This morning I sat down and read through a very interesting paper on how to split up networks into modules.

I know, I know.  Pretty boring stuff, you might think, but I will beg to differ and it’s my blog, so I’ll write what I want on it

Well, this paper was seeking to split networks into modules.  You don’t kneed to know how many modules you have in the network, or indeed, you don’t kneed to know if there are any modules - the entire network might be one single coherent module and cannot be split up.

A network can be anything - the internet, a road map, a set of friendships, a collection of blogs.  the only thing they really need is a series of links between the objects.  So, in the case of blogs, you might, for instance, put a blogroll on the front page of your blog and on this blogroll, you might put a series of links to other blogs, written by like-minded people or that you think are interesting.

Right off the top of my head, I suspect that if we took the network of all the blogs in the world that have links on their front page to other blogs, we would easily see that they split off into modules based on the language in which the blog is written.  Other than that, you could find sub-modules based on discipline, political perspective, sexual orientation.

And there might be others and that would be very interesting.

I’m sure somebody has analysed the data.

Curiously, my Amigo Pere has a link to my blog from his blog and his blog is written in Catalan (Hi Pere). So, as you can see, networks can be very complicated thingies and understanding them is really an interesting thing to do.

There is a famous network that is continuously analysed and it is based on a ‘friendship’ network within a karate club.  This was published in the 1970s in the anthropology literature.  The interesting thing was that the network could be easily split up into two modules according to who was friends with who and very soon after publication of this information, the karate club split up over an internal dispute and it split along the lines of the two modules.

The paper I read this morning was a new method of splitting networks into modules and I think that it has opened up a whole new way of looking at biology for me.  I’m going to have a think about it for a while, but definitely, it was interesting.  I have thought about multiple sequence alignments as though they were communities and the rationale for splitting sites in an alignment into groups.

And that is what I did this morning.