User:Brigette D. Black/Notebook/Brigettes Notebook/2009/10/12/Malachite Green Errors
The big problem hovering over the bulk assays is why I can't seems to get any decent measurements from the new, good kinesin (they are about the same as the when I was using the Cytoskeleton kinesin).
I think one of the big problems is with the assay itself. The PhosFree phosphate Assay is really great for measuring Pi concentrations over a vast number of possibilities. However, I am beginning to think that it is not sensitive enough to accurately measure the small changes that would result from a kinetics assay.
To test the error in the PhosFree Assay, I have decided to do a standard curve, again, this time only measuring the absorbance values close to our background (about 50 uM). Also, I will be making 3 samples for each point so that we can have a clearer idea of the precision of the PhosFree assay (and help determine if the issues I am having are in fact due to lack of precision).
I made 3 cuvettes of each of the concentrations (in uM of Pi): 0, 37.5, 50, 51.25, 52.5, 53.75, 55, 62.5, 100, 112.5, 125, 137.5
In the traditional fashion I let the malachite green develop for 16 minutes and made a measurement. For the last cuvette measured in each group, I made 6 measurements over the course of 30 seconds (every 5 seconds) to see if there were additional measurements errors with the nanodrop and to see if the dye was still developing at that time.
What I found confirmed my suspicions. I saw that for concentrations within about 5 uM of one another, the results were indistinguishable from one another. Also, even as far apart as 10 uM there are significant overlaps in the absorbance values. This explains in part why every kinetics assay has so much noise in it and why it is hard to gain any real information about the system.
- Andy Maloney 01:13, 14 October 2009 (EDT): First off, great graph. I would really like it if you started showing your graphs with one data point with error bars. It is much easier to read graphs that way.
The solution to this would be to use the more sensitive CytoPhos assay, which can detect very subtle changes in the Pi concentration and is sensitive over a range of 1uM to 15 uM. The problem here is that the background reading of the 1mM ATP alone is about 50uM Pi. Thus, the more sensitive CytoPhos assay is not useful to us as the necessary background components are well out of range of the CytoPhos assay. The user manual for the assay even suggests to keep ATP concentrations below 0.6 mM so as not to over saturate the system with Pi.
Tomorrow, I will try the CytoPhos assay and see if it can distinguish significantly finer intervals (maybe 2 uM?) at half the Pi concentration. If so, then I think the best thing to do would be to dilute the kinetics assays by half and start using the CytoPhos assay.
Also, I wasn't very happy about the values I was getting when I quickly repeated measurements. The numbers were not too far off (about 0.05), but it did seem as though the dye was still developing even that far away. Today I was very very cautious about measuring all of the cuvettes after exactly 16 minutes, so the errors above are not related to time. However, it is a likely source of error for previous kinetics assays where I have measured some samples about 30 seconds late. So I may also do a test of a sample where I measure every few seconds for 20 minutes or so to see how much time is affecting the goodness of the samples.