Pathogenic variant burden in the ExAC database: an empirical approach to evaluating population data for clinical variant interpretation

Abstract

Background: The frequency of a variant in the general population is a key criterion used in the clinical interpretation of sequence variants. With certain exceptions, such as founder mutations, the rarity of a variant is a prerequisite for pathogenicity. However, defining the threshold at which a variant should be considered "too common" is challenging and therefore diagnostic laboratories have typically set conservative allele frequency thresholds.

Methods: Recent publications of large population sequencing data, such as the Exome Aggregation Consortium (ExAC) database, provide an opportunity to characterize with accuracy and precision the frequency distributions of very rare disease-causing alleles. Allele frequencies of pathogenic variants in ClinVar, as well as variants expected to be pathogenic through the nonsense-mediated decay (NMD) pathway, were analyzed to study the burden of pathogenic variants in 79 genes of clinical importance.

Results: Of 1364 BRCA1 and BRCA2 variants that are well characterized as pathogenic or that are expected to lead to NMD, 1350 variants had an allele frequency of less than 0.0025%. The remaining 14 variants were previously published founder mutations. Importantly, we observed no difference in the distributions of pathogenic variants expected to be lead to NMD compared to those that are not. Therefore, we expanded the analysis to examine the distributions of NMD expected variants in 77 additional genes. These 77 genes were selected to represent a broad set of clinical areas, modes of inheritance, and penetrance. Among these variants, most (97.3%) had an allele frequency of less than 0.01%. Furthermore, pathogenic variants with allele frequencies greater than 0.01% were well characterized in publications and included many founder mutations.

Conclusions: The observations made in this study suggest that, with certain caveats, a very low allele frequency threshold can be adopted to more accurately interpret sequence variants.

Keywords: ACMG ISV guidelines; Allele-frequency threshold; ExAC; Variant interpretation.

Fig. 1

Concept diagram for using pathogenic variant frequency distributions to establish allele frequency thresholds. Depicted is a density plot of pathogenic variants in a hypothetical gene. x-axis: allele frequency; y-axis: number of pathogenic variants. The arrows (labeled A, B, and C) highlight three different scenarios of how allele frequencies of previously uncharacterized variants can be evaluated in the context of the pathogenic variant frequency distribution. In scenario A, the uncharacterized variant has an allele frequency that is highly consistent with known pathogenic variants. Because many benign variants are also rare or private, the allele frequency of this variant provide little weight towards either classification. In scenario B, the uncharacterized variant has an allele frequency that is consistent with known pathogenic variants, but is more common than the vast majority of them. The likelihood of such a variant being pathogenic is substantially reduced. In scenario C, the uncharacterized variant has an allele frequency outside of the pathogenic variant frequency distribution. The likelihood of such a variant being pathogenic is extremely low, as it is more common than any other previously characterized pathogenic variant

Fig. 2

Allele counts of BRCA1 and BRCA2 pathogenic and likely pathogenic variants. Histograms of the allele counts of pathogenic and likely pathogenic variants in the ExAC dataset for (a) BRCA1 and (b) BRCA2. x-axis: allele count; y-axis: number of unique sequence variants. The red portion represents pathogenic NMD_positive variants and the orange portion represents pathogenic NMD_negative variants (missense, intronic, in-frame indels, and truncations expected to avoid NMD)

Fig. 3

Number of publications for variants observed at an allele frequency greater than 0.01%. a The top box plot, in blue, summarizes the publication of a randomly selected set of 100 variants with a consensus classification of benign in ClinVar and having an allele frequency greater than 0.01%. The bottom box plot, in red, summarizes the publication counts of all 129 variants with a consensus classification of pathogenic in ClinVar and having an allele frequency greater than 0.01%. The box represents the 25th–75th percentile range, with the median publication count depicted as the horizontal line in the middle. The “whiskers” represents the maximum and minimum values. b Scatter plot of allele frequencies and publication counts of the same set of variants. Each red circle represents a pathogenic variant and each blue circle represents a benign variant. x-axis: allele frequency; y-axis: publication counts. Four pathogenic variants that are extreme outliers were excluded for display purposes: (1) GJB2 NM_004004.5:c.35delG (p.Gly12Valfs*2) has an allele frequency of 0.60% and reported in 496 publications; (2) CFTR NM_000492.3:c.1521_1523delCTT (p.Phe508del) has an allele frequency of 0.68% and reported in 966 publications; (3) SERPINA1 NM_001127701.1:c.1096G > A (p.Glu366Lys) has an allele frequency of 1.2% and reported in 56 publications; and (4) BTD NM_000060.3:c.1330G > C (p.Asp444His) has an allele frequency of 3.2% and reported in 46 publications