Improved processing of microarray data with probabilistic models

Abstracts are not currently available in GtR for all funded research. This is normally because the abstract was not required at the time of proposal submission, but may be because it included sensitive information such as personal details.

Microarray analysis allows the simultaneous evaluation of thousands of gene expression levels. In recent years both cDNA microarrays and oligonucleotide based arrays have become established technologies for genomic analysis across biology. Once the biological processes are complete the raw date generated from both the oligonucleotide arrays and the cDNA arrays consists of an image. This proposal concerns itself with the processing of the arrays once the image has been generated. We will use probabilistic models to extract information from the data, beginning with the image analysis, followed by normalisation and finally data mining. We will use Bayesian methods to integrate and propagate errors inherent in each stage of the probabilistic modelling process, eventually providing the biologist with levels of confidence for inferences made from the data. (Joint with BBS/B/00778)