Reducing uncertainty in genetic testing with Saturation Genome Editing

Figures

• 

  Figure 1. ClinVar variants in commonly sequenced genes. The number of variants reported as of September 2022 is shown for each of eight genes for which secondary findings are returned [37]. Variants of uncertain significance (VUS) or with conflicting interpretations are shown in gray. For most genes, VUS outnumber variants deemed pathogenic or benign.

• 

  Figure 2. The Saturation Genome Editing method for testing human variants. Variant libraries are designed to include all single nucleotide variants in a genomic region of interest, such as an exon of a gene harboring many VUS. Variants are synthesized in an oligo pool, cloned into a library, and integrated to the genome of human cells with CRISPR technology. In the HAP1 essentiality assay (right), variants are created in a gene required for cell growth. Loss-of-function variants lead to growth defects and are depleted from the population over time. Next-generation sequencing is used to quantify variant abundances at multiple timepoints, such that variant effects can be deduced from sequencing data.

• 

  Figure 3. SGE of BRCA1 improves clinical variant interpretation. SGE was used to produce a variant effect map of BRCA1 RING and BRCT domains, comprising function scores for each of 3,893 SNVs across 13 exons. (A) Function scores for SNVs in a single exonic region (exon 4) are plotted by position. Variants scoring similarly to nonsense variants are deemed loss-of-function variants (shaded). (B) Function scores discriminate known pathogenic variants with >95 % accuracy (top), and can be used as strong evidence for classifying VUS (middle) and variants yet to be reported in ClinVar (bottom). Data from Findlay et al. 2018.