Educating the next generation of biotechnologists

By: Tony Kulesa

In 48BC the Library of Alexandria burned to the ground, and with it, the extent of all human knowledge. Or so the legend goes. If all our knowledge burned today, where would we start rebuilding? What is the one fact worth tattooing on our forearms, such that even as our servers and notebooks are purged of all thought, all history, we will never lose?

Richard Feynman once said if he had one fact with which he would start science over again with, it would be the atomic theory of matter. I think I’d be hard pressed to disagree. At least for molecular biologists, the fact we might most hang onto is what’s going on far below where our eyes can see: this picture of molecules chaotically vibrating, but forming complex structures of the crystals in my watch, the grayness of the clouds, the grain of the wood.

Perhaps the more practical question is, not what knowledge we might save, but what we would hope to do with it. How would we imagine the generations after us rebuilding science? Contrast the vision of the last paragraph with the current progression of science education in K-12. In these formative childhood years, we are also attempting to start with a few facts accepted as truths, plus some limited years of experience in the world, and rebuild much of the extent of human knowledge.

Are we starting with the wrong facts? Why is high school physics about skateboards and rockets, magnets and batteries? Why is biology about nephrons and tundras?

Are we building knowledge in the wrong direction? Why do we save the most fundamental, physics, for post-calculus, when even without it Archimedes and Galileo gained superior understandings of our world.

To serve different interests, we may have different answers. But what our government has deemed our most worthwhile interest via the language of funding dollars, is poorly served by the current paradigm of K12 science education. Perhaps we can use this exercise to think about and discuss the skills, experience, and intuition that we deem most essential to biology (ical engineering). In my experience working the lab, this is the molecular, statistical thermodynamic intuition about biomolecules and larger biological systems.

Students are missing this molecular intuition for these systems until years after an undergraduate degree. To them, DNA is the white goop in strawberries, or abstract, lego models of double helices. Unlike electrical engineering, where one can lick a 9V battery and feel the force electrons that powers their lightbulb, how can we give students the same intuition for proteins vibrating, the squishiness of membranes, the fractal structure of the human genome, or the grip and pull that cells exert on substrates?

If we start with the fundamentals of the what the microscopic world “feels” like, we might find the answer.

Posted in Uncategorized and tagged , .


  1. Thanks for your thoughts, Tony. I’ll share my thoughts as a physics major turned biologist.

    I think the thing that a physics degree did most for me was to provide a strong sense of orders of magnitude. Most of undergrad physics is looking an immensely complex physical object (say, a spring, with all its complicated atomic, molecular, and material properties) and conceptualizing it as a little squiggle that obeys Hooke’s law. No one ever said that all the other stuff about the spring wasn’t important, we just said it didn’t matter in the circumstances we cared about (like, how far it will compress when I put a little weight on it).

    This, rather than the physical intuition, is the thing that I’ve found most helpful as a biological engineer. Biology is the study of immensely complex things. My impression (fair or unfair) of biology students is that they have an intuition that they don’t need to understand the lower strata of the physics (e.g., how entropy works) in order to understand the TCA cycle. Conversely, I’ve found a frustrating lack of intuition about the relative importance of biological objects, that is, what things in the organism or ecosystem can we ignore or treat as unchanging for this experiment.

    This is the most important skill I’ve gained in BE: figuring out what the sensible control and treatment groups are for an experiment. These basic experimental skills are what I think scientific education should be based on, since it will serve both the future scientist (who can layer on the facts of their particular discipline) as well as the future voter (who can be taught to be skeptical and curious about, say, the health effects of GMO crops without having to be taught specifically about genetic engineering).

    • I like both your points about 1) thinking about orders of magnitude 2) experimental design.

      Regarding orders of magnitude, I wonder what some good exercises would be then to teach this skill to younger students, especially in the context of biology? I agree from my experience interacting/learning from physicists, that this is indeed a skill that gets heavily emphasized to anyone who goes through at least an undergraduate physics major. In high school I don’t remember teachers ever breaking down systems this way though, it was only later in college.

      On experimental design – if there were 3 core biology experiments that every high school student should do, what would they be? I was really excited when I took my first MIT biology class and found that it was 100% about designing experiments. At Rutgers, the classes were just memorization and multiple choice testing, even at the senior level biology classes (super frustrating!). If I think back to high school too, I never remember doing any real experiments in biology classes – just more memorization.

  2. I want to echo the last comment made by Scott–I think the most important skill I’ve gained as a biological engineer is learning to think about how to design an experiment. Which details can we abstract and which details are important for the questions we’re asking? As a discipline, genetics very much asks these kinds of questions, often in a precise, elegant, and methodical way. And the basis for ecology is often abstraction of these details in trying to discern the important features of a system for predicting its response to perturbations. Evolutionary biology, as a whole, uses principles of abstraction to classify the dynamic behavior of genomes in flux. But these disciplines generally focus on scientific questions; they’re geared towards modeling and measuring.

    By contrast, MIT BE focuses on two additional M’s: mining and manipulating. Ecology and evolutionary biology, historically, have ventured deeply into the measuring and modeling aspects of these tenets, but tend to focus less on mining and manipulating. In BE, we seek to design solutions to biological problems, and that approach requires significant investment in these other two components. I think this difference in mindset shapes the questions we ask and the outcomes we look for in our research.

    • Which details can we abstract and which details are important for the questions we’re asking? As a discipline, genetics very much asks these kinds of questions, often in a precise, elegant, and methodical way.

      Do you have examples of these genetics experiments? Could we turn them into low cost exercises that could be done in high school classrooms?

  3. I wonder, how did you pick up your intuition, Tony? I think intuition often comes from seeing it applied. I remember a specific instance where I asked another grad student in my lab about how she tabulated the variance for replicate samples which were themselves measured in duplicate or triplicate. (It’s turtles all the way down.) She said, “I think the variability contributed in the measurement is much less than the variability between replicate samples, so, I take the mean measurement of each sample, and use that to calculate the variance of all the samples.” For our preliminary experiments, this seemed reasonable to me, and has given me an intuition when I do other experiments – where is most of the variability going to be? In the water, in the cells, or in the different glassware I’m using? My colleague’s advice just helped me hone in on which variables I should work hardest to keep the same, given an imperfect world where things do change.

    Seeing the world as constantly vibrating might be intuitive if young students physically saw it. Looking under a microscope and seeing Brownian motion is what comes to mind for me – a concrete experience that I can extrapolate to the molecules in the graphite of my pencil, in the belly of a firefly. To me, those molecules are jumping.

    • Seeing the world as constantly vibrating might be intuitive if young students physically saw it. Looking under a microscope and seeing Brownian motion is what comes to mind for me – a concrete experience that I can extrapolate to the molecules in the graphite of my pencil, in the belly of a firefly. To me, those molecules are jumping.

      That’s a good question! One experiment might be to tether lambda phage genomes (~50kb) with beads on one end to the bottom of a glass flow cell. By manipulating the flow the students could pull out and extend the DNA (via watching the bead), or if no flow is applied, just watch the brownian motion of the bead. I think “feeling” the DNA, rather than seeing it as amorphous white goo in strawberries would be a very different experience for students – imagine doing that and then learning about DNA origami!

  4. Doesn’t the atomic theory of matter underlie how we teach biology? Elements into molecules; molecules into amino acids, nucleotides, steroids or carbohydrates; amino acids into proteins like topoisomerase? I think conformational flexibility and molecular vibrations are important, but not so critical that they should be taught before the central dogma or evolution or the hierarchy of protein structure.

    • I agree with you that it underlies how we teach biology – but in my experience the knowledge was never carried through or emphasized in any meaningful way. Yes, we should teach the central dogma, evolution, et cetera – this is biology not physics – I just think it would be much more effective if it was taught in the light of the dynamics and distributions.

      So much biology is taught teleologically – e.g. the topoisomerase untangles the DNA, or the kinesis carries the vesicles down the microtubule, et cetera. It often takes until late in college for it to sink in that in fact the reverse happens with non-zero probability – and often this is important to understanding or engineering the system!

  5. I think this gets at something that we discussed a lot in the MIT teaching course: the distinction between wanting to teach your students facts, and wanting to teach them concepts & skills. (a.k.a. defining appropriate & specific learning objectives for your class. “After this class, students should be able to…” what?) This was an especially sore point for those of us in the life sciences in the teaching course, in that students generally equate biology classes with the memorization of vast quantities of facts with complicated names (an approach that your Rutgers instructors appear to have agreed with). Alas!

    Of course, to be able to apply skills you’ll have to memorize some of the appropriate facts first: to design a genetics experiment, you need to know what a GFP reporter is and how a knockout works. It takes planning & deep thinking to be sure to construct a class such that students are constantly reminded that such facts are merely tools in an analytical toolset, and are tested on their ability to apply that toolset.

    Somewhat relatedly, it might be fun to do a post along the lines of “What I Actually Learned in ____ Class, Explained Using Only the 1000 Most Common English Words,” a la this XKCD strip: Re: your comment on our portrait of the chaotic molecular world – my takeaway from grad biochem would definitely be along the lines of “small things bang together really fast but whether something actually happens depends on what shape they are and how hot it is.”

Leave a Reply

Your email address will not be published. Required fields are marked *