Copy number variation (CNV) in genomic DNA is linked to a variety of human diseases, and array-based CGH (aCGH) is currently the main technology to locate CNVs. Although many methods have been developed to analyze aCGH from a single array/subject, disease-critical genes are more likely to be found in regions that are common or recurrent among subjects. Unfortunately, finding recurrent CNV regions remains a challenge. We review existing methods for the identification of recurrent CNV regions. The working definition of ``common'' or ``recurrent'' region differs between methods, leading to approaches that use different types of input (discretized output from a previous CGH segmentation analysis or intensity ratios), or that incorporate to varied degrees biological considerations (which play a role in the identification of ``interesting'' regions and in the details of null models used to assess statistical significance). Very few approaches use and/or return probabilities, and code is not easily available for several methods. We suggest that finding recurrent CNVs could benefit from reframing the problem in a biclustering context. We also emphasize that, when analyzing data from complex diseases with significant among-subject heterogeneity, methods should be able to identify CNVs that affect only a subset of subjects. We make some recommendations about choice among existing methods, and we suggest further methodological research.


Bioinformatics | Computational Biology