This is a test version of Biostars. For the public version, visit https://www.biostars.org.
How to merge several BED files to create a comprehensive overview file

I have a few .bed files from different companies offering exome sequencing kits.

I would like to have a file that summarizes all target regions for all these kits. The .bed file have a basic structure composed for three columns (chr#, Start, End). I would like to get an output table that shows which genomic regions are covered only by one of these kits, and which regions are covered by more than one (and which ones). The best way to illustrate this is by an example:

Kit 1

chr# Start End
1 100 300

Kit 2

chr# Start End
1 150 350

Kit 3

chr# Start End
1 80 200

I would like to merge and intersect the files for an output that divides the regions into subregions based on overlap between the input files. It should looks something like this:

chr# Start End Kit 1 Kit 2 Kit 3
1 80 100 0 0 1
1 100 150 1 0 1
1 150 200 1 1 1
1 200 300 1 1 0
1 300 350 0 1 0

I would prefer to do this in python, but I could try in R as well. I have created a #pandas dataframe containing ranges from all kits, ordered by 'chr#' and 'Start' coordinates, which looks like this:

pandas dataframe containing ranges from all kits, ordered by chr# and Start coordinates

Any help would be appreciated.

pandas granges python bioconductor
• 2,160 views • link
updated by Jorge Amigo Jorge Amigo • written by Jordi • Jordi

2 answers

use bedtools multiinter

Summary: Identifies common intervals among multiple
     BED/GFF/VCF files.

Usage:   bedtools multiinter [OPTIONS] -i FILE1 FILE2 .. FILEn
     Requires that each interval file is sorted by chrom/start.
• 0 views • link
written by Pierre Lindenbaum Pierre Lindenbaum

When comparing regions in bed format, bedtools or bedops are usually the best way to go.

If the kits you're comparing are exome sequencing bedfiles, I would suggest to add RefSeq's exons for instance to the comparison so that those overlapping intervals have some meaning at the end.

• 0 views • link
written by Jorge Amigo Jorge Amigo

Log in to answer this question.

More posts like this