Reference-free and robust copy number inference in single-cell tumor transcriptomes
- 1Bilkent University
Abstract
Copy number changes are a hallmark of cancer and reveal which cells are malignant and how tumors evolve. Single-cell RNA sequencing is widely used to infer them in individual tumor cells. Existing methods do so by normalizing each cell against a reference of normal cells. As a result, their output depends on which cells are analyzed together, they often label normal cells as aneuploid, and they struggle in samples without normal cells. Here we present REFCON, a deep-learning model that infers copy number from each cell's own expression, without a reference. We evaluated REFCON on more than 410,000 held-out cells from cell lines, organoids and 112 tumors of nine cancer types, profiled with seven sequencing technologies. It agreed more closely with single-cell DNA than existing methods, was unaffected by sample composition, and labeled far fewer normal cells as aneuploid. In patient-derived organoids, which contain no normal cells, it gave consistent profiles across passages and tracked clonal evolution. REFCON addresses long-standing limitations of reference-based inference and enables robust, reference-free copy number calling from single-cell transcriptomes.
Related
- Scalable and semi-automated Boolean network inference from multi-condition single-cell transcriptomes with scBOLT
- Robust Inference of Phage Host Infection Dynamics from Sparse Single-Cell Transcriptomic Profiles in Pseudomonas aeruginosa
- Optimizing copy number profiles for tumor cell phylogeny inference from low-coverage single-cell sequencing data
- ScanFold3: a robust and rapid approach for identifying structured RNA targets in genomes and transcriptomes
- Decoding Cell-level Circadian Phase and Rhythm-state Remodeling in Single-cell Transcriptomes With scRhythmBayes
BibTeX
@misc{gencturk2026referencefree,
title = {{Reference-free and robust copy number inference in single-cell tumor transcriptomes}},
author = {M. M. Gencturk and A. E. Cicek},
howpublished = {bioRxiv},
year = {2026},
month = aug,
doi = {10.64898/2026.08.03.742406},
url = {https://www.biorxiv.org/content/10.64898/2026.08.03.742406v2},
note = {CC BY-NC},
}
RIS
TY - UNPB
TI - Reference-free and robust copy number inference in single-cell tumor transcriptomes
AU - M. M. Gencturk
AU - A. E. Cicek
PY - 2026
DA - 2026/08/03
PB - bioRxiv
DO - 10.64898/2026.08.03.742406
UR - https://www.biorxiv.org/content/10.64898/2026.08.03.742406v2
AB - Copy number changes are a hallmark of cancer and reveal which cells are malignant and how tumors evolve. Single-cell RNA sequencing is widely used to infer them in individual tumor cells. Existing methods do so by normalizing each cell against a reference of normal cells. As a result, their output depends on which cells are analyzed together, they often label normal cells as aneuploid, and they struggle in samples without normal cells. Here we present REFCON, a deep-learning model that infers copy number from each cell's own expression, without a reference. We evaluated REFCON on more than 410,000 held-out cells from cell lines, organoids and 112 tumors of nine cancer types, profiled with seven sequencing technologies. It agreed more closely with single-cell DNA than existing methods, was unaffected by sample composition, and labeled far fewer normal cells as aneuploid. In patient-derived organoids, which contain no normal cells, it gave consistent profiles across passages and tracked clonal evolution. REFCON addresses long-standing limitations of reference-based inference and enables robust, reference-free copy number calling from single-cell transcriptomes.
ER -