FloREN: Decoding Immune Regulatory Networks through Interpretable Graph Transformer Patient Representations
Abstract
Summary Single-cell RNA sequencing (scRNA-seq) enables detailed characterization of cellular heterogeneity, yet understanding the full cellular and regulatory environment of complex tissues remains challenging. In the era of large single-cell atlases, this technology has become increasingly accessible, and datasets have grown in scale and statistical power. As a result, sample representation methods have emerged as a promising strategy to summarize patient-level biological variation. However, most existing approaches rely on unsupervised learning frameworks with ambiguous biological interpretability. Here we present a Framework for Learning Over REgulatory-Embedding Networks (FloREN), a supervised and interpretable sample representation method. FloREN models single-cell data as a heterogeneous network integrating cells and genes together with gene regulatory and cell-cell communication relationships. Through condition-aware embeddings and interpretable attention networks, FloREN enables improved sample stratification and biomarker discovery. In addition, the framework supports downstream analyses that found specific immune network mechanisms in immune-mediated inflammatory diseases (IMIDs).
Related
- Interpretable machine learning coupled to gene regulatory networks uncovers subcircuits underlying cell fate decisions
- Limitations of Genomic Foundation Models for Decoding Regulatory Mechanisms in ALS
- Functional decoding reveals a hidden regulatory layer of the Salmonella transcriptome during infection
- Association profile conditioning in a set-temporal transformer for cross-session intracortical motor decoding
- Interpretable Decoding of Frequency-Resolved Functional Connectivity
BibTeX
@misc{quere2026floren,
title = {{FloREN: Decoding Immune Regulatory Networks through Interpretable Graph Transformer Patient Representations}},
author = {Chloé Quéré and Valérie Devauchelle-Pensec and Philippe Thuillier and Florent Besson and Théo Guichaoua and Arthur Le Madec and Raphaële Seror and Marie Pean de Ponfilly and Daniel Wendling and Hatem Boulahdour and Tristan Pascart and Aurore Pacaud and Alban Bailliez and Thomas Eugène and Sebastien Ottaviani and Philippe Dieude and Khadija Ben Ali and Frederic Paycha and Augustin Latourte and Laurinda Carre and Bruno Fautrel and Patrice Cacoub and Serge Desarnaud and Pierre-Yves Salaün and Ronan Abgral and Solène Querellou},
howpublished = {European Journal of Nuclear Medicine and Molecular Imaging},
year = {2026},
month = jul,
doi = {10.64898/2026.07.12.738088},
url = {https://hal.science/hal-05837568v1},
note = {CC BY-NC-ND},
}
RIS
TY - UNPB
TI - FloREN: Decoding Immune Regulatory Networks through Interpretable Graph Transformer Patient Representations
AU - Chloé Quéré
AU - Valérie Devauchelle-Pensec
AU - Philippe Thuillier
AU - Florent Besson
AU - Théo Guichaoua
AU - Arthur Le Madec
AU - Raphaële Seror
AU - Marie Pean de Ponfilly
AU - Daniel Wendling
AU - Hatem Boulahdour
AU - Tristan Pascart
AU - Aurore Pacaud
AU - Alban Bailliez
AU - Thomas Eugène
AU - Sebastien Ottaviani
AU - Philippe Dieude
AU - Khadija Ben Ali
AU - Frederic Paycha
AU - Augustin Latourte
AU - Laurinda Carre
AU - Bruno Fautrel
AU - Patrice Cacoub
AU - Serge Desarnaud
AU - Pierre-Yves Salaün
AU - Ronan Abgral
AU - Solène Querellou
PY - 2026
DA - 2026/07/20
PB - European Journal of Nuclear Medicine and Molecular Imaging
DO - 10.64898/2026.07.12.738088
UR - https://hal.science/hal-05837568v1
AB - Summary Single-cell RNA sequencing (scRNA-seq) enables detailed characterization of cellular heterogeneity, yet understanding the full cellular and regulatory environment of complex tissues remains challenging. In the era of large single-cell atlases, this technology has become increasingly accessible, and datasets have grown in scale and statistical power. As a result, sample representation methods have emerged as a promising strategy to summarize patient-level biological variation. However, most existing approaches rely on unsupervised learning frameworks with ambiguous biological interpretability. Here we present a Framework for Learning Over REgulatory-Embedding Networks (FloREN), a supervised and interpretable sample representation method. FloREN models single-cell data as a heterogeneous network integrating cells and genes together with gene regulatory and cell-cell communication relationships. Through condition-aware embeddings and interpretable attention networks, FloREN enables improved sample stratification and biomarker discovery. In addition, the framework supports downstream analyses that found specific immune network mechanisms in immune-mediated inflammatory diseases (IMIDs).
SN - 1619-7070
SN - 1619-7089
ER -