I am in Edinburgh to attend a UK neuroinformatics workshop. Perhaps one could argue that neuroscience is one of the few biological disciplines where theoretical biology can play significant roles. Here I intentionally avoided the word “theoretical biologists” because they are not necessarily the people who come up with the great theories in neuroscience. One famous example is Hodgkin-Huxley equation which are developed by experimental biologists who also knew enough mathematics and computing to come up with the mathematical model. Today significant number of experimental neuroscientists have done some modelling or interested in doing it.

Therefore, it should be reasonably straightforward to organise collaborative activities between experimental and theoretical neuroscientists and set up databases and computational tools for sharing and disseminating experimental data. Didn’t look like that, I am afraid, from what I have heard from this workshop. In bioinformatics there have been the data, such as genome sequence, protein structure, or microarray data which nearly everybody thinks are important and clearly require informatics for efficient analysis. I am not sure if we can agree on that for neuroscientific data. In molecular biology, there is the central dogma and since it was proposed by Crick most of the molecular biological research haｓ been focused around it. In neuroscience, on the other hand, it is very difficult to come up with a similar dogma because cognition cannot be described as a linear process. As a result neuroinformatics tends to be very diverse and diffused without a clear focus and if it continues, little will come out from large collaborative research which needs a well-defined direction.

I am not really qualified to suggest what should be the central topic of neroinformatics, but at least I can suggest something simpler. If you search the word “neuroinformatics” by Google, you will find a lot of databases, software and other research activities related to it. Funnily enough though, there seem to be little connections between them despite the popular mantra that neuroinformatics is about connecting data and people with different backgrounds together. Perhaps it would be useful if we can have a unified portal site from which neuroscientists can find data they need as well as right research groups who share common or complementary interest. Maybe I should set up the site. Is there anyone who wants to collaborate with me?