Ruminating about blogging an area that doesn’t easily relate to the public.[1]

Last year I introduced the new science blogging network at Scientific American.

David Kroll wrote a post introducing this network, too, and noted the lack of chemists in their line-up.

Bora replied in the comments:

I did struggle about it. People with chemistry background whose blogs I like (and think they fit in my network vision) tend not to blog about chemistry much. Or are taken by other networks, or unwilling to join one. But majority of chemistry bloggers write for each other, very inside baseball I cannot understand, thus not really fitting my vision (or SciAm focus on broad audiences).

(Bora co-ordinates the Scientific American blogs.)

Janet Stemwedel also added some thoughts:

a further question to explore, beyond who’s blogging and who isn’t (and why that might be), is who’s blogging as a chemist versus who’s blogging about chemistry. (Undoubtedly, this would prompt further discussion about what exactly counts as ’blogging about chemistry’ – the whole current peer reviewed research versus life in the tribe of chemists versus musings on chemistry in everyday life line-drawing. Because we seem to like line-drawing for some reason.) [Her emphasis.]

This struck a chord with me.

I’m a computational biologist (or bioinformatician) but I currently rarely write on my own subject. I’d like to write more about it, but I feel it introduces several conflicts. In particular, sciblogs is supposed to be about outreach & education so I’d like at least some, or most, of what I write here to relate to a general readership, rather than just my computational biology peers. Ironically, wanting to write more computational biology, in turn, conflicts with wanting to use the blog as an opportunity to explore other areas of science and pass on what I find to readers! But let’s let this slide, as it’s an aside in the context of article.

Another aspect of this is that I prefer to start with the biology and fit the bioinformatics within that. It’s one reason I prefer the title ‘computational biologist’ to ‘bioinformatician’. There are people whose focus it more-or-less exclusively on the informatics aspects. Some aspects of this interests me too. I’ve been in the field over 20 years so I’ve seen these things evolve so it’s hard not to take at least some interest in them and practical things do have to be made, well, practical and practical things affect what you do.

(This starting from a biological vantage point is also true of my work. If someone approaches me wanting me to do x, y and z to their data, my usual approach to encourage them to back up a bit and tell me about the biology and what the biological aims they are hoping to address. Aside from being my preferred starting point, it helps ensure I’m addressing the problem they want tackled.)

Many times when I write about something with a bioinformatics component I have focused on the biological story, with references to any bioinformatics in passing, rather than about the bioinformatics as such.

I could write about the conceptual models of computational methods in a way that biologists might get some basic understanding of what they’re after. It would be hard to reach a general audience this way, but it would be a wider audience that writing to my peers.

Compounding that is that–I believe–many, but not all, biologists have a poor view of computational models. There is an undercurrent that computational models can’t ‘prove’ anything of use to some biologists. For example, I once explained some biophysical work I wanted to do only to be asked what I would suggest for an experimental ‘proof’ – unwittingly the person asking the question had revealed he didn’t consider that a (or the) computational model could stand on it’s own. It’s a bias I’ve seen before, that only a ‘wet’ experiment can serve as a proof.[2]

Physics fares slightly better in the general public arena than computational biology does. Mathematics, however, struggles.

One question this raises is does presentation of science to the public tend reflect things general readers will already have some interest in or connection with and in that sense simply re-enforces this and with that present a skewed view of science.

By skewed a view of science, I mean that general readers see little or nothing about entire areas of science, that great swathes of science is simply out of sight.

As you can see my general thoughts are that this issue is not limited to chemistry, as David related, but has to do with the ease that some areas of science can be related to the public compared to others.

Bioinformatics, my field (although I would properly call myself a computational biologist), has similar issues. It clearly is important – bioinformatics / computational biology underpins a lot, if not most, modern genetics and molecular biology. Despite that most people writing about computational biology (or bioinformatics) are talking to their peers, some to biologists, but very few to the general reader.[3]


1. Yeah, I know, this is meta-blogging and meta-blogging is yeech and all that. It’s also another old post from the drafts heap that I’ve revisited. It might seem I’m lacking originality, but it’s that while trimming out some drafts* I can’t see myself ever returning to this post resonated with thoughts that have been on my mind for the past month or two; see Footnote 3.

2. It’s a real pity as I value working alongside biologists, each contributing to the final result from different strengths and, of course, I am a biologist myself; the whole idea is to bring biophysical insights to biological systems. I just feel too often there isn’t a true peer-level partnership with the upshot that opportunities are missed to let computational biology or biophysics bring what it can to a project. Or statistical aspects if biostatistics is your thing. Some areas of biology are becoming increasingly dependent on biophysics (or statistics) more than I think some fully appreciate or perhaps want to admit to – one in mind is genomics and gene regulation as the three-dimensional arrangements of genes and genomes will come to be of increasing importance as biophysical aspects are an integral part of these analyses.

3. Part of the reason behind this rumination is that, long story short, I still feel conflicted about the mix of material and how to best present it. Part of the issue is that we can’t use categories in our blogs as they’re already taken by the group platform, so how best to indicate different sub-topics, areas or, in particular, audiences. Not for the first time I’m mulling over a few ways to address this as I’d like to try bring a few series or thematic topics to the blog without confusing readers. I can imagine, for example, three different articles about some aspects of the three-dimensional arrangement of genomes: one to a general audience, one to biologist and yet another to computational biologists – and that’s before I start thinking of the very wide range of topics I cover!

Other articles at Code for life:

The inheritance of face recognition (should you blame your parents if you can’t recognise faces?)

Temperature-induced hearing loss

Online lecture series on genomics and bioinformatics

How did you learn to critique the scientific literature?

Coiling bacterial DNA

Career paths, redux — the academic research career is the exception