Inter-disciplinary work vs transitioning fields
Suppose you’re now under a degree program that does not necessarily satisfy your intellectual curiosity, how far can self-studying take you? Online classes are ways you can receive education from the best teachers in the world without attending their institution. But this comes with a cost. It takes tremendous motivation, effective planning and time management to reap the full benefit of these online classes. It’s too easy to take too many classes, going through all the videos without retaining much information afterwards.
I’ve tried telling myself that I could settle in this degree by saying I’d just take all these online classes alongside my major classes. Online classes are not for everyone and you could just be fooling yourself into thinking you can be proficient in that subject matter by ‘completing’ the said online course.
In the context of Bioinformatics, you need knowledge in biology, computer science and statistics. The weighting of each of these discipline depends on the projects you’re undertaking and the role you play in those projects. Some people made it with a informatics main background but little knowledge in biology, this is entirely possible. But here I wish to quote from Daniel Dvorkin
Bioinformatics rests equally on three legs: biology, computer science, and statistics. It’s pretty much impossible to become a true expert in all three. But if any of them is too weak, the whole table falls down.
Inter-disciplinary work is not the most popular thing because it takes a lot of time to build up the basics and foundation to finally be able to work on the questions you want to answer. A lot of people don’t have the patient to do this. I have been in the position where I tried to produce without a solid understanding in the basics. But things become disastrous at a later stage.
If you’re like me, who’s not lucky enough to get in a program that’s tailored to your interest, what can you do to eventually land the position/project you crave?
What you need to know as a self-taught bioinformatician
I recommend anyone who’s looking to teach themselves bioinformatics to have a look at this paper:
Searls DB (2012) An Online Bioinformatics Curriculum. PLoS Comput Biol 8(9): e1002632.
If you’re coming from a pure biological background, most likely you’re lacking some of the follow subjects to work your way towards a competent bioinformatician.
- Algorithm analysis or automation theory
- Calculus (two or three semesters)
- Computer organization
- Discrete math
- Linear algebra
- Object-oriented programming like Java and C++ (two semesters)
- Operating systems
- Probability and statistics
While you can just grab a book with that particular title, not all textbooks are written to be read in a self-taught manner. If you’re in a full time program like me, you probably don’t have the luxury to take a formal class in your own university. I have failed in trying to teach myself everything, and I don’t suggest you to do so.
How should you try to fill in those knowledge gaps?
If you only take away one thing from my post, I would say DO NOT TRY TO RUSH THROUGH THINGS. As much as your supervisors or people around you want you to pick up the basics ASAP, that’s simply not the way you deal with inter-disciplinary work. Knowing things from multiple fields doesn’t give you the excuse to skimp through a particular field, you can’t ignore the foundations you should have. Remember this, you’re a student here, you’re not expected to know everything on day 0. I think this is something I’ve only understood when I was almost a year into doing full time research work. I don’t know what kind of labs you are in, so I can’t give you a generalized recommendation, perhaps you’re luck enough to be a lab where your supervisor likes to take things slowly.
Always allot productive time when you would sit down to focus on reading/solidifying your bioinformatics skills every week. It’s all too easy to just fall into the habit of simply writing scripts, cleaning data and generating data for your project without working on your skills. You must not think that you’re already good enough and thus doesn’t need any more training. I think a long haul project tend to sway students from their skill set development.
Here’s what you could do:
- Choose a couple of core subjects that you feel like you most desperately need, for me that’s probability, computer organization and calculus.
- Plan a weekly time table for those courses, do it in the