How do I solve my problem like a bioinformatician?

Short answer to the title’s question:  Google it.

Longer answer: Consult the internet using your preferred search engine, limiting your search query to the barest minimum of key terms and phrases that describe what you are trying to do, refining your search as required.

Longest answer: Running into difficulties with bioinformatics can be frustrating and demoralising (as in computing, and life more generally). It’s easy to be left feeling that your problems are unsolvable, that bioinformatics is an impenetrably complicated field, and that everything would be easier if you had more knowledge and training, like a proper bioinformatician.

There’s a common misconception that bioinformaticians are good at what we do because we know everything that we need to; that we are familiar with all the latest analysis methods and how to implement them.

Reality: no one knows everything

It’s certainly true that we will be knowledgeable about whatever fields, methods and tools we have personal experience with. But the field of bioinformatics is wide and ever-expanding, and the reality is that very few experts will have enough knowledge and experience that they never run into difficulties when running analyses.

So how do we solve coding and other analysis problems when we (regularly) run into them? As I mentioned above, we can’t know everything but the global science community is numerous, and the internet is a big place.

It’s safe for us to assume that we will not have been the first to encounter whatever problem or error is causing us stress. Someone, somewhere, some time ago will have struggled and cried out in frustration to the internet, and the global bioinformatics community is generally helpful and generous enough to have answered.

The challenge for us then, sitting in our busy office (or alone in the dark at home with our third coffee at 23:30), is finding that answer.

The answer is almost certainly out there somewhere. Finding it quickly and efficiently is a skill in and of itself. Like any skill it can be improved with practice.

Understand what you are trying to do

The most important step to finding your answer is first taking time to understand exactly what it is you are trying to do. What function are you trying to carry out? What software, package, or language are you using? In what context did you encounter this problem?

Knowing how much information to contain in your query when you turn to your search engine is an instinct you will gradually develop. If you think your problem is directly related to the research context of your work, then you should include it. But if you have encountered an error when attempting to merge some tabular data whilst using a specific R package (for example), then excluding the package information will allow your query to return results from a much wider base of users.

The more information your query contains, the more you restrict the search results. In this vein, it is usually advisable to leave out any extra words that make your search query a grammatically-correct question. Simply type a string of key terms.

Of course, following this advice is difficult if you don’t know the precise technical vocabulary to describe what you are doing. So again, simply ask! Ask your colleagues. Ask the internet. Ask your InFLAMES Bioinformatics Support team!

Nicholas Booth, bioinformatician