About a year ago, I was having a severe bout of career depression. It seemed to me that people in my situation – biologists turned bioinformaticians in traditional academic environments – faced a special set of problems when it came to career prospects. I wrote down a few thoughts that were intended for the blog, then put them aside.
Recently I rediscovered the notes and as my situation has improved somewhat, I thought I’d write them up with a more positive gloss. All non-tenured young biological researchers, computational or otherwise, worry about their career path at some stage, but my suggestion is that there are problems specific to “people like us”. Read on and see if you agree.
The traditional career path for a researcher in academia goes something like this:
- Do an Honours degree.
- Do a Ph.D. in an area that interested you during the Honours degree.
- Do a postdoc in the lab of someone that you met during the Ph.D.
- Do n postdocs where n is an arbitrary number, most probably in the same field.
- Publish enough papers that you can apply for a permanent position.
- Start your own group based on your interests and experience.
- Take on an Honours student in the hope that they will stay on for a Ph.D.
- and so the cycle continues…
This is what I call “the old school science career path”. It’s based on some kind of master-pupil principle, passing the torch from generation to generation. I guess it was established when universities were established and it’s still going strong today.
Leaving aside the fact that lucky young people today can do Honours and Ph.D. degrees in bioinformatics, the career path for the rest of us goes more like this:
- Steps 1-3 as per the traditional path.
- Discover bioinformatics and all that it entails: programming, web-based collaboration, the “open” philosophy.
- Realise that this affords the opportunity for genuine, multidisciplinary science without the restrictions of academic disciplines.
- Struggle to become established within a system that isn’t ready for this radical concept.
In no particular order, these are some of the problems and attitudes that I’ve had to contend with as a biologist-turned-bioinformatician surrounded by biologists-not-turned-bioinformaticians.
1. So what is it that you do exactly?
Just as there are non-scientists who wonder just what it is that scientists do all day, so there are non-computational biologists who wonder the same of their computational colleagues. I’ve encountered a substantial fraction of people who seem to believe genuinely:
- that if it’s in a computer, it isn’t “real”
- that if it’s not in a wet lab, it’s not biology
- that if you didn’t generate the data yourself, it’s less worthy
- that computing is not an intellectual or creative activity; you just feed numbers into the black box and numbers come out
- and that have no concept of programming a computer whatsoever; i.e. understanding that you can make it do what you want, as opposed to being a passive user
If there are too many of these people in your department, you need to be somewhere else.
2. Defining your interests
Many scientists seem to have latched onto a rather specific topic early in their career (at Ph.D. or perhaps even Honours level) and stayed with it for the rest of their lives. If you ask them what they’re interested in they’ll have a snappy one-line answer such as “the structure of myoglobin”, “the visual system of bees”, “ways to improve brewing efficiency” and so on.
Biologists-turned-bioinformaticians have a rather different way of looking at things. When asked what my interests are, I often respond “any interesting biological problem to which computational techniques can be applied”. It’s the most honest answer that I can come up with – I really don’t care whether the biological system is cold-adapted microbes, cancer, yeast, fruit flies or humans. In fact to my mind, the main strength of a computational approach is that it provides you with a generic toolkit to address whatever you like.
The problem is that other people, especially in a job interview situation, will judge this response as rather vague and unfocused, as though you haven’t really thought about just what it is that’s important, interesting or worthy. It also makes you think about the problem of what kind of group you could possibly run, should you ever get to that position. I wonder if one solution might be to feign an interest in something more specific (“yes, I’m totally into pathogens!”), with the goal of getting your foot in the door and then doing what you want later.
3. Falling into a technical support role
Do any of these sound like you?
- people are always asking if you can just “write a quick script” to do some data munging
- you’re happy to oblige because otherwise, they’ll take 3 months to achieve what should take 30 minutes
- on the other hand, you can’t escape the feeling that people are taking advantage of your good nature and that if only a few more people would make put some effort into improving their computer literacy and learning some basic scripting skills, it would benefit both them and you
- you’ve toyed with organising a short course to teach these skills but (i) you don’t think there’s sufficient general interest and (ii) you’re not employed in an official teaching capacity – it’s just another distraction from your own research and career development
If so, you’re a “technical support” computational biologist. That can be great if that’s your role – if your job title is something like “bioinformatics engineer” in a company or large institute. But in an academic setting where you’re also expected to perform novel research of your own – you’re in trouble.
There are far too many academic environments like this – where researchers are treated as resources for furthering the careers of others, rather than appreciated for the technical skills that they bring. No wonder that the ideal environment for many young bioinformaticians is not a biology department, where they are most needed, but an institute that specialises in bioinformatics – where they feel more appreciated.
4. Learning on the job
I began my time in science as a biochemist and later, acquired some expertise in molecular biology. This involved learning quite a large number of skills – cloning DNA, culturing microorganisms, purifying proteins, performing enzyme assays, spectroscopy etc. and learning how to operate various apparatus and machinery. However, once those skills were learned, there was very little need to cultivate them or to learn new ones. Regardless of the biological system or organism under study, I could have spent my life applying essentially the same skill set and generated numerous papers of the type “gene X encodes protein Y involved with process Z”.
In contrast, as a bioinformatician I’m required to learn something new almost every day. This could be as trivial as some facet of the Linux operating system that I’ve not used before or as complex as the theory behind a statistical method with which I’m not familiar. Personally I find this very satisfying, as someone who has always maintained the philosophy that science is a process of lifelong self-education.
On the other hand, this approach is not without its problems. Would my life be easier if I knew exactly what I was doing all the time, because I only ever used a limited range of techniques? Would I be more productive if I had formal training in computational or statistical methods, rather than having to spend a couple of weeks reading up on e.g. machine learning before I can use it? Would the boss be happier if I gave the impression that I knew exactly what I was doing 100% of the time? Maybe so. But is that really research? Remember, science is hard.
How then, are we biologists-turned-bioinformaticians in academia to remain positive and attain our goals (namely a permanent job)? These are my career tips:
- Find the right working environment. You need a critical mass of like-minded people to support you. If you feel that you’re surrounded by closed-minded people who just don’t “get it”, get out. Don’t hang in there in the hope of converting them to the cause or because you fear that quitting will damage your future prospects. You’ll end up insane. Just run away.
- Work on important, biologically-relevant problems. Many biologists just can’t see past the organism/system, regardless of whether your computational methods have wider significance. They need to be shown that what you do is useful.
- Publish. Publications are the sole criterion that others will use to judge your success. Not blogs, not websites, not participation in online communities. Think always in terms of your next paper.
- Find ways to be proactive and maintain your enthusiasm. One of the best ways is to supervise students – you may not be able to influence your peers or superiors, but a student is looking to you for guidance. Be active in your group or department. When you notice people having problems that could easily be solved with a simple perl script, make everyone aware of it in your next presentation.
- Learn from successful researchers. If you know someone who has the bioinformatics job of your dreams, get friendly and ask them how they got there.
- Have faith in your ability and convictions. Did I say that science is hard?
- There’s no substitute for hard work. Put in the hours. Keep expanding your skill base. Take advantage of any career development seminars or courses on offer.
- Be aware of alternatives. After years of postdocs, it’s easy to forget that there are places where bioinformaticians can work other than universities. Find them and try to picture yourself there. See how that makes you feel.
Remember, science is supposed to be fun. If it isn’t, something’s wrong. So fix it!