So maybe THAT'S why we didn't get results...

Reader's Digest version: we've been using one-tenth of what we thought we were using, and concentrations that low probably can't do anything against microbes. And I figured it out!

Longer, richer version: Potential good news today. I was working up in the lab today and made a discovery that bodes well. First, though, a brief overview on the project for those who don't know: I work in a Natural Products Research lab, which means we look at things that come from plants. Specifically, I'm working on a project where we're looking for medicinally-active plants from Guatemala. We grind and mash up the plants to get at the good stuff that's inside, then test that good stuff (the "drug") at different concentrations against five different microbes. We've probably run 60-some-odd plants, and we haven't really had any significant inhibition of microbial growth. Unsurprisingly, this has been a little disheartening.

Recently we had discussed tripling the concentrations that we use for the test to see if more drug will provide better results. Being somewhat anal, I went through (on paper) the calculations to make sure that the method I would use to triple the concentrations would work. I checked my method against the normal way we had been doing it, and all went well at first. Then on one of the steps, my numbers weren't working out. So I tried it again, still no dice. I tried again, using a calculator--still nothing. My calculations showed that we were using only 10% of the concentrations we thought we were using! That means that instead of 1000, 500 and 250 ug/ml, we were using 100, 50 and 25. That's pretty much nothing.

I've been known to mess up on my math before, so I thought about it qualitatively. We were trying to go from 400 ug/ml to 1000 ug/ml by diluting. Yeah, that's right--that's impossible. So I tried to find someone else to check my math with, and was finally able to find my professor who I work with. I showed him what I had found, and he was shocked. This was a fairly simple calculational error in our protocol that had snuck past seven or eight people for nearly a year. But it explained a lot.

When Aaron, Mark and I began working in the lab, we were trained by folks working on another similar project in the lab, with plants from Morocco. They were running these plants for a third time to verify previous results. The PhD. student in the lab had recently shown them a faster way to make their dilutions (different concentrations of their drugs) via serial dilutions (that's for the nerds out there), and the new method is how we were trained. Unfortunately, that was where the error came from. Those two students were surprised when their third set of results didn't match up with their first and second sets. We also were unable to duplicate previous results with a plant from Arizona. And we also have gotten pretty crappy results on our Guatemalan plants.

Today all of that strangeness made sense, and I happened to be the one who figured it out. I was rather pleased with myself. Unfortunately, we'll have to re-run all of the plants that we've done so far. Fortunately, I'm paid by the hour. How's that for job security? And this time we may actually get some good positive results that we can publish and present. Fingers are crossed.