Getting it out in the Open

"Bollocks to waiting 10 years for progress. I want people to know about it now, and then do something about it" - Dr Paul Fisher
UPDATE: Fisher didn't like the picture I used originally, so here's a better one.
I'm not taking the other one down though >>
Dr Paul Fisher ain't so happy

My friend and former colleague Paul Fisher has some workflows that link QTL (quantitative trait loci) data to pathway information in KEGG, with a bit of rudimentary text mining, to suggest which biochemical mechanisms underpin the trait. You can get a more complete description of the methods here, but my understanding of it is basically that QTLs are chromosomal regions known to be associated with a phenotypic trait. If Fisher finds that many genes in the QTL region map onto the same pathway, then that pathway might underpin the trait. So with the workflows we can move from a list of genes towards a mechanistic hypothesis.

Here the trait is Bilateral Perisylvian Polymicrogyria, a neurological condition in which the surface of the brain is irregular. Professor Wikipedia can tell you about it better than I, but then that's kind of the point of these workflows - they were developed for completely different traits but can easily be adapted.

So apart from it being the work of my friend, why have I focused on this? Fisher is a bioinformatician in a computer science department, not a clinician or an expert on BPP, who are the people who really need to see his results. Normal practice would be to publish them, most likely in a bioinformatics journal where the BPP experts probably won't see it, and this would all take far longer than necessary. So Fisher has decided to do away with that and just stick the data up on the Interweb straight away. Good for him.

It clearly is silly to slow down the release of valuable data by publishing everything in peer reviewed journals. The workflows are published and reviewed, and the data has been taken from the literature, so why not just get the new data from the workflows out as fast as possible? Oh yes, university politics again - research assessments and whatnot. Well I second Fisher's assertion - bollocks to it.

If we all thought more about the most appropriate way to disseminate our data then we'd surely conclude it's rarely wise to publish in slow, eye-wateringly expensive journals and being more open is the better choice. The king of all things Open Science is, of course, Cameron Neylon (if not the king then he is at least entitled to a very snazzy hat), who is always worth listening to. Following Fisher and Cameron, I have to work out how to get my data out there live and kicking, and have I got an Arduino-driven plan up my sleeve...

Fisher once made me go geocaching down a path in Cambridge. This is where you use your phone to look for some crap left somewhere by someone. It's a bit like dogging for people with no libido. After a good half hour rummaging in the bushes, I realised we were in fact poking around next to a nursery and this might have looked a bit dodgy. Perhaps some things shouldn't be done out in the open after all.


  1. In my defence to the geocaching fiasco - I did eventually find it, despite my GPS signal being way-off target, and the persistant rain ruining play.