Network inference with Granger causality ensembles on single-cell transcriptomics

Abstract

Highlights d Pseudotime estimates order cells in a dynamic process using single-cell gene expression d SINGE infers gene regulatory networks from gene expression trends over pseudotime d SINGE's ensembling considers many smoothed versions of irregular pseudotemporal data d Uninformative pseudotime values can be detrimental to network reconstruction

Depicting pseudotime-lagged causality across single-cell trajectories for accurate gene-regulatory inference

“…Building upon several deep learning-based methods 18, 22 , we have demonstrated that combining fully-supervised deep learning with joint-probability matrices of pseudotime-lagged single-cell trajectories can overcome certain limitations of current Granger causality-based methods of gene-regulatory inference 9, 12 , such as their inability to infer cyclic regulatory motifs. Moreover, although our convolutional neural network DELAY remains sensitive to the ordering of single cells in pseudotime, the network can accurately infer direct gene-regulatory interactions from both timecourse and snapshot datasets, unlike many supervised methods that rely strongly on the number of available time-course samples 15, 16 .…”

“…Several methods for gene-regulatory inference rely on pseudotime in Granger causality tests, which try to determine whether new time series can add predictive power to inferred models of gene regulation 8, 9 . However, Granger causality-based methods can be error-prone when genes display nonlinear or cyclic interactions, or when the sampling rate is uneven or too low 9–12 . Because pseudotime trajectories exhibit these problems, Granger causality-based methods often underperform model-free approaches that exploit pure statistical dependencies in gene-expression data 9, 13, 14 .…”

“…Because pseudotime trajectories exhibit these problems, Granger causality-based methods often underperform model-free approaches that exploit pure statistical dependencies in gene-expression data 9,13,14 .…”

“…However, Granger causality-based methods can be error-prone when genes display nonlinear or cyclic interactions, or when the sampling rate is uneven or too low 9–12 . Because pseudotime trajectories exhibit these problems, Granger causality-based methods often underperform model-free approaches that exploit pure statistical dependencies in gene-expression data 9, 13, 14 .…”

Breaking the Immune Complexity of the Tumor Microenvironment Using Single-Cell Technologies

“…Various methods have been introduced in single-cell data analyses to study the intra- and inter-cellular interactome through the inference of gene regulatory networks (GRNs) and ligand-receptor pairs to understand the biological processes underlying these mechanisms. Several methods for GRN inference have been proposed to decipher intracellular networks, including tools originally designed for bulk transcriptomics ( Huynh-Thu and Sanguinetti, 2015 ; Moerman et al, 2019 ) and techniques specifically designed for single-cell transcriptomics that exploit additional information, such as pseudo-temporal ordering ( Matsumoto et al, 2017 ; Specht and Li, 2017 ; Deshpande et al, 2021 ) and information about transcription factors and their targets ( Aibar et al, 2017 ). The first class of methods tries to learn the gene regulatory structure without prior information, and the second class uses pseudo-temporal information to better explain gene regulation during cell differentiation and development.…”