Teaching Tips from Seminar

I've learned that scientists, as a group, give pretty crappy talks. They love data and want to show you all the bits and pieces, even if there's really not time for that. They tell you things they have been working on so long and in such detail that they have forgotten what it's like to be their audience. They just dump information onto the digital page and expect you'll understand it. Instead of understanding in an instant, the audience is bombarded with new information that they need to process very rapidly before the next slide comes up with yet another information dump. This makes seminar a gamble. You spend an hour or more sitting in a dark room hoping to hear something interesting and knowing there is a possibility the best thing you'll get is a lukewarm cup of tea.

Friday's biophysics seminar was presented by Peter Chien of U Mass Amherst. He told us about recent work in his group that completes an 8-year story: they have been chasing down a mechanism for how certain proteins are broken down1 in Caulobacter crescentus bacteria.2

It was excellent.

 Model of a  Caulobacter crescentus  bacterium. Image by Flickr user  AJ Cann .

Model of a Caulobacter crescentus bacterium. Image by Flickr user AJ Cann.

Dr. Chien did tell us a lot of new things, but he didn't overwhelm. Here are some of the things I thought he did really well.

  1. He used good (and funny) metaphors. When explaining how a protease3 works, he compared it to a 3-year old playing with a garbage disposal. "The 3-year old decides what goes in, but it doesn't matter what it is: once it's in there, it's gone." His comparison made the semi-abstract molcular interaction more tangible: the audience could imagine the kid shoving things into the sink and immediately understood how the different parts worked together. You might not remember the particular name of the protease (ClpXP, if you're curious), but you'd definitely remember the 3-year old.
  2. He got his audience to participate. As he was explaining Caulobacter's asymmetric cell division,2 he paused and asked the audience. "Who has heard of Caulobacter?" A fair number of people raised their hands. He then asked "For comparison, who has heard of E. coli?" People laughed, and basically everyone raised their hands. The poll of the room didn't really teach us anything new, but it kept us engaged. We were now part of his presentation, and not just passive listeners.
  3. He used colors and images as labels. When he talked about the many different proteins involved and the states they were in (e.g. phosphorylated vs dephosphorylated), he represented each protein with a different shape and color. Sometimes I think biologists will represent anything and everything as an elliptical blob, but shapes and colors are much easier to distinguish at a glance than protein names—which he kept referring to as "alphabet soup." It could have been alphabet soup, but it wasn't; it was very clear.
  4. He was consistent. The first time he showed some gels, he explained which row and column was which. The next time he showed us gels, they were in the same place, with the same size, style and layout of labels. He was consistent with diagrams, too. The protein that was represented as a green circle on slide 10 was still a green circle on slide 30, and nothing else had been a green circle in between. This meant that we could learn the new information about the biophysics and not also have to learn how to interpret the current slide.
  5. He didn't let the audience get lost. He would tell us about one piece of the mechanism in detail, then refer back to a diagram of the mechanism, adding elements each time he introduced a new piece and never putting too much up at once. I'm pretty sure you could have walked in 40 minutes late and you could still have figured out what was going on.
  6. He told a story. It had a beginning (Here's what was known when we started working on this), a middle (We tried some things and they didn't always work) and an end (Here's what we think is happening now). It had a cast of characters, too, both human (Dr. Chien, his colleagues and his students) and molecular (ClpXP, CpdR, CtrA, PdeA, etc.).

Dr. Chien's talk was an excellent seminar, but I think it was also an example of excellent teaching. The things that made Dr. Chien's seminar great would make any class better:

  1. Explain with metaphors to turn abstract ideas into something tangible and relatable.
  2. Encourage participation to have learners take an active role.
  3. Use visuals to express and organize ideas.
  4. Be consistent to promote content learning over deciphering.
  5. Provide context to make connections to related ideas.
  6. Tell a story to give the learning a familiar structure.

If I were given the opportunity to take a class from Dr. Chien, I think I'd do it, pretty much regardless of topic.4 I'm pretty sure I'd learn as much about effective speaking and teaching as about the course material.


1: Cells don't just make proteins, they also need to break proteins down. Cells make proteins in response to different events or stimuli, and when the proteins aren't needed, they are disassembled. If proteins are allowed to build up in the cell, they can cause lots of problems, so degradation machinery is pretty important.

2: Caulobacter crescentus are bacteria that look kind of banana-shaped. If you remember high school biology, you may recall learning that bacteria are single-celled organisms, and they reproduce by splitting into two identical cells. Those two cells split into two more cells for a total of four cells, and on an on exponentially.

This model of cell division is pretty common (think E. coli), but it's not universal among bacteria. Caulobacter split asymmetrically. One cell splits into two, but one of those gets a stalk (which is why I think they look like lumpy bananas), and the other gets a tail (flagellum). The cells with stalks are called "stalked cells" (surprise, surprise) and the cells with tails are called "swarmer cells." Here's the next interesting part: only stalked cells can divide. Swarmer cells aren't a dead end, though, they can turn into stalked cells and then divide also.

3: protease: a protein that breaks down other proteins

4: I had a physics professor at IUP (Dr. Haija) who was so engaging that I'm sure I'd have learned something fascinating even if he'd started lecturing about belly button lint. He taught the Modern Physics class I took, and I liked it so much that I took Optics just because he was teaching it.