In this hands-on session you will go through some basic applications in network biology.
You will learn
- Where to find biological interaction data?
- How to use Cytoscape?
- How to visualize experimental data on a network?
- How to open a pathway as a network?
- How to extend a network with regulatory information?
The questions in part 2, 3 and 4 are meant to guide you through the practical session. You will receive all the answers at the end of the course. Feel free to ask the instructors if you need help with any steps in the practical.
Part 1: Learn how to use Cytoscape
We will go through this part together with you to show you how Cytoscape works.
Cytoscape (www.cytoscape.org) is a popular and commonly used open source tool for network biology. The program has lots of different functionalities making it possible to perform a variety of network analysis, from the basics to advanced analysis. Since it is as open community project, many different researchers and developers contribute to Cytoscape. This is demonstrated by the wide variety of Cytoscape apps.
For the following two parts you will need different Cytoscape functionality that is explained in one of their tutorials online. You don’t need to go through the tutorials when you followed the introduction session.
Part 2: Using networks to analyze transcriptomics datasets
In this session we will continue using the dataset that you analyzed in the pathway analysis hands-on. You can find the dataset.xlsx file in the data-tutorial3 folder (in the zip file).
I selected the 9 significantly differentially down-regulated genes (logFC > 1 AND adj p-value < 0.1) and we will use GeneMania to find known interactions between the genes.
- Copy the gene symbols and use them as the query on GeneMania.
- Click on ‘Show advanced options’.
- Make sure that only ‘Pathway and Physical Interactions’ are selected. (Hint: sometimes you need to click multiple times to really deselect an interaction type)
- Make sure that you increase the number of related genes to 50.
Go to the “Functions” tab on the right side. GeneMania performs an over-repesentation analysis and provides a list of GO terms.
In the next step, we want to import this network in Cytoscape to visualize our experimental data on the gene nodes. Therefore, we need to download the network as text.
Then, we go to Cytoscape and import a network from the downloaded file (genemania_network.txt) – File → Import → Network → File.
First, we need to remove the first few lines from the file which are just comments. All those lines start with a “#”. So go to advanced options and add ignore all lines that start with #. The preview should now already change.
Then, we need to define the source and target column, so the columns that define the nodes in the network. We can do that by clicking on Entity 1 and select source (green filled circle) and on Entity 2 and select target (orange circle). Then just click okay and your network will be generated.
Now, we want to import and visualize the gene expression dataset (data-tutorial3/dataset.xlsx) in the network. Follow the same instructions as in the introduction tutorial for importing data.
- Go to File → Import → Table → File → Select dataset.xlsx in data-tutorial3 folder.
- Make sure that you use the correct key column. Cytoscape does not provide an identifier mapping solution. The network contains gene symbols, so we need to make the GeneSymbol column the key column that maps the data to the network. Click on the GeneSymbol column and select the key symbol.
You can save the whole session with File → Save. This Cytscape session file (*.cys) can be opened at a later stage.
Part 3: Biological pathways as networks
Open a new session in Cytoscape: File → New → Session.
In this part, we are going to open a pathway from WikiPathways as a network and look at the known network properties to identify important nodes in the network. We will use the WikiPathways app and the NetworkAnalyzer tool.
First we are going to open the Serotonin Receptor 4-6-7 and NR3C Signaling pathway in Cytoscape. You can find the GPML file in the data-tutorial3/selected-pathways folder.
- File → Import → Network → File…
- Select data-tutorial3/selected-pathways/Hs_Serotonin_Receptor_4-6-7_and_NR3C_Signaling_WP734_74438.gpml
- Make sure you select “Network” in the first properties dialog:
Use the NetworkAnalyzer to calculate the different network properties that were discussed during the lecture. Go to Tools → NetworkAnalyzer → Analyze Network. Treat the network as undirected.
Next, we are going to use visualize the node degree and betweenness on the nodes to easily find hub proteins in the network.
and the node degree as node color gradient (Map node color to → Select “Degree”).
- In the NetworkAnalyzer dialog, click on “Visualize Parameters”.
- Select the node betweenness as node size gradient.
- Select the node degree as node color gradient.
Part 4: Regulatory interactions
Open a new session in Cytoscape: File → New → Session.
In this part, we will look for known microRNAs and drugs targeting the transporters and receptors in the vitamin B12 pathway that we created in the first tutorial. For this step, we will use the CyTargetLinker app for Cytoscape.
The regulatory information is provided as networks, called Regulatory Interaction Networks (RegINs). For this practical the miRTarBase and the DrugBank RegINs are stored in the downloaded zip file in the directory data-tutorial3/RegINs. You can find other available RegINs for the future here.
First we need to open the vitamin B12 pathway. File → Import → Network → File…
Select data-tutorial3/selected-pathways/UpdateTransportVitaminB12.gpml. Import it as a network!
Now we are going to use CyTargetLinker to extend the network:
- Click on Apps → CyTargetLinker → Extend Network
- Select User Network → choose the network name of the pathway network
- Select Network Attribute → choose the name of the column containing a biological identifier (“GeneID”)
- Select Datasource Networks → Browse for the directory which contains the RegINs (network-analysis/RegINs)
- Select Direction → “Add both”
For the next questions, it is easier if you hide the microRNA-target interactions from miRTarBase. Go to the CyTargetLinker tab on the left and use the drop-down box next to the miRTarBase RegIN to hide the interactions.
For the next questions, it is easier if you hide the drug-target interactions from DrugBank. Go to the CyTargetLinker tab on the left and use the drop-down box next to the DrugBank RegIN to hide the interactions. Make sure that the miRTarBase interactions are shown!