.. Any suggestions?

Thanks and sorry for the long posting.

P.S: I incubated these membranes with primary antibody A, read it, then primary antibody B, read it, then lastly with housekeeping gene.

Welcome to the world of westerns, they aren't really quantitative in my experience, you would have to have exactly the same conditions on both gels and blots for it to work.

By exactly, I don't just mean running at the same time from the same gel prep and in the same buffer etc.. I mean your gels would have to be identical and run with identical amounts of buffer and blotted in the same position on the same blot apparatus in the same buffer, blocked, washed, etc. for exactly the same times and so on. Rant not aimed at you, just general frustration at similar issues I have had in the past! In the end I learned that the technique is OK for one blot only, you can't compare between blots.