Bioinformatics Zen

Bioinformatics - a wide set of transferable skills

// Wed July 27 2011

Working in academia feels like a lifestyle rather than simply a job. I feel that being a researcher, in particular a bioinformatician, is part of my identity. I carry my research around in my head outside of working hours but at the same time I enjoy the autonomy and mastery of my work. Therefore if you become unhappy in academia, as thoughtfully described by Massimo Sandal, the decision to leave can be very difficult. The wrench of leaving a career that required the best of your twenties for post-graduate study will be painful. I think, however, once this decision has been made bioinformatics skills are eminently transferable to begin a new career.

Knowing how to program is often the first class a computational biologist is taught when coming from a biological background. Programming is a skill which you realise the worth once you have it. There is a great feeling of satisfaction when creating a piece of software with nothing but a text editor and an open-source programming language. Ruby programmers are very much in demand, and if you already know one language it's easy to learn a second.

A good way to improve as a programmer is to create and develop an open source project. Something to tinker with, try new software libraries or a different programming language. Furthermore an open source project shows that you can program and hopefully program well. Interesting companies may also tend to hire people based on such projects. Auxiliary skills such as test driven development, a version control system like git, or mastery of a customisable editor like vim can make you more productive as a developer. Working to add new skills that augment yourself as a bioinformatician are also the same skills that many companies looking for in programmers.

Computational biologists are more than simply a programmer who works in a a biology department though. Developing in silico hypotheses requires storing, manipulating and analysing biological data. Recent years have seen the amount of this data increase and increase. Therefore data-munging skills, the bread and butter of a bioinformatician, will be attractive to companies. Particularly those dealing with "Big Data."

Deriving meaning from meaning from data is therefore a valuable skill for a bioinformatician. At a basic level bioinformatics requires classic statistics such as ANOVA and regression to prove or disprove a hypothesis and a solid grounding is therefore useful. Simple statistics are also the easiest to understand and implement. A good understanding of statistics always impresses me when I discuss a topic with someone.

The intersection of programming and statistics is what what excites me the the most in bioinformatics. When a good understanding of statistics is combined with solid programming ability great things can happen. There are complex probabilistic algorithms used regularly: generating a multiple sequence alignment, a phylogenetic tree, or an assembled genome sequence. These algorithms search for the best solution in a countably infinite number of possible solutions to describe a phylogenetic tree or genome assembly. Having maximum likelihood estimators or Bayesian MCMC in your tool box when you want to search a solution space or large data set is very useful. Especially when you develop an open source tool solving a difficult problem.


You hope when Embarking on career after university that it will last at least the next few years. Hopefully even a lifetime. Unfortunately at some point you may find that have to apply your skills to a different career.

Wet-lab biologists may spend years developing specialised skills rarely applicable outside the lab. Bioinformatics on the other hand I think develops an excellent set of transferable skills. In summary I think bioinformatics is an excellent career because it's an exciting growing field but also because the skills learnt are useful in other professions if you decide your career lies elsewhere.