Scalable cell-specific coexpression networks for granular regulatory pattern discovery with NeighbourNet

Gene networks provide a fundamental framework for understanding the molecular mechanisms that govern gene expression. Advances in single-cell RNA sequencing (scRNA-seq) have enabled network inference at cellular resolution; however, most existing approaches rely on predefined clusters or cell states, implicitly assuming static regulatory programs and potentially missing subtle, dynamic variation in regulation across individual cells. To address these limitations, we introduce NeighbourNet (NNet), a method that constructs cell-specific coexpression networks. NNet first applies principal component analysis to embed gene expression into a low-dimensional space, followed by local regression within each cell's k-nearest neighborhood (KNN) to quantify coexpression. This approach improves computational efficiency and stabilizes coexpression estimates, mitigating challenges posed by small sample sizes in KNN regression and the inherent noise and sparsity of scRNA-seq data. Beyond coexpression, NNet supports scalable downstream analyses, including (i) clustering and aggregating cell-specific networks into meta-networks that capture primary coexpression patterns and (ii) integrating prior knowledge to annotate coexpression and infer active signaling interactions at the individual cell level. All functional modules of NNet are implemented with an efficient algorithm that enables the application to large-scale single-cell data sets. We demonstrate NNet's effectiveness through three case studies on transcription factor activity prediction, early hematopoiesis, and tumor microenvironments. Provided as an R package, NNet offers a novel framework for exploring cellular variation in coexpression and integrates seamlessly with existing single-cell analysis workflows.