bindseek

Bindseek is a workflow that enables the identification of miRNA binding sites to the sequences of the 3'UTR of mRNA target genes. The workflow can be successfully used to search for miRNA binding sites in non-model organisms.

Table of Contents

• System Requirements
• Input Files
  • Gene List
  • Species List
  • miRNA Seed Sequence
  • Source of Target Sequences
  • Mature miRNA Sequence
• Installation and Running
• Output
  • Result Table
• Citation

System requirements

Bindseeker has been tested on Python 3.8 version. For the workflow to run, it is necessary to install: pandas, openxl and scipy Python packages, Perl and RNAhybrid. All dependencies are included in Dockerfile (Docker installation required).

Input files

Five inputs are required to run the workflow:

• list of genes that should be scanned through (.txt file),
• list of species of interest (.txt file),
• 8-characters long miRNA seed sequence (e.g. "UUCAAGUA"),
• source of target sequences (e.g. "3utr_human"),
• 22-characters long sequence of mature miRNA (e.g. "UUCAAGUAAUCCAGGAUAGGCU").

gene list

The gene list should contain one gene name (Gene Symbol) per line:

AARS
ACADM
ACLY
ACSL4
ACTBL2
ACTN1
ACTN4
ACTR1A
ACTR3
AHCY

An example of gene list provided in test_data directory ("gene_names_short.txt").

species list

The species list should contain one species name per line:

sus_scrofa
sus_scrofa_largewhite
sus_scrofa_berkshire
sus_scrofa_hampshire

The order of the given species names is important due to the fact that if the gene sequence for the first species is not found, the program will start looking for the gene in species lower on the list. An example of gene list provided in test_data directory ("sus_names.txt").

miRNA seed sequence

The seed sequence should be provided as 8-characters long combination of four letters (A, C, G, U).

source of target sequences

Used for a quick estimate of extreme value distribution parameters. You can choose between 3utr_fly, 3utr_worm and 3utr_human for better equitation within these species.

mature miRNA sequence

The sequence should be provided as 22-characters long combination of four letters (A, C, G, U).

Installation and running

The workflow is based on docker image. First, clone GitHub repository:

git clone https://github.com/compcore-irzbz/bindseek.git

Change directory to bindseek and run docker image build:

cd bindseek
docker build -t bindseek_img .

Run workflow as docker container:

docker run -v "/path/to/input/dir:/results" bindseek_img --genes_file gene_names.txt --species_file species_file.txt --motif UCCCUGAG --rnahybrid_param 3utr_human --mirna_sequence UCCCUGAGACCCUAACUUGUGA

The output of the workflow will be stored in given path, like: /path/to/input/dir.

Output

The Bindseek workflow provides two output files. The main result is a results table called complete_results.tsv, which provides a detailed summary of the miRNA binding sites identified in the 3'UTR sequences of target genes. The second output file is the utr.fa file, which stores all the 3'UTR sequences of the provided target genes, in the FASTA format.

Result table:

Gene	Binding	22-nt binding sequence	Start	End	Length	Prob.	GC content	MFE
ACLY	6mer	CAGTGTCTCTTTGTGTCAGGGG	584	589	877	0.19178	54.545	-18.5
ACSL4	6mer-A1	AATTTATCTTTGATAACAGGGA	949	954	2724	0.48516	27.273	-14.8
ACSL4	6mer-A1	AATATTGTTAAGGGACCAGGGA	2037	2042	2724	0.48516	40.909	-23.1
ACTN4	6mer-A1	TCTGGGGGGCGGGGGGCAGGGA	135	140	1075	0.22992	81.818	-28.8

Header description:

• 'Gene' - gene symbol,
• 'Binding' - type of binding identified,
• '22-nt binding sequence' - 22 nucleotide sequence derived from 3'UTR of tagrget gene (last characters include binding site),
• 'Start' - starting location of binding site in the 3'UTR sequence,
• 'End' - terminal location of binding site in the 3'UTR sequence,
• 'Length' - the length of 3'UTR of target gene,
• 'Prob.' - probability of finding a binding site motif in a 3'UTR sequence,
• 'GC content' - GC content of 22-nt binding sequence,
• 'MFE' - minimum free energy of 22-nt binding sequence and mature miRNA duplex.

Citation

Using bindseek workflow, please cite:

 Myszczynski, Szuszkiewicz, Krawczynski, Sikora, Romaniewicz,  Guzewska, Zabielski, Kaczmarek In-depth analysis of miRNA binding sites reveals the complex response of uterine epithelium to miR-26a-5p and miR-125b-5p during early pregnancy  (in press)

As well as RNAhybrid:

Rehmsmeier, Marc and Steffen, Peter and Hoechsmann, Matthias and Giegerich, Robert Fast and effective prediction of microRNA/target duplexes RNA, RNA, 2004