Module 2 imposed upon us some constrained writing, though the constraint was less amusing and probably more sensible. We were tasked to produce a research article from a set of data that was in many ways incomplete, and to draw out from that data some sort of useful observation in a convincing way. Our data set lacked controls for cut topology between different putative NHEJ inhibitors, so we weren't in a position to compare them. Plenty of the data generated no useful results, likely from operator error. My first instinct upon seeing the data said to do more experiments. There's nothing interesting to say here--we need more data.
And that was probably true. The studies we produced were far from publishable quality as far as data is concerned. But that wasn't the point of this exercise. I can see two plausible reasons for structuring the assignment the way it was. First, it forced me to realize that there was in fact something moderately interesting to say, even with a questionable amount of data of sub-par quality. Second, it pushed us to write under a pretty formidable and realistic constraint: say something both interesting and valid. Make the data inviting and amenable to analysis, but do it in a way that doesn't fudge or cut corners. I'd say I did at least a passable job at this, but it took a lot more work than I anticipated, and that's ok with me. At least when at some point in the future I prepare a dataset for publication, I can think back to this assignment and appreciate the luxury I will have then of being able to improve my data quality so that the job of writing it up becomes easier.
This assignment was tough but fair.
Yes, a hard, a crazy difficult one.
I had to round pi to make that one make sense.