What Should Biology Graduates Be Learning? A Cautionary Tale

Maybe…biology? Uncharacteristically, I’m going to, in vague terms, talk about my career trajectory because I think it offers an important lesson (maybe. Perhaps not. Fuck it, it’s my blog…) An anecdote in a post by John Hawks about the potential overemphasis on bioinformatics in the education of biology students inspired me to answer this question, because it mirrors my experience:

…don’t these students understand that in a few years all the software they wrote to handle these data will be obsolete? They certainly aren’t solving interesting problems in computer science, and in a short time, they won’t be able to solve interesting problems in biology.
I said, well, yeah. I’ve been through this once already — fifteen years ago, the hot thing was setting up a wet lab for sequencing — or worse, RFLP. That sure looked like a lot of data at the time, and a lot of students spent a lot of time figuring out how to do it. Some of them successfully started careers, got grants, and moved on with the times. Others fell by the wayside. Meanwhile, clusters of people at the DOE, Whitehead Institute, Wellcome Trust and several private companies were spending their time figuring out faster and faster ways of automating sequencing. Now one machine can do the work of ten thousand 1990’s graduate students.

Back in the day of S35 sequencing, I was in a microbiology lab that focused on molecular evolution of bacteria. Instead, I decided to focus of the phenotypic and evolutionary ecology aspects of the system. I even told my advisor, “Someday there will be machines that will do this for us, and much faster.”* To me, relating phenotype and evolution of bacteria in natural environments was much more interesting. Far fewer people were doing this, so providing some ecological and fitness data seemed to be pretty important. And even then, sequencing was becoming something lots of people were doing in a rather routine way.
Flash forward to today where I help develop bacterial genomics projects for a sequencing center (admittedly, there was a post-doctoral detour where I did a lot of sequencing). Guess what? We do have lots of machines to do sequencing on a genomic scale. But forcing myself to learn microbiology was a good thing. I’m now also involved in strain selection for some of the reference genomes of the Human Microbiome Project–had I become a ‘sequencer’, I’m sure I wouldn’t know what I need to know to do that.
I’m also leading (in name anyway) one of the data analysis groups for the Human Microbiome Project, which is essentially molecular ecology of the human body. Good thing I learned all of that evolutionary ecology stuff. Most sequencing centers don’t have someone trained as an ecologist (even if the ecology iswas rusty). I’m shocked how often I’m now re-reading all of my ecology books–I certainly didn’t expect to be doing this. In this case, knowledge of statistical techniques plays a role in doing this right (conclusion: we don’t have any good statistical techniques, but we’re working on it), but the key thing is familiarity with ecological theory and questions. Techniques will change, but many of the questions won’t.
I also saw students graduating just a few years after me who were being trained as ‘sequencers’, and they couldn’t get jobs. (Yes, basic experience with PCR, sequencing, and also Perl are good. Then again, you should also know some statistics too. That’s not what we’re talking about here). If you’re asking what you should be learning, I haven’t a clue: I’m the Mad Biologist, not Mr. Wizard. But I would definitely advise caution in adopting the techniques that everyone is using–that might be good for your advisor**, but it might not be good for you.
*Anyone who knows me will tell you that I would say this to my advisor. Oddly enough, I was known as a bit of rabble rouser in grad school….
**My advisor, a few years after I left, began to focus on the phenotypic aspects of the system we were working on and that I first investigated. Sometimes, I think advisers are too focused on using grad students to complete their research to their own detriment (mine, to said adviser’s credit, wasn’t).

This entry was posted in Education. Bookmark the permalink.

6 Responses to What Should Biology Graduates Be Learning? A Cautionary Tale

  1. DrA says:

    It’s always a mistake to focus on a technique or data-gathering approach rather than a question about basic biology. People dedicated to answering questions seem to “keep up” with changing techniques, but I’ve seen many technical specialists run up against obsolescence. It’s a evolutionary tale really, the generalist with long-term success or the specialist with short-term success. And these days the half-life on techiques is growing ever shorter.

  2. frog says:

    They should learn a hell of a lot more than strict biology — since so many of them had specialized “pre-med” training. Learning to use a specific software or technique is silly — learning how to develop new software or techniques is important. Be an inventor and not just a user, since it’s the development of new techniques that answers important questions — makes them tractable important questions — and not just chasing some problem.

  3. Jim Thomerson says:

    Back in the early ’90’s I got a call from my DNA collaborator, “Jim, I just got a new machine. This afternoon I replicated all of my previous three year’s work.”
    At the bachelor’s level, a student should have a broad background in biology. They should have had hands-on experience with a wide variety of observations, experiments, measurements, analyis, etc.

  4. Mokele says:

    Honestly, molecules are nice and all, but if there’s one course that’s needed, it’s paleontology. Biology without paleontology is like trying to understand politics without knowledge of anything before last Tuesday afternoon. Plus, we *really* need to improve the quality of macroevolution education, if only to counter ID.

  5. fvngvs says:

    Mike et al,
    I’ll weigh in on the ‘basics’ side, too. Observing my own field, engineering, shows that the same problems crop up forever – unwanted resonances, scaling problems, the tired old chains vs trees debate,…
    Learn the basic problems and you’ll be able to contribute your own solutions.

  6. I spent three years of my PhD making nested deletions and running radioactive sequencing gels (Sequenase! w00t!). Today that entire effort would cost a few hundred bucks and take one day.
    In the late 1990s/early 2000s I tried and tried and tried to convince biology graduate students in my post-doc department not to become computer jockey “genomicists”, because their long-term job prospects would suck total shit.
    I explained to them the history of neurophysiology: The first wave of neurophysiology faculty hired in the 1970s were mostly electronics whizzes who could build their own shit. By the mid 1980s, the equipment was standardized and mass-produced, and being an electronics whiz who could build a voltage-clamp amplifier from parts bought at Radio Shack became completely worthless except for the handful of people who work in the engineering departments of the two or three companies in the world that now mass produce this shit.
    Genomics computer jockeys are now in exactly this position, and none of them can get faculty positions in the biosciences because their skills are worthless and they aren’t real biologists. They can’t even get jobs teaching biology because they don’t know any biology.

Comments are closed.