Since once in a while the topic semi-quantitative Western blot analysis pops up and many people in biological sciences who do Western blots or gels are into that, I wanted to share some information on that. So the following topic is not directly linked to ImageJ necessarily but many of this community might still be interested in. And I think the topic is important in the aspect of reliability and reproducibility in science.
Reason no. 1: Many students try to find a guideline for Western blot “quantification” online while using ImageJ. That’s why I posted it here!
Everything I have found in the web so far was mostly going a in the wrong direction leading to non-reliable measurements!
Reason no. 2: The ►Analyze ►Gel tools and the ►Analyze ►Calibrate… function provide theoretically everything what you would need but a clear guideline is still missing. So, even while often suggested, drawing boxes around a band is not the proper way!
So, here a few recommendations and very helpful links (since this is not my invention):
1.) have a look at the following video seminar from Aldrin Gomes as an introduction to the problem on the experimental side:
2.) Here a few publications which help in getting a better understanding before the actial image analysis
- Quantifying Western blots: Pitfalls of densitometry
- A Defined Methodology for Reliable Quantification of Western Blot Data (Here you see the analysis methos indicated)
- Evaluating Strategies to Normalise Biological Replicates of Western Blot Data
- Western Blotting Inaccuracies with Unverified Antibodies: Need for a Western Blotting Minimal Reporting Standard (WBMRS)
- The design of a quantitative western blot experiment
3.) Since I first tried to code a plugin which would combine the necessary functions for ImageJ, I stumbled over an existing and extremely good tool (also Java) with all the things you would need to run such an analysis (including calibration). So, why reinvent the wheel. It is freely available but not completely open source with access to the source code. I will try to get in contact with the authors and see if there might be a possibility to link it as plugin into ImageJ (e.g. as an upate site). It also produces publishable figures which indicate exactly the quantification areas which is extremely important for a transparent measurement.
Sorry, for being a little off-topic but I think this community is the right place for this information and some others might add some comments as well so that those seeking for a guideline get an anker point.