FlowGRN: Scalable and Dropout-Robust Gene Regulatory Network Inference via Flow Matching-Based Trajectory Reconstruction (Technical Report)
2026-08-10 • Computational Engineering, Finance, and Science
Computational Engineering, Finance, and Science
AI summaryⓘ
The authors developed FlowGRN, a new method to understand how genes control each other using data from single cells, even when the data is noisy or missing information over time. They combined techniques to better reconstruct the paths cells take as they change and to build gene networks more accurately. FlowGRN uses a special way to compare cells that works well despite data dropouts. Tests showed that FlowGRN performs very well on both artificial and real data. The authors also showed that both their new cell comparison method and the way they rebuild cell trajectories are important for accurate results.
Gene Regulatory NetworksSingle-cell RNA sequencingDropout noiseTrajectory reconstructionConditional flow matchingScore matchingdynGENIE3Cell similarity measureBEELINE benchmark
Authors
Tsz Pan Tong, Jun Pang
Abstract
Inferring gene regulatory networks (GRNs) from single-cell RNA sequencing (scRNA-seq) data offers insights into cellular behavior, but is complicated by the lack of temporal information and the prevalence of dropout noise. To address these challenges, we present FlowGRN, a method that integrates conditional flow matching and score matching for robust trajectory reconstruction with dynGENIE3 for scalable GRN inference. FlowGRN incorporates a novel cell similarity measure that is resilient to dropout effects in high-dimensional scRNA-seq data. Evaluation on the BEELINE benchmark demonstrates that FlowGRN achieves state-of-the-art performance on both synthetic and experimental datasets. Ablation studies validate the importance of both the dropout-robust similarity measure and the trajectory reconstruction step, highlighting FlowGRN's ability to accurately model dynamic regulatory relationships.