36. Classifying benign multigenic CNVs: Exploring available population databases

      Copy number variants (CNVs) are common in the normal population, and most have no known clinical effects. The pervasiveness of CNVs makes interpretation of chromosomal microarrays challenging. The new technical standards for interpretation and reporting of constitutional CNVs published by ACMG and ClinGen address the issue of CNVs in the normal population. The technical standards utilize a point based classification system and suggest that CNVs that “overlap with common population variation” are scored 0 to -1 points, with -1 (automatic benign classification) given to variants observed at ≥1% general population frequency. The document also states that one may attach less weight to variants observed at <1% in the general population and details are left to the user's judgment. Two databases are frequently used for CNV frequencies in the normal population: Database of Genomic Variants (DGV) and gnomAD. The DGV Gold Standard (DGV_GS) track contains merged calls from multiple high resolution and high quality studies, covering 0.52 billion nucleotides. Recently, gnomAD added approximately 445,000 CNVs covering 1.46 billion nucleotides. While gnomAD CNVs cover ∼3x more of the genome than DGV_GS, gnomAD variants are significantly smaller and allele frequencies are significantly lower than DGV_GS (p<0.001). Reliance on either database alone limits automatic benign classification for multigenic CNVs, resulting in reduced efficiency and increased reports of uncertain clinical significance. Furthermore, the choice of database may result in different classifications for many variants. To better understand the utility of each database, we will assign different weights (between 0 and -1) for various DGV_GS and gnomAD CNV frequencies and test the effect on final classification of CNVs previously classified in our laboratory. As CNV classification becomes standardized, there is a need for consensus on the weight given to normal population databases in the assignment of benign criteria.
      To read this article in full you will need to make a payment


      Subscribe to Cancer Genetics
      Already a print subscriber? Claim online access
      Already an online subscriber? Sign in
      Institutional Access: Sign in to ScienceDirect