Table 2.

Recently developed network inference tools, input data, and their core algorithmic approaches. Links to code/tools are provided in Supplementary Table 2.

Tool	Input data	Algorithmic approach	Example application	Reference
WGCNA	Snapshot expression data	Correlation	Identify differences in brain co-expression networks underlying the pathology of autism[73]	[48]
ARACNE	Snapshot expression data	Information theory	Resolve the regulatory structure of co-expression modules and identify disease drivers in post-traumatic stress disorder[74]	[50], [51]
CLR	Snapshot expression data	Information theory	Inference of miRNA co-expression network in type 2 diabetes[75]	[52]
MRNET	Snapshot expression data	Information theory	Network analysis of WNT1-regulated network in neurodegeneration[76]	[77]
GENIE3	Snapshot expression data	Regression trees	Identification of TEADs as regulators of invasive cell states in melanoma[78]	[54]
GRNBoost2	Snapshot expression data	Regression trees	Investigating the role of NKD2 in kidney fibrosis[79]	[55]
PLSNET	Snapshot expression data	Partial least squares regression	Identify new ERRα partners in breast cancer[80]	[56]
FastGGM	Snapshot expression data	Gaussian graphical model	Applied to gene expression datasets in asthma and Alzheimer’s disease[58]	[58]
BiDAG	Snapshot expression data	Bayesian network; Dynamic Bayesian network	Identify network transition structures in colorectal cancer[81]	[64]
LBN	Snapshot expression data	Bayesian network	Evaluated on DREAM3 challenge dataset and the E. coli SOS DNA repair network[65]	[65]
ATEN	Time-series expression data	Boolean network	Evaluated on Drosophila segment polarity gene regulatory network[82]	[82]
GABNI	Time-series expression data	Boolean network	Benchmarked on DREAM dataset and yeast cell cycle dataset[83]	[83]
dynGENIE3	Time-series expression data	Ordinary differential equations + Regression	Construction of GRN governing circadian clock[84]	[85]
Inferelator	Time-series expression data	Ordinary differential equations + Regression	Learning the Bacillus subtilis regulatory network[86]	[87], [88], [89]
Ma et al.	Snapshot and time-series expression data	Ordinary differential equations	Benchmarked on DREAM4 dataset and E. coli and S. cerevisiae datasets[65]	[90]
BETS	Time-series expression data	Granger causality	Predicting the transcriptional response of A549 cells to glucocorticoids[91]	[91]
SWING	Time-series expression data	Granger causality	Infer time-delayed edges in a SOS response network in E. coli following DNA damage[92]	[92]
iRafNet	Snapshot expression data; prior knowledge	Regression trees	Studying clear cell renal cell carcinoma using phosphopeptide data[93]	[94]
PostPLSR	Snapshot expression data; multi-omics data	Partial least squares regression	Studying the effects of circadian rhythm on gene expression[95]	[95]
PANDA	Snapshot expression data; prior knowledge	Message passing	Construction of sex-specific networks in chronic obstructive pulmonary disease[96]	[97]
Lemon-Tree	Snapshot expression data; multi-omics data	Bayesian network	Identifying changes in gene interactions during spontaneous migraine attack[98]	[99]
MERLIN+Prior	Snapshot expression data; prior knowledge	Bayesian network	Infer a general neural lineage network in early hindbrain and spinal cord development[100]	[101]
GRACE	Snapshot expression data; prior knowledge	Regression trees; Markov random fields	Construction of developmental GRNs in Arabidopsis thaliana and Drosophila melanogaster[102]	[102]
KiMONO	Snapshot expression data; prior knowledge	Penalized regression	Identification of host factors for COVID-19[103]	[104]
Inferelator-AMuSR	Snapshot expression data; multiple conditions	Penalized regression	Inferring shared and cell-type specific interactions in cortical interneuron development[105]	[106]
fused LASSO	Snapshot expression data; multiple conditions	Penalized regression	Construction of GRNs in early immune response to Avian influenza A viruses (IAV)[107]	[108]
GRADIS	Snapshot expression data; known TF-target interactions	Support vector machines	Evaluated on DREAM4 and DREAM5 datasets[68]	[68]
DeepSEM	Single-cell RNA sequencing (scRNA-seq) data	Deep learning	Identification of marker genes in the mouse cortex[72]	[72]
DeepDRIM	scRNA-seq data	Deep learning	Identifying genes relevant to COVID-19[70]	[70]
CNNC	scRNA-seq data	Deep learning	Evaluated on mouse embryonic stem cells, bone marrow-derived macrophages, and dendritic cells to predict new TF-target relationships[69]	[69]
GraphReg	DNA sequence data; Chromatin interaction data	Deep learning	Evaluated on ENCODE data from human cell lines and data from mouse embryonic stem cells to predict CAGE expression signals[109]	[109]
Enformer	DNA sequence data	Deep learning	Variant effect prediction[110]	[110]
Basenji	DNA sequence data	Deep learning	Identifying causal eQTLs[111]	[112]