Overview
The Cohort
Five late-stage, treatment-naïve primary epithelial ovarian carcinomas were profiled, including high-grade serous (HGSOC) and clear-cell (CCOC) subtypes. nih
The five samples were: OV440, OV773, OV594 (HGSOC), and OV511, OV150 (clear-cell ovarian carcinoma, CCOC).
Platforms & Methods
This is a dual-modality study integrating two genomic approaches on the same tumors:
1. Single-cell whole-genome sequencing (scWGS)
- Per-cell copy-number (CN) profiling at 10 Mb resolution
- Allele-specific CN analysis using CHISEL — enabling detection of loss of heterozygosity (LOH) at the single-cell level
- Cells clustered by maximum likelihood genetic clustering across successive rounds
- WGD status inferred per cell based on the fraction of genome with major CN ≥ 2
- Benchmarked against matched whole-exome sequencing (WES) and validated via variant allele frequency (VAF) concordance of germline and somatic mutations
2. Spatial transcriptomics (ST)
- An 11 mm × 11 mm tissue region profiled per sample
- CN inference from ST data using inferCNV with HMM-based CN prediction at the subcluster level (Leiden clustering)
- Malignant subgroups mapped spatially and correlated with histology (H&E)
Notes
Cancer Research Communications 2026. Paired scWGS + spatial modality extremely rare. Contains notable finding: spontaneous functional reversion of CTNNB1 driver mutation.