miRinGO: Prediction of GO terms indirectly targeted by human microRNAs

This R Shiny application can be used to predict Gene Ontology (GO) terms indirectly targeted by human microRNAs. In contrast to direct targets which are predicted to have binding sites, indirect targets are regulated transcriptionally through transcription factors (TFs).

Running the application

There are two ways to run this tool,

Online through Shinyapps.io platform

can be accessed directly from this link https://fadeel.shinyapps.io/miRinGO

Locally using RStudio

1. Download R shiny app source code from GitHub https://github.com/Fadeel/miRinGO/archive/master.zip
2. Open miRinGO.R in RStudio.
3. Make sure the following packages are installed (shinythemes, shiny, DT, dplyr, stringr, ggplot2,tm and wordcloud)
4. You can run the application by clicking the 'Run App' button

Input Parameters

Parameter	Description	Possible values
Tissue type	29 different tissues	Brain, colon, lung, ..
input miRNAs	miRBase miRNA ID	hsa-miR-9-5p
GO category	Choose the category of GO annotation	Biological process OR cellular component OR molecular function
Targeting mode	Choose mode of miRNA targeting	Direct OR indirect
Percentage of TargetScan target genes	percentage of top-ranked targeted genes (sorted by TargetScan v7.2 context++ score)	[20% - 100%], step size 20%
min. number of genes per GO term	GO terms with number of genes less than this number will be removed	integer value (default is 5)
Number of GO terms/pathways for visualization	Number of GO terms/pathways to be used for bar plot and word cloud	integer value (default is 10)

User interface

This app has two panels, left one for input data and parameters selection and the right one for displaying the results as shown in below (a) input parameters; (b) Table with miRNA GO enrichment analysis; (c) Bar plot and word cloud summarizing the top enriched GO terms.