NeighbourNet: Scalable cell-specific co-expression networks for granular regulatory pattern discovery
We introduce the latest work from our PhD student Yidi Deng (who has already secured a postdoc position at ANU!).
Understanding how genes interact at the single-cell level is key to unlocking the complexity of biological systems. Traditional GRN inference often overlooks the nuances within individual cells by relying on predefined clusters or assuming static regulation.
NNet has been designed to construct cell-specific co-expression networks using local neighbourhoods in the gene expression space. NNet addresses common scRNA-seq challenges like noise and sparsity by:
– Embedding gene expression via PCA
– Applying local regression within each cell’s k-nearest neighbours
– Stabilising co-expression estimates and scaling efficiently to large datasets
With NNet, researchers can:
– Aggregate and cluster networks into meta-networks
– Integrate prior knowledge to infer active signalling
– Explore dynamic regulation in contexts like haematopoiesis, tumour microenvironments, and transcription factor activity
Yidi Deng, Jiadong Mao, Jarny Choi, Kim-Anh Lê Cao. NeighbourNet: Scalable cell-specific co-expression networks for granular regulatory pattern discovery. bioRxiv 2025.03.27.645629; doi:https://doi.org/10.1101/2025.03.27.645629
Abstract Gene regulatory networks (GRNs) provide a fundamental framework for understanding the molecular mechanisms that govern gene expression. Advances in single-cell RNA sequencing (scRNA-seq) have enabled GRN 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 co-expression 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 neighbourhood (KNN) to quantify co-expression. This approach improves computational efficiency and stabilises co-expression estimates, mitigating challenges posed by small sample sizes in KNN regression and the inherent noise and sparsity of scRNA-seq data. Beyond co-expression, NNet supports scalable downstream analyses, including (i) clustering and aggregating cell-specific networks into meta-networks that capture primary co-expression patterns, and (ii) integrating prior knowledge to annotate co-expression and infer active signalling interactions at the individual cell level. All functional modules of NNet are implemented with an efficient algorithm that enables application to large-scale single-cell datasets. We demonstrate NNet’s effectiveness through three case studies on transcription factor activity prediction, early haematopoiesis, and tumour microenvironments. Provided as an R package, NNet offers a novel framework for exploring cellular variation in co-expression and integrates seamlessly with existing single-cell analysis workflows.
