STGRNS: an interpretable transformer-based method for inferring gene regulatory networks from single-cell transcriptomic data

Abstract

Motivation Single-cell RNA-sequencing (scRNA-seq) technologies provide an opportunity to infer cell-specific gene regulatory networks (GRNs) which is an important challenge in systems biology. Although numerous methods have been developed for inferring GRNs from scRNA-seq data, it is still a challenge to deal with cellular heterogeneity. Results To address this challenge, we developed an interpretable transformer-based method namely STGRNS for inferring GRNs from scRNA-seq data. In this algorithm, gene expression motif (GEM) technique was proposed to convert gene pairs into contiguous sub-vectors which can be used as input for the transformer encoder. By avoiding missing phase-specific regulations in a network, GEM can improve the accuracy of GRN inference for different types of scRNA-seq data. To assess the performance of STGRNS, we implemented the comparative experiments with some popular methods on extensive benchmark datasets including 21 static and 27 time-series scRNA-seq dataset. All the results show that STGRNS is superior to other comparative methods. In addition, STGRNS was also proved to be more interpretable than “black box” deep learning methods which are well-known for the difficulty to explain the predictions clearly. Availability The source code and data are available at https://github.com/zhanglab-wbgcas/STGRNS. Supplementary information Supplementary data are available at Bioinformatics online.

Single-cell transcriptome profiling reveals the spatiotemporal distribution of triterpenoid saponin biosynthesis and transposable element activity in Gynostemma pentaphyllum shoot apexes and leaves

“…Moreover, with the assistance of spatial information, subtypes such as palisade versus sponge mesophyll cells or upper versus lower epidermal cells could be resolved ( Xia et al., 2022 ).With the rapid advancement of single-cell transcriptomics technology, emerging tools have opened new avenues for diversified data analysis and utilization. For instance, STGRNS effectively infers cell-specific gene regulatory networks through the analysis of single-cell transcriptomic data ( Xu et al., 2023b ), while scmFormer delves into the intricate relationships between gene expression and protein levels by integrating single-cell proteomic data ( Xu et al., 2024 ). Looking ahead, the integration of single-cell transcriptomics with single-cell multi-omics technology will become a trend, providing us with more comprehensive and profound cellular analysis tools.…”

Spatially contrastive variational autoencoder for deciphering tissue heterogeneity from spatially resolved transcriptomics

“…Meanwhile, gene co-expression networks provide a way to enhance SRT data. Therefore, drawing on the current gene regulatory networks inference methods [ 40–42 ], the next crucial work is to effectively infer gene regulatory networks through spatial contrast modeling and investigate spatially specific regulatory factors.…”

Interpretable deep learning in single-cell omics

“…Alternatively, scGeneRAI attempts to infer GRNs by predicting the expression of a gene from a set of other genes using scRNA-seq data and layer-wise relevance propagation (LRP), a post-hoc and model-specific feature attribution technique ( Keyl et al 2023 ). Similar to scGeneRAI, STGRNS aims to reconstruct GRNs by predicting TF expressions using gene sets but relies on an intrinsic and model-specific approach where a transformer network with a multi-head attention layer is trained on scRNA-seq data ( Xu et al 2023 ). More recently, methods such as DeepMAPS have been developed for intrinsic and model-specific interpretable learning from multimodal single-cell omics data ( Ma et al 2023 ).…”