Reconstruction of Structural Variation Breakpoint Junctions by Whole Genome Sequencing Split-Read Analysis

Document Type

Article

Publication Date

2026

JAX Location

JGM

Abstract

Breast cancer is the most diagnosed cancer in women1. BRCA1 gene deficiency is a major driver of hereditary breast cancer, particularly triple-negative breast cancer (TNBC)2. Previously, the Liu Lab reported that BRCA1-deficient tumors exhibit the tandem duplication phenotype (TDP), a genome-wide instability profile characterized by tandem head-to-tail genomic duplications (TDs), and are specifically enriched ~11 kb “class 1” TDs3. Recently, the Liu Lab has developed an in vitro longitudinally passaged clonal series using a Brca1/Trp53-deficient TDP tumor mouse model (B1P) and shown that class 1 TDs continue to arise over time in this model, demonstrating that the TDP is an ongoing process. Here, to interrogate whether de novo and pre-existing class 1 TDs arise through similar mechanisms, we developed a whole-genome sequencing (WGS) split-read analysis method to reconstruct TD breakpoint junction sequences and identify informative sequence patterns. Applying this approach to a subset of 50 structural variations (SVs) from the B1P dataset, we successfully reconstructed 48/50 (96%) breakpoint junction sequences, while traditional PCR followed by Sanger sequencing validated 41/50 (82%). Furthermore, 40/41 (97.6%) SVs showed concordance between junction sequences identified by both methods, establishing WGS split-read analysis as a robust and substantially more efficient approach for SV breakpoint reconstruction. Even from this SV subset, we already see that most de novo class 1 TDs show breakpoint homology, similar to pre-existing class 1 TDs—suggesting they share a mechanism of TD formation. Using this high throughput WGS split-read analysis, expansion to the full dataset will determine whether this pattern and TD mechanism are consistent over time.

Please contact the Joan Staats Library for information regarding this document.

COinS