As you now know from our last entry, I’ve got lots of data. I’d like to graphically represent my data so that I can show lots of people my findings and have them go “Ooooo, Aaaaa. Tell me, young scientist, what does it all mean?” I can then look like a smarty pants by explaining to them that by plotting some quantity on the X-axis (the horizontal line) vs. something else on the Y-axis (the vertical line) we can make BROADER GLOBAL IMPLICATIONS with this data. Scientists love to graph things. Anything and everything looks better on a graph. It also helps the scientist presenting the graph to summarize their findings in order to avoid needless rambling. Scientists love rambling almost as much as they love graphs.
Let's build a graph together to better illustrate my point. We talked earlier in the week about auto-analyzers and all that jazzy business of precision and why auto-analyzers are rad if you have a buttload of samples, but not that rad if you want high precision. The data points we're gonna use are gotten by pumping samples with known concentrations through your auto-analyzer. The auto-analyzer spits out what we’ll call an Arbitrary Number. Your results might look like this:
38.41 uM -----------------------0.205
76.81 uM -----------------------0.410
101.51 uM ----------------------0.560
What we have in the left column is the concentration of our chemical in micro-Molar units. That’s 10^-3 moles per liter of solution and remember that a mole is 6.022 x 10^23 units (molecules or atoms), so we’re really talking about how many molecules are in our sample.
On the right is the Arbitrary Number that your auto-analyzer might display once it “digests” your sample. I say digest as an purely imaginative description because these machines do not actually feed on your sample. Or do they? Oh my.
Anyways, the next step is to arrange our graph. We’re going to put the Arbitrary Number on the X-axis and the Concentration on the Y-axis. If you reach WAY back into your mathematical memory, you’ll find an itsy-bitsy morsel of knowledge which tells you that when making a plot, the independent value goes on the X-axis and the dependant value goes on the Y-axis. Don’t remember? Well, I won’t hold it against you. Here is your data plotted by itself.
What happens next is that you, being the intrepid and fashionable young scientist you are, use the linear fitting function in Excel or some similar program to get a line running through (or close to) all your data points.
This line has an equation to it. Without getting into the nitty gritty of it, you’ll be able to use this equation to translate all the rest of your data into concentrations. All that’s left to do is eat Cheetos and let the auto-analyzer do its thing.
I think congratulations are in order because you just made your first calibration curve!! YAY!