CycloCross: Adversarial single-cell RNA-seq data translation across species
- 1Kyoto University
Abstract
Single-cell RNA sequencing (scRNA-seq) has transformed our understanding of cellular heterogeneity, yet most datasets are limited to a handful of model organisms, leaving critical gaps in cross-species biology. Existing computational methods for integrating multi-species scRNA-seq data often rely on one-to-one ortholog mapping, which fails to account for gene duplications, losses, or functional divergences. To address these challenges, we introduce CycloCross, an adversarial method based on the CycleGAN architecture, designed to translate scRNA-seq data between species without requiring a priori gene correspondences. Unlike generative models that synthesize data from noise, CycloCross leverages real data from a source species, enabling more realistic and data-efficient translations. We demonstrate its effectiveness in germ cell development, translating data between mouse, macaque, and human with conserved yet divergent transcriptional programs. CycloCross not only preserves temporal dynamics but also enables interpolation and extrapolation of gene expression at missing time points, offering predictions for experimentally inaccessible states. Furthermore, CycloCross generates biologically plausible samples even with limited data, outperforming traditional generative models.
Related
- Single-cell transcriptomic insights into conserved and divergent oligodendrocyte lineage programs across species
- Neural prediction decorrelation reveals that adversarial robustness substantially improves DNN prediction accuracy across the entire human auditory cortex
- Trait variation driven by tree size more than environment and shows limited translation to species vital rates
- Mechanistic modeling of bacterial translation initiation across growth conditions
- Sparsomycin inhibits translation through a conserved mechanism across all domains of life
BibTeX
@misc{hacquard2026cyclocross,
title = {{CycloCross: Adversarial single-cell RNA-seq data translation across species}},
author = {O. Hacquard and Y. Tokuta and Y. Imoto and T. Nakamura and S. Deguchi and M. Nagano and M. Saitou and Y. Hiraoka},
howpublished = {bioRxiv},
year = {2026},
month = sep,
doi = {10.64898/2026.09.24.754263},
url = {https://www.biorxiv.org/content/10.64898/2026.09.24.754263v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - CycloCross: Adversarial single-cell RNA-seq data translation across species
AU - O. Hacquard
AU - Y. Tokuta
AU - Y. Imoto
AU - T. Nakamura
AU - S. Deguchi
AU - M. Nagano
AU - M. Saitou
AU - Y. Hiraoka
PY - 2026
DA - 2026/09/24
PB - bioRxiv
DO - 10.64898/2026.09.24.754263
UR - https://www.biorxiv.org/content/10.64898/2026.09.24.754263v1
AB - Single-cell RNA sequencing (scRNA-seq) has transformed our understanding of cellular heterogeneity, yet most datasets are limited to a handful of model organisms, leaving critical gaps in cross-species biology. Existing computational methods for integrating multi-species scRNA-seq data often rely on one-to-one ortholog mapping, which fails to account for gene duplications, losses, or functional divergences. To address these challenges, we introduce CycloCross, an adversarial method based on the CycleGAN architecture, designed to translate scRNA-seq data between species without requiring a priori gene correspondences. Unlike generative models that synthesize data from noise, CycloCross leverages real data from a source species, enabling more realistic and data-efficient translations. We demonstrate its effectiveness in germ cell development, translating data between mouse, macaque, and human with conserved yet divergent transcriptional programs. CycloCross not only preserves temporal dynamics but also enables interpolation and extrapolation of gene expression at missing time points, offering predictions for experimentally inaccessible states. Furthermore, CycloCross generates biologically plausible samples even with limited data, outperforming traditional generative models.
ER -