BLOOM: Biologically Informed Flow Matching for Transcriptomic Virtual Embryo Modeling
- 1The University of Hong Kong
Abstract
Constructing a virtual embryo requires a generative representation of how heterogeneous cell populations change across developmental time. Here we introduce BLOOM, a biologically informed framework for conditional generation of developmental cell populations. BLOOM integrates temporal population context with developmental and regulatory priors through conditional flow matching to generate cell-expression distributions at queried stages in either temporal direction. Evaluation spans three mouse single-cell atlases covering prenatal development, visual-cortex maturation and multi-organ endothelial specialization. In comparisons with 22 time-series forecasting baselines, BLOOM ranks first for gene- and cell-expression correlations across all three datasets. Measured-stage evaluations support recovery of developmental expression patterns, population organization and marker-specific temporal relationships. Biological analyses connect these predictions to cell-state identities, gene programmes and regulatory associations. At unmeasured stages, interpolation and autoregressive history and future rollouts extend simulated expression dynamics between and beyond measured timepoints. Analyses of marker genes and temporal descriptors identify candidate differences in the timing of muscle-associated, neural and vascular-specialization programmes. Input-based gene perturbations further probe predicted responses of edited genes and associated molecular programmes across time. Together, BLOOM supports transcriptomic virtual embryo modeling by bridging temporal sampling gaps, exploring extended developmental trajectories and prioritizing gene-dependent hypotheses for experimental investigation.
Related
- Biologically Informed Multi-Omics Integration Reveals Clinically Meaningful Patient Representations in Acute Myeloid Leukemia
- A likelihood-based framework for simultaneously learning both noise and growth dynamics using biologically-informed neural networks
- EG-BINN: Parameter-Efficient Ensembling of Gated Biologically Informed Neural Networks for Epigenetic Age Estimation
- Reliable mechanistic operator recovery with biologically-informed neural networks: principles for architecture and optimisation design
- Rigorous validation of graph-based network analysis reappraises biologically coherent ADHD-associated transcriptomic modules in peripheral blood
BibTeX
@misc{liang2026bloom,
title = {{BLOOM: Biologically Informed Flow Matching for Transcriptomic Virtual Embryo Modeling}},
author = {Z. Liang and S. Li and B. Li and Y. Huang and L. Yu},
howpublished = {bioRxiv},
year = {2026},
month = oct,
doi = {10.64898/2026.10.26.761895},
url = {https://www.biorxiv.org/content/10.64898/2026.10.26.761895v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - BLOOM: Biologically Informed Flow Matching for Transcriptomic Virtual Embryo Modeling
AU - Z. Liang
AU - S. Li
AU - B. Li
AU - Y. Huang
AU - L. Yu
PY - 2026
DA - 2026/10/26
PB - bioRxiv
DO - 10.64898/2026.10.26.761895
UR - https://www.biorxiv.org/content/10.64898/2026.10.26.761895v1
AB - Constructing a virtual embryo requires a generative representation of how heterogeneous cell populations change across developmental time. Here we introduce BLOOM, a biologically informed framework for conditional generation of developmental cell populations. BLOOM integrates temporal population context with developmental and regulatory priors through conditional flow matching to generate cell-expression distributions at queried stages in either temporal direction. Evaluation spans three mouse single-cell atlases covering prenatal development, visual-cortex maturation and multi-organ endothelial specialization. In comparisons with 22 time-series forecasting baselines, BLOOM ranks first for gene- and cell-expression correlations across all three datasets. Measured-stage evaluations support recovery of developmental expression patterns, population organization and marker-specific temporal relationships. Biological analyses connect these predictions to cell-state identities, gene programmes and regulatory associations. At unmeasured stages, interpolation and autoregressive history and future rollouts extend simulated expression dynamics between and beyond measured timepoints. Analyses of marker genes and temporal descriptors identify candidate differences in the timing of muscle-associated, neural and vascular-specialization programmes. Input-based gene perturbations further probe predicted responses of edited genes and associated molecular programmes across time. Together, BLOOM supports transcriptomic virtual embryo modeling by bridging temporal sampling gaps, exploring extended developmental trajectories and prioritizing gene-dependent hypotheses for experimental investigation.
ER -