BlueFuse

High throughput platforms, such as microarrays and mass spectrometry, enable a systematic view of biology to be developed because they enable the relative abundance of genes, proteins and metabolites between samples, such as disease and healthy, to be compared.

Analysis of this data remains challenging due to issues of experimental variability, noise and other confounding artefacts in the data, non-linearity in the underlying biology, a lack of a standardised framework for assessing quality and problems of reducing to a computationally and mathematically tractable size the amount of data involved.

BlueFuse exploits, extends and customizes the most recent advances in Bayesian statistical modelling and machine learning methodology to provide an automated solution for the extraction and combination of higher quality information from such data sets. By modelling the physical processes generating the data BlueFuse is able to deliver, fully automatically, highly optimised and reproducible results supported by mathematically robust estimations of quality.

BlueFuse is applicable to most high throughput experimental data and has been applied by BlueGnome to genomic, proteomic and metabolomic data generated by microarrays, mass spectrometry, 2d gel, NMR and a range of cellular assays.

© cambridge bluegnome 2006 info@cambridgebluegnome.com