Multiple lines of evidence indicated that microarray analysis has a better diagnostic yield of clinically significant genetic changes than do conventional methods in patients with constitutional abnormalities. However, interpretation of microarray data is complicated by the presence of both novel and recurrent copy number variants (CNVs) of unknown significance. To address this issue, Hehir-Kwa et al. described a new computational method for determining the pathogenicity between benign and mental retardation (MR)-associated CNVs among patients with MR. This study demonstrated the value of objectively prioritizing MR-associated CNVs using structural and functional genomic features in diagnostics. In this regard, we discuss an evidence-based summary of how to classify pathogenic or benign status of a CNV in clinical genetics and advocate that there is a need for algorithmic adjustment between constitutional cytogenetic and prenatal diagnosis settings.