Biocurious is a weblog about biology, quantified.

Ten simple rules for selecting a [lab]

by PhilipJ on 3 December 2006

PLoS Computational Biology seems like lists as much as I do. This time around in Ten Simple Rules for Selecting a Postdoctoral Position they’re offering advice to those soon to be finishing their PhDs. In fact, the majority of the list is good reading for those who are about to start a graduate degree as well, so click here to read the whole list, no subscription required.

Since I’m in the proccess of figuring out labs in which I’d like to apply for my own PhD, I’ve spent a lot of time thinking about the combinations of rules 5 (Choose a Project with Tangible Outcomes That Match Your Career Goals) and 10 (Learn to Recognize Opportunities). Taking on exciting new projects by definition means you are charting in unknown waters, and tangible outcomes are no longer guaranteed (and even for a lot of me-too science). The editorial notes,

[f]or a future in academia, the most tangible outcomes are publications, followed by more publications.

How does one successfully ballance tangible outcomes (publishing papers [that people will read]) with choosing a lab doing novel and interesting science, and what are some safeguards from overly-ambitious “opportunities” that take too long to produce those tangible results? I’ve seen a number of really excellent talks (most recently at the Frontiers in Biophysics retreat), only to realise by the end that the number of dead ends graduate students (over a few generations) followed on these projects numbered much larger than the successes.

I’ve started learning the lessons, but I don’t yet have the answers.

  1. TheBrummell    4244 days ago    #

    test comment – am I blocked?

  2. TheBrummell    4244 days ago    #

    OK, that last comment seems to have gone through. I was blocked from commenting here a few weeks ago.

    First, I’d like to say that I really like this blog, and I read it fairly often. Regardless of where you end up for you PhD, please continue to contribute.

    I’m also at SFU, and I just finished the process of selecting a lab for my PhD – I had a choice between the University of Guelph and the University of Zurich. Surprisingly, I ended up deciding to go to Guelph. I discuss the details (many details, too many details) on my blog, but the short story is I based my decision on the project. I still wonder about the alternate choice, and probably will for many years, but I can honestly say that I feel I made the best choice. I don’t stay up at night, worrying.

    So my (unasked-for) advice is to try to base your decision primarily on what you’ll actually be doing day-in, day-out. A place might be very attractive for whatever reason, but you’ll probably spend the overwhelming majority of your time sitting at a desk, staring at a computer screen, regardless of what’s happening outside.

    I was faced with two choices, and I resolved this partly by imagining the two different projects taking place in opposite locations – and I decided I’d rather do the Guelph project, regardless of whether I would be doing it in Guelph or in Zurich.

    Your dilemma of trade-offs between “exciting” and “likelyhood of publication” fits into the general discussion of “sexy” science vs. “reliable-but-boring”, and discussions of “exploratory” science vs. “hypothesis testing”. This is a much larger discussion than can be summarized in a comment, but I’m happy to see your thoughts on these issues. I’m looking forward to learning how you will make your choice, and what your opportunities will be.

  3. luke Quinton    4243 days ago    #

    I suspect the “exciting” here leans more on the side of “can I wake up today and not utterly dread the thought of going to work?”

    The “switching” of the two locations (Guelph & Zurich) seems like a very helpful idea.

  4. PhilipJ    4243 days ago    #

    TheBrummell — died a little while ago, which used to take care of our comment spam. By died, I mean it started thinking everyone was a spammer, so we got rid of it. We seem to fly under the spammer radar here at Biocurious quite well, so it wasn’t necessary anyway. If it ever happens again though, please let me know.

    I really like you’re method of swapping projects between the institutes to decide which project is more interesting — that’s definitely a good way to try and differentiate on the science you’d like to do. My dillemma, though, is how can one actively differentiate between a high impact, exciting project that has a decent likelihood of working from those that don’t? How do you analyze the projects to decide which are more or less inclined to succeed?

    In all cases there is a large amount of chance involved, but I would love to hear from you and others how you’ve tried to gauge the likelihood of success of a project.

    Luke – no, in this case I actually mean high impact, in a scientific community kind of way. Lots of things are interesting enough to make me not dread going to work as long as there is some kind of progression, but in thise case, when you’re dealing with a PhD project, you may as well go all out and try to find something that the community at large will find exciting too.

    Once things become a little more concrete (there are no offers anywhere yet, application deadlines aren’t for a couple of months), I’ll expound a little on what kinds of things I’m looking for in a lab. A post on exploratory vs hypothesis testing science is definitely in the cards now, however, and I look forward to more discussion soon!

  5. Andre    4243 days ago    #

    I think it’s hard to tell if a particular project is likely to succeed, and for that reason I don’t think you should try too hard. Especially because your project has a decent chance of changing during your PhD (mine certainly has… several times). I think the thing you should watch for is the success of former students from the group. Looking at the groups around me this seems to be as much a function of the group as the individual. If the group’s students consistently publish good work during their PhDs, you probably will too.

  6. Rosie Redfield    4242 days ago    #

    Base your choice on the quality of your colleagues.

    You shouldn’t be doing a PhD to get good publications, or even to work on exciting problems, but to learn to be a good scientist. This is much easier if the people around you (students, postdocs and faculty) have high abilities and high standards.

  7. Andre    4242 days ago    #

    I don’t think there’s a single factor that you should base your choice on. I would definitely agree that colleagues are important. I know people that have gone into labs that publish in high profile journals despite having heard horrible things about the advisor/environment thinking “I’m going to be different, he/she will respect me” (or something like that) only to find that everybody’s advice about the lab environment was right and that it was a bad choice.

    Learning to be a good scientist might be the primary goal, but working on a problem that’s not exciting and then having little to show for it after 6 years is not a good outcome. That does happen, but I think it’s important to make sure that it doesn’t happen to students in the lab you’re considering. After all, it’s hard to become a good scientist in the long term if your career ends after graduate school.

    I should also point out that I didn’t mean for my previous comment to sound like general advice. I know some of the groups Phil’s considering and I think they would both be good choices in most terms and I was really just responding to the question of project choice. If I wrote it again I would say something like “ of the things you should watch for…”

  8. TheBrummell    4240 days ago    #

    I’m flattered people like my project-location-switching idea; it was approximately the last thing I thought of before I sent the flammable-bridge email (“Thank you for your offer, however I have decided to pursue my studies with…”).

    I find it worked quite well when I got to that last stage, the final moment before the bifurcation.

    PhilipJ said: Once things become a little more concrete (there are no offers anywhere yet, application deadlines aren’t for a couple of months), I’ll expound a little on what kinds of things I’m looking for in a lab.

    I found that the process of looking for potential labs was highly influential in determining what I wanted to do – there are so many possibilities that it’s unlikely you’ll simply say “I absolutely want to do THAT” and then pursue that option single-mindedly. Personally, I was simply browsing through “position available” ads, and one jumped out at me – Guelph. Zurich was similarly serendipitous – a former post-doc of my M.Sc. advisor came for a visit, and told me about a new assistant prof. at U of Zurich who had seen my talk at an previous conference.

    Luke said: “can I wake up today and not utterly dread the thought of going to work?�

    That’s a good strategy for an entire career, as well as one particular portion of it. I try to base my decisions on a principle like that as much as I can. My evaluation of potential Projects was heavily based on hypothetical answers to that question.

    PhilipJ said: My dillemma, though, is how can one actively differentiate between a high impact, exciting project that has a decent likelihood of working from those that don’t? How do you analyze the projects to decide which are more or less inclined to succeed?

    This probably comes down to the very specific, so I don’t know how much help my experiences will be. But I’m supposed to be marking lab reports right now, and this is more interesting than broken English.

    The Project I would have gone to do in Zurich would have initially involved (talking in the future very imperfect is really annoying) three sub-projects. One was essentially a technical exercise, guaranteed publication, and the production of a useful tool for other researchers (so, exciting from a certain point of view), that was very similar to the work I did for my Master’s. Ho-hum, but I know for sure I’ll get a (first-author) paper. The second was helping another PhD student there with a BAC library (lots of boring molecular work). Again, guaranteed publication, and the other PhD student is (in my opinion) both an excellent person and a scientist destined for great things – but the sub-project would not be particularly interesting to me. The third part would use the first two as a foundation for some really interesting (to me) research that would likely have carried on into (potentially) a post-doc position at another lab. But PhD projects mutate rapidly, best-laid-plans-of-mice et cetera. Also, I seem to be abusing my parentheses keys.

    I think it’s very difficult to predict “exciting” research, since part of what makes something exciting is that it’s new and unexpected. Probability of success, as least as measured by publication, might be a little easier to predict, but still looks very uncertain to me.

    Rosie said: Base your choice on the quality of your colleagues.

    Rosie Redfield is awesome, and she has excellent taste in beverages. I suggest you listen very carefully to everything she says. Personally, I found it impossible to rank my two potential choices relative to each other based on “the quality of [my] colleagues”. I find it very hard to weight various factors in a meaningful way – does a potential advisor’s own PhD advisor matter as much as their last three publications, or who their current collaborators are? I didn’t bother making a decision matrix for factors like these – there was simply too much uncertainty.

    Please keep us posted – I’m finding this conversation very interesting. Good luck!

    P.S. I can’t seem to make italics or bold work. Textile help didn’t. Any suggestions?

  9. PhilipJ    4239 days ago    #

    That textile help page really isn’t very helpful. I’ll look into that. The way to make things bold is to use paired asterisks, and to italicise something, you can use paired underscores around the word. Works for me!

    My own search was carried out by surfing around the various departmental sites, looking for labs that do research that I admire. It varies widely from lab to lab if they actively post available positions, so I simply contacted the PIs directly through email. We’ll see how my applications pan out!

  10. TheBrummell    4239 days ago    #

    Let’s try this out…

    PhilipJ said: ...looking for labs that do research that I admire.

    That’s how I found my Master’s position, and it’s a good plan – most PhD positions are probably not advertised, and any lab you contact is probably one you’re already at least somewhat familiar with.

    When I was looking for my M.Sc. position, I got many responses like “I’m glad you’re interested, and you seem suitable for our lab, but do you have any funding?”. Since I didn’t have an NSERC (and don’t have one now), it was difficult to find a lab. Many profs are limited by funding constraints in taking on new students; if you have your own funding (partial or full), you stand a much better chance of facing a dilemma like I did, of having to choose between two exciting projects. You probably already knew that, though.

    When I visited labs looking for my M.Sc. position, and when I visited Guelph and Zurich recently looking for a PhD position, I was told I was doing exactly what I should be doing – going and meeting profs and their labs, and figuring out what I’d like to spend the next 2-4 years doing. Email is the obvious place to start – will you be able to visit any of the labs you’re contacting? Some profs are willing to pay some or all of your expenses for a visit – I got 1/2 my (awesome) trip to Zurich paid for by the (very generous) prof there I was visiting.

  Textile help