Closing remarks

Well, it’s over. Congratulations on making it to the end.

Congratulations!

Here’s a few more pieces of opinionated advice, drawn from bitter experience37:

  • Don’t feel too bad about making subjective decisions in the choice of parameters. Much of scRNA-seq data analysis is exploratory, which is inherently guided by our own interests. We’re just generating new hypotheses at this point so we don’t need to be too rigorous.
  • Because exploration is so open-ended, scRNA-seq data analysis can be quite time-consuming. For example, you might redo the analysis with different parameters, perform subclustering, etc. to examine the data from different perspectives. Make sure you get appropriate recognition for all this effort38.
  • Always validate conclusions with independent replicates and a different (i.e., non-sequencing-based) assay technique. We had a fun time with all the subjective data exploration to generate new hypotheses, but at some point, we need to pay the piper and test all the stuff we made up.

And hey, if all else fails, there’s nothing wrong with a bit of stamp collecting39. You’ve already been paid to generate and analyze the data, so you might as well get it published; perhaps it might help someone else down the line.

  1. As I often told my manager, “fools learn from experience, wise men learn from history.” In the end, I guess the distinction didn’t matter as we both got fired.↩︎

  2. For a study with a major single-cell component, a joint first/corresponding authorship seems to be fair market price for the primary analyst.↩︎

  3. For our younger readers: back in the day, when you wanted to send a message to someone, you would write your message on a piece of paper, put that paper in an envelope (a paper-based packaging device), and request your country’s postal service to physically deliver it to the recipient’s address. Payment for delivery would be denoted by purchasing an adhesive “stamp” and sticking it on the envelope. These stamps would often have interesting decorative features that made them desirable to collectors. In a scientific context, the infamous adage “all science is either physics or stamp collecting” is often attributed to Sir Ernest Rutherford, dismissing the importance of other research fields to their inability to generate testable hypotheses. We feel that this comparison is disrespectful to the fun and noble hobby of stamp collecting.↩︎