BABAR: an R package to simplify the normalisation of common reference design microarray-based transcriptomic datasets

The paper describes an algorithm and software package that facilitates the normalisation of 2-colour microarray data prior to differential gene expression or network inference analysis.

BABAR is of particular benefit when applied to large, diverse datasets that are ‘heterogeneous’ in nature (i.e. comprising different result file formats, microarray layouts, and dye-swaps) and/or where there is a marked difference in the scale of gene expression between microarrays. Using real and simulated datasets we show that BABAR allows the correct interpretation of the data to be made via the application of the limma-implemented cyclic loess algorithm.

We believe our paper will be of interest because transcriptomic data are increasingly being used for network analyses. Such ‘systems’-level approaches rely upon large, diverse datasets that cannot be handled by existing applications. BABAR will automatically combine heterogeneous datasets which cannot be done by any existing software. BABAR enables such datasets to be normalised and made comparable, allowing existing data to be revisited and easily processed. BABAR will offer computational biologists a new way to compare published data which will be highly informative.

Downloads:

Reference

BMC Bioinformatics 2010, 11:73 - doi:10.1186/1471-2105-11-73