scReGAT: Leveraging Knowledge of Regulatory Interactions to Predict Long-range Gene Regulation at Single-cell Resolution.

Inferring Gene Regulatory Networks in Stem Cells: Methods and Applications.

Mapping enhancer–gene regulatory interactions from single-cell data

A family of classification models, scE2G, is introduced that predict enhancer–gene regulatory interactions from single-cell datasets and enable mapping of these interactions across diverse cell types and tissues and will enable accurate mapping of enhancer–gene regulatory interactions across thousands of human cell types.

RegFM: an interpretable context-aware foundation model for human transcriptional regulation

Transcriptional regulation is governed by interactions between cis-regulatory elements (CREs) and trans-acting regulators in a context-specific manner. Although DNA and single-cell foundation models have enabled modeling regulatory biology at scale, most represent either sequence or cellular state alone, limiting their ability to capture context-dependent gene regulation. Here we present RegFM, a context-aware foundation model for human transcriptional regulation. RegFM treats transcriptional regulation as a dialogue between cis-regulatory sequences (e.g., CREs) and trans-acting regulators (e.g., transcription factors (TFs) and chromatin regulators (CRs)) by coupling long-range CRE representations with TFs and CRs activity. Trained on large-scale ENCODE and CELLxGENE transcriptomic profiles, RegFM learns gene-centered regulatory representations that generalize across unseen cellular contexts. In a wide range of tasks, including gene expression prediction, cis-regulatory element annotation, bivalent promoter and dosage-sensitivity classification, and perturbation-response prediction, RegFM consistently improves over existing methods. RegFM emerges as a scalable and interpretable framework for modeling human transcriptional regulation and provides insights into context-dependent gene regulatory programs.

IAGRN: An Interleaved-Attention Graph Neural Network for Gene Regulatory Network Inference

A structure-aware interleaved-attention graph learning framework, termed IAGRN, is proposed for GRN inference from scRNA-seq data that interleaves topology-constrained local attention with distance-aware global attention, enabling effective integration of structural priors and long-range regulatory signals.

ScanNet: Single-cell annotation informed by transcriptional regulation Network via iterative heterogeneous graph learning

The ScanNet framework is a scalable, transferable, and mechanistically informed framework for accurate cell type annotation across diverse single-cell data modalities and can be flexibly transferred to single-cell ATAC-seq data by mapping chromatin accessibility to gene level.

Causal assessment of Bayesian gene regulatory networks from single-cell transcriptomics.

Gene regulatory network (GRN) inference is an essential tool for revealing dysregulated relationships between genes in different cell types from single-cell transcriptomic (SCT) data. GRNs based on Bayesian networks (BNs) learned from SCT data can elucidate directed regulatory relationships representing complex disease mechanisms and their interplay through graphical modeling. However, software for learning BNs from SCT data is not widely available, nor is software for evaluating the BNs' structural accuracy in representing causal relationships between genes. Here, we describe the scstruc R package. This package provides a suite of BN structure learning algorithms specifically designed to handle SCT data, to evaluate the resulting networks based on the causal relationships they represent regardless of the availability of established molecular interaction networks, and to compare regulatory relationships between conditions. We demonstrated that scstruc can identify biologically relevant differential regulatory relationships between groups on a per-cell basis.