The Art of Presenting Failiure

When I saw that we were using the bacterial photography system developed by people in Chris Voigt's lab for module 2, I was pretty excited. I was actually familiar with this system from freshman year, when we generated images using the wild-type strain during the biological engineering FPOP (shout out to my awesome first ever lab partner, Ellie Laukaitis!) So when we started the module, I was confident about the performance of this system. Plus, from our experiments in module one, it seemed like results in 20.109 were fairly predictable.

This didn't quite turn out to be the case... Module 2 was also very well organized similar to module 1, and there was an unified goal for the experiments we were performing: optimization of the photography system. But the biggest difference was that experimental outcomes for each group were meant to be unique. And since not all outcomes can be successful, some groups were bound to get results that did not fulfill the projected goal.

But even before reaching different (and potentially inconclusive) outcomes, we-as the whole TR section- faced a few fundamental "road blocks":

After our whole section got results that completely contradicted any expected outcome, we realized that there might be an anomaly either in our bacterial strains, or in our methods for mutant generation. So all the results we had until this realization were suddenly deemed erroneous.

But there was hope! We ended up running experiments testing system performance using  mutants strains generated successfully by a former 109 section. So my lab partner Alyssa and I picked 2 of these mutant strains at random and set up our experiments.

It turns out our mutants didn't result in system optimization. But that was fine! There was a large amount of class data  to analyze, and I was expecting one mutant strain in particular to definitely improve the system. Guess what: this strain was also unsuccessful in reaching our goal for the module.

So there I sat at the beginning of the long weekend, with large amounts of data demonstrating how we had failed at optimizing the photography system. How was I supposed to base a WHOLE research article on failed results?

Well, it turns out, there are actually whole journals dedicated to experimental failures (Check one of them out!) I had heard people in lab mention these, but up until this weekend, I thought it was a joke. But after actually taking a look at some articles, I understood that presenting failed results could actually be quite beneficial for the scientific community. People can learn from the mistakes of others, and save time by avoiding experiments they know will fail.

Also, presenting failed results is significantly more difficult than writing about expected outcomes. But it is absolutely a very useful skill, as it allows researchers to see their experiments and results in a more objective manner. So even though things didn't go as planned, and we didn't get the results we wanted in Mod 2, we definitely acquired new skills we will need for better scientific writing. And on that note: