Likelihood-based calibration improves the clinical utility of JAG1 functional data for variant classification.

Calibration of JAG1 MAVE data into ACMG/AMP evidence weights via likelihood ratios

The authors recalibrate an existing MAVE dataset for JAG1, the primary cause of Alagille syndrome, by computing pathogenicity likelihood ratios translated directly into ACMG/AMP evidence weights. Calibration improved separation of benign and pathogenic variants and increased the number of classifiable abnormal missense variants from 486 to 610 (strong n = 1, moderate n = 340, supporting n = 269). Applied retrospectively to 29 individuals with a JAG1 VUS, it yielded evidence in nine (31%), six (21%) of which were upgraded to likely pathogenic or pathogenic. The framework is proposed as generalizable to other MAVEs.

Synthesis written by Geno'X. For the full original abstract, please refer to the source publication.

A concrete example of how high-throughput functional data (MAVEs), once calibrated to ACMG/AMP, reduce the VUS burden, a major bottleneck of exome/genome diagnostics. Reclassifying 21% of tested VUS has tangible clinical impact. Generalization will depend on assay-specific calibration quality.

Analysis by Dr Thibaut Benquey