Learning the language

Sometimes I think about how just a year ago, I was still in 7.013 and drawing Punnett squares. My knowledge of biology was largely surface-level and limited to terms and definitions, and I had no real understanding of what biological engineering actually entailed, even as I declared my major. Today, I’m a few steps closer to my goal of being a Good Scientist who does Good Science, but the incredible amount that I still have to learn really helps keep me from getting too big for my britches.

In biology, there are definitely unifying principles that go a long way towards general understanding, but there’s also staggering amount of detail unique to every single one of more specific sub-fields.  This means that every time you have to learn about something new for your UROP, or start a new module in 20.109, you’re basically a novice starting from the beginning again. And the process of going from novice to somebody who actually knows what they’re talking about is not a simple one. Here’s how that process happened for me in Mod 2, with a few illustrative graphs.

Way back at the beginning of Mod 2, we read that paper where they found a gene that had synthetic lethality with another gene that was knocked out in a certain kind of cancer. I ended up spending most my time on the introduction trying to understand what they were even saying, to say nothing of trying to understand their actual results. Part of that was down to the paper not being very user-friendly, but as I somehow forget every time, that’s also just what happens when trying to learn the language of an unfamiliar field. It tends to look something like this: 

Figure 1. It takes a long time to really wrap your head around a new concept or idea. A large amount of time is spent collecting the facts and trying to assemble them into a coherent picture, which often coalesces in a moment of brilliant insight, after which additional details can fit into the preexisting structure. Data were collected empirically; statistical significance not shown.

After this initial period, you move on to applications: in this case, experiments and data analysis. New information and data come in that have to be fit into your existing understanding of the concept. In this intermediate phase you often find that you misunderstood a detail before, or assumed something that wasn’t necessarily true, or it’s been a few weeks and you’ve forgotten exactly how synthetic lethality works and what does all this R code even mean? What is this graph saying, and how does it relate in any way to HR and NHEJ?? At which point you have to reevaluate things a bit: 

Figure 2. The initial burst of clarity fades as new information and nuance is introduced, and some time is spent re-examining the evidence before an updated picture emerges. This process may go through a few cycles before true understanding is attained, and you can never really be sure how far along the x-axis you are.

And now you’ve finished all your experiments and it’s time to write your report. Here, you’ve entered the final stage of synthesizing your data and the theory into a coherent story. You’ve finally become conversational in the language, and are able to make logical leaps based on your understanding of what’s going on in the cell and what the data are showing. Although the data are comically contradictory, you’re able to write thoughtfully about it and hopefully put forth some interesting hypotheses. And that’s it; Mod 2 is over and you’ve written a whole fancy-looking research article! One step further along the path to Good Science.