Finishing Up and Saying Goodbyes (Weeks 9 and 10)

Sorry I didn’t get a chance to put a blog out last week, so I am combining the last 2 weeks into 1 final blog post. Throughout these past 2 weeks, I had to do a lot of analyzing and tying up of loose ends. I figured out the memory issues, figured out the extraction of the individual peaks to get the area under them, programmed labview to be more normalized from sample to sample, ran all of the collected data through the labview program, saved all the data, analyzed the data that was outputted further (made graphs and other sorts of things), and put together a final presentation. With all of the troubles with computer memory and installations, I was super happy that the program that I wrote was consistently working the way that I wanted it to.

The two main aspects that I looked into further were the peak heights and how they related to each other from sample to sample as well as the area under the extracted individual peaks. As the particle size increased, so did the peak height voltage readings. There were sort of distinct ranges that the different size of particles fit into with the different voltage peak heights, but there is too much variation to have set defined ranges that they have to fall in to. The other interesting thing with the peak height was comparing the latex bead data with the phytoplankton data of similar sizes. The phytoplankton of similar sizes to the beads produced more scattered peak heights in the upper voltage ranges. I concluded that this was due to the phytoplankton forming chains of 2, 3, 4, or more and when these chains go through the microflow cytometer, it reads it as a bigger particle (i.e. higher peak height).

Another thing that I investigated was the area under the individual extracted peaks. What I found was that as the particle size increased, the area under the peak increased as well. I also found that when binned into a histogram, the areas follow a normal distribution. Good cultures of phytoplankton follow a normal distribution when looking at size, so this was a good observation to see in the area calculations. The last thing to note with the area is that the different morphologies of the phytoplankton had different shifts in the areas even if they were similar in size. The longer barrel shaped Thalassiosira sp. had larger areas under their peaks than the Chlorella, for instance, that are spherical. This too, could be explained by the formation of the chains in the Thalassiosira sp.

The final major calculations involved the concentrations (particles/ml) with both the FlowCAM and the microflow cytometer. Since I ran the 2 machines using the same pump, they had the same flow rates. I used this flowrate along with the particle count from the FlowCAM and the number of detected peaks (i.e. particles) from the microflow cytometer and the elapsed time while the sample data was collected to calculate concentrations for both the FlowCAM and the microflow cytometer. The microflow cytometer had consistently higher concentrations than the FlowCAM for the same sample. This was also a good finding as the FlowCAM doesn’t detect all of the particles that are in the sample. The microflow cytometer is more accurate at determining particle counts if you have the proper tools to count the number of particle (which is what my labview program will do).

So by using the peak heights in relationship to the areas under the peaks, you can identify what types of particles are in the water and by doing the calculations, you can find out how concentrated your water sample is. This is important as it can help to identify harmul algal blooms and for routine water assessments (how clean/dirty the water is).

This summer went by super fast and now it is over. I met a lot of nice people here at CMOP and it is bittersweet to leave. My principal investigators Joe Needoba and Tawnya Peterson were great to work with as well as everyone else in my lab.