Back

Saturation genome editing of BARD1 resolves VUS and provides insight into BRCA1-BARD1 tumor suppression

Woo, I.; Casadei, S.; Snyder, M. W.; Smith, N. T.; Best, S.; Tejura, M.; Gupta, P.; McEwen, A. E.; Post, M.; Hamm, A.; Dawood, M.; Hosokai, A.; Xu, A.; Garge, R. K.; Fayer, S.; Brannan, T.; Richardson, M. E.; Pendyala, S.; Heidl, S.; Muffley, L.; Fowler, D. M.; Starita, L. M.

2025-11-06 genetic and genomic medicine
10.1101/2025.11.03.25339440 medRxiv
Show abstract

BARD1 encodes a dimeric partner of BRCA1 and is required for homology-directed double strand DNA break repair1-5. Germline loss-of-function variants in BARD1 are linked to elevated risk for breast cancer and neuroblastoma6-8 and PARP inhibitors have shown effectiveness against BARD1-deficient tumors. However, the majority of BARD1 variants identified by genetic testing for cancer risk are variants of uncertain significance (VUS)9 limiting clinical utility. We used saturation genome editing10 to assess the impact of nearly 11,000 single-nucleotide variants and three base-pair deletions across all 11 coding exons of BARD1 on cellular fitness and gene expression. The cellular fitness data are nearly perfectly concordant with known pathogenic and benign BARD1 variants9 (AUC > 0.99) and loss-of-function missense variants in all three functional domains were associated with elevated risk for breast cancer6,8. When used for clinical variant classification, our data resolved 95.4% of existing BARD1 variants of uncertain significance. Comparison of cellular fitness data to known structures11-14, further solidifies that BARD1s role in homology-directed repair is required for tumor suppression in humans15. These results will immediately improve clinical genetic testing and decision-making for patients with BARD1 variants and deepen our understanding of BARD1s role in maintaining genome integrity.

Matching journals

The top 3 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.