Package index • scGraphVerse

Network Inference

Core functions for inferring gene regulatory networks from single-cell data

infer_networks(): Infer Gene Regulatory Networks from Expression Matrices

create_consensus(): Create a Consensus Adjacency Matrix from Multiple Networks

compare_consensus(): Compare Consensus and Reference Graphs or STRINGdb Networks

PCzinb(): Structure learning for count data using PC algorithms

Data Preparation

Functions for preparing and organizing input data

create_mae(): Create MultiAssayExperiment from Multiple Single-Cell Datasets

build_network_se(): Create a SummarizedExperiment for Network Storage

Network Analysis

Tools for analyzing network structure and properties

compute_topology_metrics(): Compute Network Topological Properties

compute_community_metrics(): Compute Community Assignment Similarity Metrics

community_similarity(): Compare Community Assignments and Topological Properties

classify_edges(): Classify Edges as TP, FP, or FN

edge_mining(): Edge Mining of Gene Interactions Using PubMed

community_path(): Community Detection and Pathway Enrichment Analysis

Visualization

Plotting functions for networks and analysis results

plotg(): Visualize Graphs from Adjacency Matrices

plot_network_comparison(): Visualize Network Comparison

plot_community_comparison(): Visualize Community and Topology Comparison

plotROC(): Plot ROC Curves for Inferred Networks

Network Utilities

Helper functions for network manipulation and processing

symmetrize(): Symmetrize Adjacency Matrices in a SummarizedExperiment

generate_adjacency(): Generate Adjacency Matrices from Gene Interaction Tables

cutoff_adjacency(): Threshold Adjacency Matrices Based on Shuffled Network Quantiles

stringdb_adjacency(): Build Adjacency Matrices for Physical Interactions from STRING (POST API)

selgene(): Select Top Expressed Genes from Single-Cell Data

pscores(): Compute Performance Scores for Predicted Adjacency Matrices

Setup and Configuration

Functions for initializing Python dependencies and downloading reference data

init_py(): Initialize Python Environment for GRNBoost2

download_Atlas(): Download and Load an RDS File from a URL

Example Data

Toy datasets for testing and examples

toy_counts: Toy MultiAssayExperiment for Network Inference

toy_adj_matrix: Toy adjacency matrix for examples

zinb_simdata(): Simulate Zero-Inflated Negative Binomial (ZINB) Count Matrices with Sequencing Depth

Internal Functions

Internal helper functions (not typically called directly by users)

earlyj(): Modify Cell Names and Combine Datasets

nb.loglik(): Log-likelihood of the negative binomial model Given a vector of counts, this function computes the sum of the log-probabilities of the counts under a negative binomial (NB) model. The NB distribution is parametrized by two parameters: the mean value and the dispersion of the negative binomial distribution

nb.loglik.dispersion(): Log-likelihood of negative binomial model, for a fixed dispersion parameter

nb.loglik.regression(): log-likelihood of the NB regression model

nb.loglik.regression.gradient(): Gradient of the log-likelihood of the NB regression model

nb.OptimizeDispersion(): (NB) model. The NB distribution is parametrized by two parameters: the mean value and the dispersion of the negative binomial distribution

nb.regression.parseModel(): Parse ZINB regression model

zinb0.noT(): Structure learning with zero-inflated negative binomial model (mean only)

zinb1.noT(): Structure learning with zero-inflated negative binomial model

zinbOptimizeDispersion(): (ZINB) model. The ZINB distribution is parametrized by three parameters: the mean value and the dispersion of the negative binomial distribution, and the probability of the zero component.

zinb.regression.parseModel(): Parse ZINB regression model