miRNAtpa

Micro RNA target prediction analysis (miRNAtpa) is a tool for predicting the importance of each miRNA in gene regulation when specific parameters are given. The analysis considers both intermolecular interactions and tissue-specific expression levels of miRNA.

There are 2 parameters that can be customized:

• list of body regions
• list of target genes

NOTE: This tool does not provide fully automated analysis. Manual download of data is required.

Interaction data

Data about interactions miRNAs have with specific genes and their gene products is obtained from 5 biological databases:

• DIANA microT
• miRDB
• miRmap
• miRWalk
• TargetScan

Expression data

The expression of miRNA varies by tissue and developmental period of the organism. Expression data was obtained from 2016 study Distribution of miRNA expression across human tissues by Ludwig et al. (DOI: 10.1093/nar/gkw116).

Requirements

• Git
• Python 3.6 or later

Install

git clone https://github.com/ualich/miRNAtpa.git
cd miRNAtpa
pip3 install .

Usage

1. Prepare file architecture:

   • Select analysis name. Use alphanumeric characters.
   • Create analysis directory inside analyses, use analysis name.
   • Create a subdirectory input inside analysis directory.

2. Create a subdirectory inside input for every biological database and download interaction data (see directory anayses/example/input):

   • diana - download results as CSV file
   • mirdb - download results as HTML file (save webpage)
   • mirmap - download results as CSV file
   • mirwalk - export results as CSV file
   • targetscan - download table as TXT file

3. Run the command

   python3 mirnatpa/main.py <analysis_name> <regions>

   Where

   • analyses_name - alphanumerical name of the analysis
   • regions - space separated names of regions used for expression analysis, possible values can be found in the file expression/regions.txt

4. Results will be generated inside analysis directory, in subdirectory results.