Description

The tracks listed here contain data from the ClinGen ENIGMA BRCA1 and BRCA2 Expert Panel Specifications to the ACMG/AMP Variant Interpretation Guidelines for BRCA1/BRCA2 Version 1.2. The ENIGMA VCEP has adapted the ACMG-AMP codes for the BRCA1 and BRCA2 genes. These include the codes PVS1 (modified PVS1 decision tree), PS3/BS3 (functional data), PP4/BP5 (multifactorial data), PM5_PTC (PTC data, at exon level), the (potentially) clinically important functional domains defined by ENIGMA, and prediction programs (SpliceAI and BayesDel for PP3/BP4).

The data required for the application of these ENIGMA codes are displayed in 5 data tracks:

• BRCA1/BRCA2 protein domains 1.2 - Shows the (potentially) clinically important functional domains for the genes BRCA1 and BRCA2 as defined by ENIGMA. Data taken from Figure 1 of the specifications document, version 1.2.
• BRCA1/BRCA2 exon weights 1.2 - This track shows exon-specific weights for PM5_PTC ACMG code with evidence strong used for application of this code for novel protein termination codon (PTC) variants in an exon where a different proven pathogenic PTC variant has been seen before. Data taken from Supplementary Table 1 of the specifications appendix, version 1.2.
• BRCA1/BRCA2 functional assays 1.2 - Summary of BRCA1 and BRCA2 functional assay results reviewed for application of PS3 and BS3 ACMG codes. Data taken from Table 9 of the specifications, version 1.2.
• BRCA1/BRCA2 splicing 1.2 - Summary of ACMG codes applicable for variants considered against the BRCA1 and BRCA2 PVS1 decision trees. Includes PVS1 and PM5 codes recommended for initiation, nonsense/frameshift, deletion, duplication, and splice site (donor/acceptor ±1,2) variants, organized by exon. Data taken from Table 4 of the specifications, version 1.2.
• BRCA1/BRCA2 likelihood for PP4 and BP5 - Summary of BRCA1 and BRCA2 multifactorial likelihood analysis scores (displayed as Combined LR score) for ACMG codes PP4 and BP5. Combined LR scores were recalculated from 5 sources, Parsons et al. 2019, Caputo et al 2021, Li et al 2020, Easton et al 2007, and the case-control likelihood ratios from Zanti et al. 2025. All individual scores can be seen by clicking on the items. See the Methods section below for more information.

Display Conventions

• BRCA1/BRCA2 protein domains 1.2 - Items in fuchsia show clinically relevant protein domains. Mouseover on items shows the name of the domain and the location.
• BRCA1/BRCA2 exon weights 1.2 - Mouseover on exons (black items) show the transcript, exon number and the PM5_PTC code strength.
• BRCA1/BRCA2 functional assays 1.2 - Variants assigned with BS3 code in light blue, PS3 code in dark cyan, or no code assigned in black. Mouseover on items show variant HGVS nomenclature, assigned ACMG code and code weight.
• BRCA1/BRCA2 splicing 1.2 - Items show variant positions, red indicates exon deletions, blue exon duplications, purple items indicate variants with supporting RNA-based functional evidence. Mouseover on items show transcript, exon, variant position, possible variant type at that position and assigned ACMG code to the variant.
• BRCA1/BRCA2 likelihood for PP4 and BP5 - Variants assigned with PP4 code in brown, BP5 code in orange, and non-informative variants (no code assigned) in grey. Mouseover on items show variant HGVS nomenclature, combined likelihood ratio (LR) score, and assigned ACMG code and strength.

Data Access

The most up-to-date VCEP specifications for application of ACMG/AMP criteria for BRCA1 and BRCA2 genes are freely available at the ClinGen Criteria Specification (CSpec) Registry. This registry is intended to provide access to the Criteria Specifications used and applied by ClinGen Variant Curation Expert Panels and biocurators in the classification of variants.

Methods

These data were created and adapted from the files referenced above. Some custom scripting was employed in tasks like mapping variants, adding colors and mouseovers, and producing the desired format. The version 1.2 source tables were downloaded from the ClinGen CSpec Registry entries for BRCA1 (GN092) and BRCA2 (GN097), released January 9, 2025.

The multifactorial likelihood analysis scores were recomputed from Parsons et al. 2019 suppl table HUMU-40-1557-s001_Parson_Multicatorial.xlsx, as well as data from Caputo et al 2021, Li et al 2020, and Easton et al 2007. Variants present in both Parsons and Easton were removed from one dataset so as to only be counted once. All likelihood ratios of each matching type (Family LR, Co-occurence LR, segregation LR, pathology LR, and case control LR) were multiplied together and finally those values were then multiplied to create a combined LR score for the variant. Likelihood ratios are shown with five significant figures. All individual scores can be seen in the item description page.

Case-control evidence update (2026). The case-control likelihood ratio component of the PP4/BP5 track was updated in 2026. The earlier case-control LR from Parsons et al. 2019, derived from the iCOGS dataset and available for 20 variants, was replaced by the case-control likelihood ratio (ccLR) from Zanti et al. 2025, a case-control analysis of more than 400,000 women from the BRIDGES, CARRIERS, and UK Biobank studies. The older values were removed rather than kept alongside the new ones because the iCOGS dataset overlaps the cohorts analyzed by Zanti et al., and keeping both would count much of the same evidence twice. The other evidence types (family history, co-occurrence, segregation, and pathology) are unchanged, and the ccLR is multiplied in as the case-control term of the combined LR.

Case-control LRs from too few carriers (2026). Zanti et al. 2025 report a ccLR for every variant they analyzed, but label some of them "No evidence*" in their suggested evidence column. These are variants seen in fewer than three individuals across the combined dataset or, for BRCA2, with data from only one dataset (Supplementary Data 4, footnote h). Following guidance from the ENIGMA VCEP, these ccLRs are treated as not informative: they are still shown on the item's detail page, but they are not multiplied into the combined LR. A variant whose only evidence is such a ccLR has no combined LR and is shown in grey.

Interpreting the combined LR. The combined LR is a product of separate evidence types, and these can disagree. For some variants the case-control evidence from Zanti et al. 2025 points in the opposite direction from the earlier multifactorial evidence, and the combined score shows only the net result of multiplying them together. A variant with conflicting evidence can therefore have a combined LR that is weaker than, or opposite in direction to, one of its individual components. The per-source likelihood ratios are shown on each item's detail page, and we recommend reviewing them alongside the combined value, especially when they disagree. Also note that the evidence types may not be fully independent for all variants: some family history and co-occurrence values in Parsons et al. 2019 trace back to Easton et al. 2007, which is also included as its own source. Variants present in both were removed from one dataset so each is counted only once, but some residual overlap may remain.

Variant names and positions (2026). The source studies sometimes write the same variant in different ways, for example c.4574_4575del and c.4574_4575delAA, or a duplication written as an insertion. Variants are matched by the change they make to the genome sequence rather than by name, so each variant appears once and the likelihood ratios from all of its sources are multiplied together. Names follow current HGVS style: deleted or duplicated bases are not written out, and an insertion that copies the neighboring sequence is written as a duplication. Variant positions follow the same convention as the ClinVar track: a substitution covers its base, a deletion covers the deleted bases, a delins covers the replaced bases, and an insertion or duplication covers the two bases on either side of the insertion point. When a deletion or insertion falls in a repeat and could be placed at several positions, it is shown at the leftmost one.

For the complete details on the data processing see the makedoc on GitHub.

Changelog

• November 2026 - Following guidance from the ENIGMA VCEP, case-control likelihood ratios that Zanti et al. 2025 label "No evidence*" (too few carriers to assign an evidence strength) are no longer included in the combined LR. They are still shown on each item's detail page. This changed the ACMG code of 4,242 variants, 3,584 of which had no other evidence and are now grey (no code). See the Methods section for details.
• November 2026 - Variants that appeared twice because the source studies named them differently (for example c.4574_4575del and c.4574_4575delAA) were merged into a single item, and their likelihood ratios were combined. This affected 297 variants, and 99 of them now have a different ACMG code from any of their earlier entries. Variant positions now follow the ClinVar track convention, and likelihood ratios are shown with five significant figures. See the Methods section for details.
• August 2026 - The case-control evidence in the PP4/BP5 multifactorial likelihood track was updated from Parsons et al. 2019 (20 variants) to the case-control likelihood ratios of Zanti et al. 2025, and the combined LR scores were recalculated. The track grew from about 4,400 to about 13,500 variants. The other evidence types are unchanged. See the Methods section for details.
• August 2026 - All tracks and this page updated to ENIGMA specification version 1.2, released by ClinGen on January 9, 2025. In the splicing track, the BRCA2 c.1909+1G> and c.1909+2T> variants were upgraded from PVS1_N/A to PVS1, several splice sites gained RNA-based evidence annotations, and a BRCA2 start codon entry was added. The functional assays track now includes the PMIDs cited for published mRNA splice assay results. The protein domain and exon weight data are unchanged in version 1.2. The multifactorial likelihood track is built from published studies and is independent of the specification version.
• 2024 - Tracks updated to specification version 1.1.
• 2024 - Initial release of the hub, based on specification version 1.0.

Credits

Thank you to Luis Nassar from the Genome Browser team, Anna Benet-Pagès, Melissa Cline, Andreas Laner, Finja Hennig, Johanna Hadler, and Michael Parsons for technical coordination and consultation, and to the ENIGMA consortia for making these data available.

References

Enigma Guidelines: https://clinicalgenome.org/affiliation/50087/

Enigma Consortium: https://enigmaconsortium.org/

Parsons MT, Tudini E, Li H, Hahnen E, Wappenschmidt B, Feliubadaló L, Aalfs CM, Agata S, Aittomäki K, Alducci E et al. Large scale multifactorial likelihood quantitative analysis of BRCA1 and BRCA2 variants: An ENIGMA resource to support clinical variant classification. Hum Mutat. 2019 Sep;40(9):1557-1578. PMID: 31131967; PMC: PMC6772163

Caputo SM, Golmard L, Léone M, Damiola F, Guillaud-Bataille M, Revillion F, Rouleau E, Derive N, Buisson A, Basset N et al. Classification of 101 BRCA1 and BRCA2 variants of uncertain significance by cosegregation study: A powerful approach. Am J Hum Genet. 2021 Oct 7;108(10):1907-1923. PMID: 34597585; PMC: PMC8546044

Easton DF, Deffenbaugh AM, Pruss D, Frye C, Wenstrup RJ, Allen-Brady K, Tavtigian SV, Monteiro AN, Iversen ES, Couch FJ et al. A systematic genetic assessment of 1,433 sequence variants of unknown clinical significance in the BRCA1 and BRCA2 breast cancer-predisposition genes. Am J Hum Genet. 2007 Nov;81(5):873-83. PMID: 17924331; PMC: PMC2265654

Zanti M, O'Mahony DG, Parsons MT, Dorling L, Dennis J, Boddicker NJ et al. Analysis of more than 400,000 women provides case-control evidence for BRCA1 and BRCA2 variant classification. Nat Commun. 2025 May 25;16(1):4852. PMID: 40413188; PMC: PMC12103537

Li H, LaDuca H, Pesaran T, Chao EC, Dolinsky JS, Parsons M, Spurdle AB, Polley EC, Shimelis H, Hart SN et al. Classification of variants of uncertain significance in BRCA1 and BRCA2 using personal and family history of cancer from individuals in a large hereditary cancer multigene panel testing cohort. Genet Med. 2020 Apr;22(4):701-708. PMID: 31853058; PMC: PMC7118020