Waved aCGH: to smooth or not to smooth

Array-based comparative genomic hybridization (aCGH) is a powerful tool to detect genomic imbalances in the human genome. The analysis of aCGH data sets has revealed the existence of a widespread technical artifact termed as 'waves', characterized by an undulating data profile along the chromosome. Here, we describe the development of a novel noise-reduction algorithm, waves aCGH correction algorithm (WACA), based on GC content and fragment size correction. WACA efficiently removes the wave artifact, thereby greatly improving the accuracy of aCGH data analysis. We describe the application of WACA to both real and simulated aCGH data sets, and demonstrate that our algorithm, by systematically correcting for all known sources of bias, is a significant improvement on existing aCGH noise reduction algorithms. WACA and associated files are freely available as Supplementary Data.