Equivalent Junctions

This repository contains the python scripts for extarcting equivalent junctions (the set of neighboring junctions that result in the same junction sequence) in a reference transcriptome. The codes are general and can be used for any organism and genome assembly.

Software

• Python 3.6.1 with Biopython 1.71 installed

Pyhton scripts

getJunctionsFromGTF.py:

• determines equivalent junctions from a genome fasta file and an annotation file in the GTF format
• usage:

  python getJunctionsFromGTF.py -f genome_file.fa
                                -a annotation_file.gtf
                                -o output_file.txt
  

• output file:
  • a tab delimited file with an extracted equivalent junction sequence for each annotated exon-exon boundary:

      equiv_junc_sequence    gene_name    chromosome    strand    donor_exon_coordinate    acceptor_exon_coordinate 

• example equivalent junction in the output file:

                                 AG   LEPR   chr1    +    65420740    65425302  

getJunctionsFromTxt.py:

• determines equivalent junction sequences from a genome fasta file and annotatiaon file in the BED format containing chr, donor, and acceptor coordinates.
• usage:

  python getJunctionsFromTxt.py -f genome_file.fa
                                -t annotation_file.txt
                                -o output_file.txt
                                -c choromosome_column (defult=0) 
                                -d donor_coordinate_column (default=1) 
                                -a acceptor_coordinate_column (default=2)
                                -s strand_column (default=3)

• output file:
  • a tab delimited file with the extracted equivalent junction sequence for each annotated exon-exon boundary:

      equiv_junc_sequence    gene_name    chromosome    strand    donor_exon_coordinate    acceptor_exon_coordinate 

Note: Example genome fasta and annotation files that have been analyzed for equivalent junctions are provided in equiv_junc_genomes.

Computing the number of equivalent junctions:

After running the python script and obtaining the output_file containing the equivalent junction sequence for each exon-exon junction, the following command gives the total number of each equivalent junction sequence:

cut -f1,3,4,5,6 output_file.txt | sort -u |cut -f1 | sort |uniq -c| sort -k1nr > output_file_equivSeqCounts.txt 

Citation

Dehghannasiri, R., Szabo, L., and Salzman, J., Ambiguous splice sites distinguish circRNA and linear splicing in the human genome, Bioinformatics, April 2019, (https://academic.oup.com/bioinformatics/article/35/8/1263/5091181)

Contact

In case of questions, please contact Roozbeh Dehghannasiri ([email protected]).