The Salmonella were swimming in the mini channels of the microfluidics device that I had built. Viewing under the confocal microscope, I again set up the program to take images of the cells under different conditions for 24 positions to grow for 24 hours. At ETH Zürich, I had the opportunity to learn how to do single cell analysis via microfluidics technique in the Ackermann laboratory. Excited to learn and use this new methodology to investigate the avirulent:virulent phenotype switching, I read up on the microscope that I would be using as well as the experimental procedure. With nine weeks to complete my experiments, I was eager to get started right away. 

Salmonella in a microfluidic chip

Soon, I would learn about the many ways that I could fail my timed microfluidics experiments. Plasma-binding the microfluidics chip to a slide was a difficult task in itself because I broke many slides. To fix these failures, I had to practice many times so that my clumsy hands would soon become proficient at handling fragile items. Many of my chips also became useless because sometimes they did not bind well or when I loaded my cells into the chip channels, the chip would unbind by itself, thus rendering it useless. I rectified these failures by communicating with my mentors and colleagues about how to best keep the chip bound to the slide. Furthermore, there are also a certain percentage of channels in each of the main channels that should be filled with cells. In this aspect, I also was also unsuccessful many times because each channel was only wide enough to fit a single line of cells. These failures could be rectified by adding a detergent so that the cells could easily slip into the micro-channels. Because the device was so small and the experiment was truly on the micro-scale, any small movement of the device would completely mess up the positions for image acquisition. I also failed in this aspect. This could be fixed by taping down the tubings so that there are no major movements that would disrupt the image acquisition. As I pumped media with different conditions through the channels that would feed the cells, another problem I ran into was the presence of micro-bubbles. These micro-bubbles would change the pressure in the channels and would kill the cells. Thus, I had to be extremely careful with not introducing these bubbles into the condition media.

Rather than easily completing my experiments and analyzing the data, I spent the majority of time troubleshooting experimental failures. I felt frustrated and upset that the experiment and method I was so eager to learn about could go wrong in so many ways. However, with each problem that arose, I communicated with the people around me and tested to solve each of these failures. I learned to be more comfortable with uncertainty, and I learned about the patience and effort that goes into every single experiment. This patience, continuous effort, communication, and creativity are what I hope to bring to all I do in the future.