As a lab, we have been banging our heads against a basic technique for a couple of months. This is a method I have successfully used countless times over the past 4 years and based a substantial part of my post-doc around. Both of my students are using this method in their respective projects and it has been a major roadblock because we cannot seem to get results. Positive controls work just fine, as if mocking us, but the data we care about defies our grasp. In one case we thought we would have to troubleshoot a bit to get publishable data, but in the other the technique is so routine that there is no conceivable reason why it would not work, even if we had to tweak it here or there.
Why no data? How are we treating the positive controls and the sample differently? Normally they are treated identically, but because of the multi-step sample prep, we tend to take a short cut with the positive controls for the final step, but there are controls along the way to indicate that the early steps are working. All the controls were fine, and so around and around we went, looking for the problem. What is the fucking difference?
We have a conference coming up next week, and this last piece of data would really demonstrate that we're not blowing smoke here. We've changed every variable we could think of and started from scratch multiple times and yesterday afternoon we were met with the same result. Nothing. One of my students and I were sitting in my office wracking our brains to figure this out and the student blurted out "What about substance X, that we take for granted because it is such a basic part of every procedure we do in the lab? It works for everything else, but it might be affecting this procedure because it's different in Y way. Were you using substance X in your old lab or an alternative?" All I could think was "holy fucking cannoli, I bet they're right. Why didn't I think of that?"
I didn't think of it because I've done this a gazillion times and there is no reason to think that substance X would be a problem because it works in everything else we do. But we don't use substance X on the final step positive controls and there is a distinct possibility that the difference between substance X and the one I was using in Post-doc lab could be blocking what we are trying to do. I will know tomorrow for sure.
Oddly enough, I am not arguing for having a positive control that is taken through the whole process, because if the positive in the final step was not working we would still be scratching our heads as to what is going on. It's the fact that we had a difference between out final step positive and the sample that made it possible to figure this out. While I'm still not sure this is the problem, it might be the only rock we haven't turned over and I'm not sure I would have thought to do so without talking it over with my students a dozen times until one of them tossed the idea out there. If this works, I owe this student some beer next week.