In the current study, we first perform a comparative analysis of genomic wave artifacts in several different high-density SNP genotyping arrays, and confirm that genomic waves are not a platform-specific phenomenon. Next, we perform exploratory analysis of local genomic features (such as GC content, gene content and segmental duplication patterns), to find potential genomic features that correlate with genomic waves. We investigate the technical cause of waves by examining potential DNA degradation or protein contamination, and by performing serial dilutions from the same sample to assess the impact of DNA quantity. These experiments allowed us to identify the property of DNA samples that leads to genomic waves and to find ways to reduce genomic waves in the experimental protocol. Finally, we present a method to computationally reduce the effects of genomic waves and show that this approach reduces the wavy patterns of signal intensities and improves the accuracy of CNV detection.
