Final Thoughts and Update About a New Skill

August 01, 2016

The summer has really flown by, and I certainly did not get as much done as I (and my advisor) had initially hoped. The goal for data collection of my work is to obtain many single molecule fluorescence scans of our dye and then conduct “blinking traces” on each individual molecule which plot the relative fluorescence intensity vs. time. We had hoped to have those all our scans finished midway through July but we had some setbacks finishing our control experiments due in part to equipment troubles. Now that my partner and I have been successful at replicating single molecule scans, we have been scrambling to collect as many blinking traces of molecules as we can before the summer ends.
Regardless, this summer was beneficial in that I was able to develop a basic understanding of the data analysis portion of our research. My project uses a Matlab code previously developed by brilliant former researchers that records the relative changes in the intensities of fluorescence in our blinking traces. It records this data as a set of values that represent the time in seconds of various types of “events” such as fluorescing (“on”) events or non-fluorescing (“off”) events. Next, we can have Matlab develop a probability distribution that represents the likelihood of a specific time of an event occurring based on the relative number of times that value occurred in the data set. Lastly, we can hope to describe the data by fitting it to various heavy-tailed distribution functions, which are used because they represent a large data set (from times as small as 10^-2 seconds to 10^3 seconds). Matlab also generates p-values for this fits to give us a better understanding of which functions fit best. I hope that when I return in the fall I can use this approach fully analyze my data and develop a better understanding of how our dye molecule behaves when exposed to our laser.