DNF: A differential network flow method to identify rewiring drivers for gene regulatory networks

Abstract

Differential network analysis has become an important approach in identifying driver genes in development and disease. However, most studies capture only local features of the underlying gene-regulatory network topology. These approaches are vulnerable to noise and other changes which mask driver-gene activity. Therefore, methods are urgently needed which can separate the impact of true regulatory elements from stochastic changes and downstream effects. We propose the differential network flow (DNF) method to identify key regulators of progression in development or disease. Given the network representation of consecutive biological states, DNF quantifies the essentiality of each node by differences in the distribution of network flow, which are capable of capturing comprehensive topological differences from local to global feature domains. DNF achieves more accurate driver-gene identification than other state-of-the-art methods when applied to four human datasets from The Cancer Genome Atlas and three single-cell RNA-seq datasets of murine neural and hematopoietic differentiation. Furthermore, we predict key regulators of crosstalk between separate networks underlying both neuronal differentiation and the progression of neurodegenerative disease, among which APP is predicted as a driver gene of neural stem cell differentiation. Our method is a new approach for quantifying the essentiality of genes across networks of different biological states.

Keywords: differential network analysis; information entropy; network flow; network topology; neuronal differentiation.

Figure 1.

Comparison of different methods in detecting perturbed nodes using simulated datasets. DNF (the red bar) is compared with 5 methods (the blue bars), including 4 state-of-the-art differential network analysis methods and the random selection method. Results are averaged by 100 pairs of simulation networks.

Figure 2.

Comparison of different methods in uncovering statistically survival-related genes for four TCGA datasets. The height of bars corresponds to the number of statistically (p-value<0.05) survival-related genes uncovered by each method (top ranked 20 genes) for each TCGA dataset.

Figure 3.

The fitted scatter plot of the local (degree centrality) and global (closeness centrality) differential network topology of detected nodes by different methods for three single-cell RNA-seq datasets. Nodes in the plot are fitted by the local polynomial regression, and the color of area represents the confidence area of 95%, this confidence area describes the statistical confidence of tendency in topological differences

Figure 4.

Network topology and gene expression analysis to identify driver genes in neural stem cell differentiation. (A) The combined network topology of three gene ontology enrichment terms and their first-order neighbors in the top 20 nodes scored by DNF. The size of nodes represents the degree of genes in the network. The red represents the term of GO:0045664, the yellow represents the term of GO:0010001, the green represents the term of GO:0048708, and the blue represents the top 20 nodes score by DNF. (B) The average gene expression of temporal single-cell RNA-seq datasets. The gene expression of three cell types is standardized into read-counts-per-million (CPM) format.