3.1 Main functions in Network Viewer
Contents
1 Overview
1.1 Background
- MicroRNAs (miRNAs) regulate most cellular processes and are promising therapeutic candidates for cancer and other diseases.
- Understanding miRNA function is challenging due to the “many-to-many” relationships between miRNAs and their target genes.
- In addition, complex interplay exists between miRNAs and other functional elements, such as transcription factors, lncRNAs, etc.
- Network-based approach enables the integration of multiple data types and interpreting results at a systems level to allow the understanding of miRNA corporative functions and regulatory mechanisms.
1.2 Overall goal
To provide an easy-to-use web application to enable complex miRNA-centric network analytics for systems-level interpretation of miRNA functions and gene regulations.
1.3 New features in miRNet 2.0
The miRNet 1.0 was mainly about miRNA–target gene network. The miRNet 2.0 has expanded to include other key players involved in gene regulatory network with new modules, new knowledge bases and new visualization algorithms to support multipartite network creation, analysis and exploration
- Added support for miRNA family mapping and enrichment analysis
- Added support for tissue and exosomal specific miRNA annotation
- Added support for more species based on user’s feedback
- Added two modules to support xeno-miRNA targets exploration & visualization
- Added a new module to support miRNA-TF regulatory network
- Added a new module to support miR-SNP annotation and target interactions
- Added a new module to support integrating multiple types of molecules for systemslevel visual analytics
- Added additional network layout algorithms, including bipartite/tripartite, concentric and backbone layout
- Significantly expanded miRNA interaction knowledgebase
- Revamped the web interface, optimized the workflow, and introduced microservices and web application programming interface (API)
1.4 miRNet & its knowledgebase
1.5 Workflow
1.6 Network analysis
1.7 Network customization
1.8 Implementations
1.9 Computer and browser requirements for analysis
- A modern web browser with JavaScript enabled
- Supported browsers include Chrome, Safari, Firefox, and Internet Explorer 9+
- For best performance and visualization, use:
- Latest version of Google Chrome
- A computer with at least 4GB of physical RAM
- A 15-inch screen or bigger (larger is better)
1.10 Comparison with other tools
2How to create network from a list of miRNAs
2.1 Choose a module
Go to the miRNet home page (https://www.mirnet.ca) and click “miRNAs” to enter the module
2.2 Data input
“Genes” and “lncRNAs” are selected in this analysis
2.3 Network creation
2.4 Network trimming
3 How to explore and customize networks
3.2 Customizing Network (I)
A lncRNA-miRNA-gene competing endogenous RNA (ceRNA) network. The blue nodes (central zone) represent lncRNAs, green nodes (middle layer) represent miRNAs and red nodes (outer layer) represent genes.
3.3 Customizing Network (II)
You can select “Node type” (new feature) under “Scope” and drag to a separate area to further investigate.
3.4 Module detection & highlighting
Modules are tightly clustered subnetworks with more internal connections than expected by chance in the whole network. Members within a module are likely to work together to perform a biological function.
3.5 Exploring enriched biological functions
miRNet provides four query types (all genes, highlighted genes, all miRNAs, highlighted miRNAs), two enrichment algorithms (hypergeometric tests and empirical sampling), nine annotation libraries (three gene-set libraries and six miRNA-set libraries) for functional enrichment analysis.
3.6 Path of interactions between queries
3.7 Batch selection & highlighting
4 How to refine and filter networks
4.1 Filter in the interaction tables
Users can apply data filters to further improve the quality of the default miRNA–target interaction data.
To do this, click the “Advanced Filter” button on the top-right of the table to bring up the dialog. Users need to specify three parameters—the column to be filtered (“Target Column”); a keyword and matching criteria—“(Character) Matching,” “(Character) Containing,” or “(Numeric) At least”; and whether to “remove” or “keep” rows that meet the specified criteria.
4.2 Filter by topological measures
Use Network Tools to refine networks
4.2.1 Degree filter
4.2.2 Betweenness filter
4.2.3 Shortest path filter
4.2.4 Manual batch filter
4.3 Compute minimum network
You can use Minimum Network or Steiner Forest Network tool to construct a minimally connected network that contains all of the seed nodes. This means that the only added nodes are ones that connect previously disjointed networks of seed nodes.
5 How to perform network visual analytics
5.1 Analysis of node importance
Sort the nodes by their degree/betweenness values in descending order. Use the checkbox to select nodes to highlight.
To highlight seed nodes, please click on highlight seed nodes icon located in the vertical tool bar on the top left corner of Network Viewer.
5.2 Analysis using modules
The main goal in module analysis is to reduce the network complexity while still keeping the most interesting functions and connections.
5.2.1 Function-first approach
- Use the color palette to set a new highlight color (e.g., yellow)
- Select the ‘p53 signaling pathway’. The corresponding nodes will be highlighted in yellow and made larger within the current network
- Click the ‘Extract’ button on the viewer’s tool bar to extract a module that will contain all nodes that are related to the highlighted pathways
5.2.2 Connection-first approach
- Click on ‘module1’ above the ‘Global Node Styles’ panel on the upper left corner to reload the main network (mirnet1) into the network viewer
- Click ‘Module Explorer’ on the bottom right to bring up the panel, and then click the ‘Submit’ button. A list of modules will be returned together with summary statistics about their sizes and P values
- Click any module to view its nodes highlighted in the current network
- Extract the modules of interest by clicking the ‘Extract’ button in the network viewer
