Sunday 3 June 2012

Seven of Fifty-One

I am Seven of Fifty-One. I know it doesn't sound quite as good as Seven of Nine, but allow me to explain.

Those of you who have been reading this blog for a while may remember that part of the work I've been doing over the last few years has been related to cancer studies (I've blogged about it here, here, here, and here). Specifically I've been involved with trying to reduce the cost of performing Genome Wide Association Studies, by reducing the number of samples required. We have been doing this by mining the published literature to try and determine the prior probability that a specific single nucleotide polymorphism (SNP) will be related to specific cancers.

Essentially the idea boils down to this (although I have to admit in advance that neither the genetics or the statistics are my strong points). If a SNP occurs near a gene that has been shown to be involved in either the specific cancer under investigation or in the healthy functioning of the organ/tissue that the cancer affects then it is more likely that this SNP might be the cause of the cancer than if it occurs somewhere else in the genome. We calculate the probability that this is the case by mining the published literature and when these priors are combined with results from both healthy controls and diseased patients the statistical power of the tests are improved. As the power of the test is improved we can determine that a SNP is related to a cancer using less samples than previously. While this will save money it is especially important when investigating rare forms of cancer where it might be difficult or impossible to ever assemble the required number of patients for testing using the standard approach.

We'd previously shown that the approach worked, but only by going back and reproducing previous results using less data. Now, however, we've used the same approach to find a new cause for cancer! The results have just been published in the journal PLoS ONE: Using Prior Information from the Medical Literature in GWAS of Oral Cancer Identifies Novel Susceptibility Variant on Chromosome 4 - the AdAPT Method. PLoS ONE is an open access journal so you can all go and read the article if you want to!

By this point you might be wondering what this all has to do with the title and opening paragraph (and I can't say I'd blame you). Well I'm author number seven of fifty-one.

Post a Comment