NetworkAnalyst is designed to support integrative analysis of gene expression data through
statistical, visual and network-based approaches:
Data inputs: one or more gene/protein lists with optional fold changes;
one or more gene expression tables from microarray or RNAseq experiments.
Network currently supports 13 species (eight model organisms & five common species).
Expression analysis & meta-analysis: support limma, edgeR and DESeq2.
The interface allows paired comparisons, time series, common reference, as well as
two-factor nested comparisons; for meta-analysis - p values, fold changes,
effect sizes, vote counts, and direct merge.
Network creation & customization: support protein-protein interactions,
TF-gene interactions, miRNA-gene interactions,
protein-drug interactions and protein-chemical interactions;
multiple functions for network refinement;
Network visual analytics: interactive visual exploration
- zooming, searching, highlighting, point-and-click, drag-and-drop; network customization -
six layout algorithms, background color, edge size/shape,
node size/color/visibility; in situ functional enrichment analysis (GO, pathways, etc.); network editing - node deletion, module extraction,
and image exporting; topology analysis - hubs, shortest
paths analysis, as well as three module detection algorithms.
Other visual analytics: interactive heatmaps, clustering (PCA & t-SNE), Venn diagram and
chord diagrams for pattern discovery & complex comparisons.
|