Random conference thoughts

When my mind wanders during a conference talk, I often find that short sentences summarising how I feel about work come into my head. Here’s what I scribbled in my notepad during the ComBio meeting this week:

Information relevant to me is communal, not owned by individuals
I wrote that down when thinking about how biologists interact (or not) at meetings. As a computational biologist, most of my day-to-day problems are programming and software issues. If I need information, I go straight to the Web. However, wet-lab biologists seem to get much more of their information by talking to other biologists. If you’re interested in an organism, a model system, a laboratory technique or if you just want to get your hands on a plasmid, you talk to someone who works on it. It strikes me that a lot of wet-lab group leaders claim some sort of ownership over the information that their lab generates, resulting in the “so-and-so is the world expert in system X, you should talk to him” mentality. On the other hand, the idea of schmoozing with “the Perl expert” is a tad silly.
That, at least, is my excuse for not networking much at biological conferences ;)

Bioinformaticians need to be free
We (or at least, I) are happiest when working on a range of problems. A main project and a bunch of fun, side projects with plenty of variety is the key to a happy bioinformatician. Conversely, getting bogged down for months or years on a single project, particularly one on which you work largely alone with little external input makes for a sad bioinformatician.
Much has been written about Google’s 20% time, where employees are encouraged to spend 20% of their time on projects that they think are fun, cool and interesting. I think this would be a great policy to implement for bioinformaticians, computational biologists and other researchers in academia.

4 thoughts on “Random conference thoughts

  1. That first thought is a huge difference between bench work and bioinformatics. When we have some questions we can just search for the answers online. In bench work you can search for the protocols online too but the manual how-to and the little tricks that make it work with your working conditions are really crucial and are not easy to find. That is why JoVE (http://www.myjove.com/) or something similar can really make a difference.

  2. I think part of it is that it is really hard to replicate and transfer difficult techniques. So while it is (relatively) easy to pass around, or replicate, say a perl script, a wet lab bench protocol, no matter how careful the person who wrote it down, could still be missing crucial details (this experiment doesn’t work in Bath because the water is too hard – and no I’m not kidding, this happens). So ‘so-and-so is the world expert in system X’ really translates to ‘if you want to do this, go and learn in their lab’ then take it home and iron out all the extra issues that appear. The question of openness revolves around how willing they are to let you do this.

    Which is not to say that the availability and usefulness of good protocols couldn’t do with a lot of improvement.

  3. A very good post, thanks. This is exactly what I’m experiencing. Whatever the reason for the 100%-attitude and ingroup vs outgroup behaviour it makes working for biologists difficult as a computer scientist. They, in turn, don’t know the “working” culture in computer science and think we aren’t really working but rather playing. Which might contain quite a bit of truth.

    Cameron is right but if biologists were like computer scientists then you would find tons of very detailed descriptions on the web and a newgroups for every strange technique with a core group of 5-10 experts that are happy to answer questions. Even the water at Bath can be analysed and once it’s part of an FAQ it’s not a mystery anymore. But this kind of reasoning seems to be difficult in biology or maybe there are simply more possible conditions to try than in computer science? No clue.

  4. I think the environment is less reliable in biology than in computer science. If you re-run a compile you will (usually) get the same thing – more to the point you’d be annoyed if you didn’t because ‘something is wrong’. Something not working is a pretty common occurance in your average wet lab so all procedures are a bit more contingent and a bit more local. But you’re point about culture is dead on. These things create a reinforcement loop where the end result is ‘Why bother putting up protocols seeing as they probably won’t work for other people’ or, worse ‘After all that work why should I give it away’.

    I would say working with computer scientists has its own challenges. People do look at me strangely when I try to use the terminology. To be fair this is more what happens in person than on the web which is much more friendly and open. Or is that just because I can’t see your expression? :)

Comments are closed.