TC-DATABASE Shiny app

Description

TC-DATABASE is a platform that combines a neo4j graph database with a shiny app to explore and analyze dynamic (temporal) biological networks. Analyses include dynamic behavior of pathway-related or GO-term-related subnetworks, gene neighborhoods, and gene modules, among others.

As default and example, we have included the dynamic coexpression network of gene expression data from small airway cells in non-smokers, healthy-smokers, and smokers with COPD (chronic obstructive pulmonary disease), during four time-points (0 months, 3 months, 6 months, and 12 months). We built the network from GEO data (GSE108134). We computed the gene-gene correlation coefficients (edge weights) using the WGCNA package for each time-point and each group. For pathway information, we used clusterProfiler::download_KEGG() to download human KEGG pathways and the biomaRt package to convert gene IDs. For Gene Ontology (GO), we used the GO.db and org.Hs.eg.db packages.

After collecting all gene-gene correlation scores, KEGG pathways, and GO terms, we imported all data into the neo4j graph database, and built the shiny app to perform the different analyses and visualizations.

Install and run TC-DATABASE:

Step1: Install R packages:

Go to R and install required R packages.

Step2: Run this shiny app:

Option a) Using neo4j from our lab:

# using our public neo4j:
library(RNeo4j)
graph = startGraph("http://www.moralab.science:3838/db/data/", username="neo4j", password="xiaowei")

#run shiny
library(shiny)
runGitHub("mora-lab/TC-DATABASE_shiny")

Option b) Using your local neo4j:
Before you run this shiny app, you need to:
(i) Start the neo4j database (version = 3.5.23).
(ii) Download our database.
(iii) Unzip the downloaded file and copy it to the $NEO4J_HOME/data/database folder.
(iv) Set dbms.active_database=neo4j-copd20201115 in the $NEO4J_HOME/conf/neo4j.conf file.

Now, you can go to R and run the following commands:

library(RNeo4j)
# you should change the username and password in this command
graph = startGraph("http://localhost:7474/db/data/", username="neo4j", password="password")

#run shiny
library(shiny)
runGitHub("mora-lab/TC-DATABASE_shiny")

Install TC-DATABASE from Docker and VirtualBox:

Click here

Tutorial:

The app consists of three tabs: Gene Relationships in KEGG Pathway/GO Term, Genes Neighborhoods Relationships and Alluvial Diagram.

1. Gene Relationships in KEGG Pathway/GO Term

This tab is for query genes relationship in each time and groups under KEGG pathway or/and GO term.

1.1 KEGG ID or KEGG pathway

You can input one or more KEGG ID or KEGG pathway to query. It means the genes are belong to KEGG pathway your query, and show that genes relationship. if you didn't input KEGG ID or KEGG pathway, the result will not show you any about KEGG information or relationship.

1.2 GO ID or GO Term

You also can input one or more GO Term. It means the genes are belong to GO term your query, and show that genes relationship. It is the same of KEGG option, the result will not show you any about GO information or relationship.

Notes
You can input GO Term and KEGG pathway at the same time.

1.3 Groups

You must choose at least one group for query.
We set this option has three groups: COPD smoker, smoker and nonsmoker. You can chose one or more groups to get whether those genes have relationship under groups your query.

1.4 Timeponts

Here has 4 timepoints, when you chose those timepoints, it means you want to query those gene relationship at that timepoint(NOT ONLY THAT TIMEPOINTS).

1.5 Weight

This option will be show if you chose the timepoints.
This weight is for WGCNA weight threshold, it has two option: the min weight and the max weight.

1.6 Plot

This option is ask you which type relationship you want plot in this network plot. It has three option: genes to genes, genes to KEGG, genes to GO. Only you chose KEGG pathway or GO, it will show you the genes to KEGG or genes to GO relationship.

1.7 Node information, Edge information, Network coordination scores and Download

That show you those nodes and edge information in the network plot. It also make some button for download those information.

2. Genes Neighborhoods Relationships

This tab is for query special genes relationship in each time and groups.

2.1 Gene symbol or entrezid

You must choose at least one gene for query. you can input gene symbol name or ENtrezid.

2.2 Groups

You must choose at least one group for query.
We set this option has three groups: COPD smoker, smoker and nonsmoker. You can chose one or more groups to get whether those genes have relationship under groups your query.

2.3 Timeponts

Here has 4 timepoints, when you chose those timepoints, it means you want to query those gene relationship at that timepoint(NOT ONLY THAT TIMEPOINTS).

2.4 Weight

This option will be show if you chose the timepoints.
This weight is for WGCNA weight threshold, it has two option: the min weight and the max weight.

2.5 Plot

This option is ask you which type relationship you want plot in this network plot. It has three option: genes to genes, genes to KEGG, genes to GO. If you didn't chose genes to genes, it means you only want to plot relationship between KEGG/GO and genes your input.

Node information, Edge information, Network coordination scores and Download

That show you those nodes and edge information in the network plot. It also make some button for download those information.

3. Alluvial Diagram

3.1 Groups

We set this option has three groups: COPD smoker, smoker and nonsmoker. You only chose one group to plot. This Alluvial plot will show you genes changing in WGNCA module in each timepoints in special group.

3.2 Alluvial data and download

Here show you the alluvial plot data and download option.

Last updated: Oct.31st, 2021