BedFusion

A small specialised tool for sorting, merging and padding bed files

Usage: bedfusion <inputs> ... [flags]

BedFusion follows the bed file standard outlined in: Niu J., Denisko D. & Hoffman M. M. (2022): The Browser Extensible Data (BED) format

Quick-Start Guide

BedFusion will both sort (lexicographically) and merge regions by default.

Example bed file examples/merge-test.bed:

1	1	4	1	A
1	5	8	1	A
1	6	8	1	A
1	5	8	-1	A
2	5	8	1	A
1	5	8	1	B
1	20	30	1	A

> bedfusion examples/merge-test.bed
1       1       8       1,-1    A,B
1       20      30      1       A
2       5       8       1       A

Contrary to bedtools merge, BedFusion merges touching regions (like the two first lines in the example bed file). If you prefer to only merge overlapping, and not touching, regions you can use the flag --overlap=-1:

> bedfusion examples/merge-test.bed --overlap=-1
1       1       4       1       A
1       5       8       1,-1    A,B
1       20      30      1       A
2       5       8       1       A

Using several bed files as input

Several bed files can be used as input as long as they contain same number of columns. These files will be joined and then merged and sorted together.

Example bed file examples/merge-test2.bed:

2	1	4	1	A
2	5	8	1	A
2	6	8	1	A
2	5	8	-1	A
1	5	8	1	A
2	5	8	1	B
2	20	30	1	A

> bedfusion examples/merge-test.bed examples/merge-test2.bed
1       1       8       1,-1    A,B
1       20      30      1       A
2       1       8       1,-1    A,B
2       20      30      1       A

Examples

• sorting
• merging
• padding
• track files
• using a configuration file

Flags and arguments

BedFusion supports three separate ways of configuration: flags, a configuration file or environmental variables. If a combination of the three is used the reading priority order is as follows:

1. flags
2. configuration file
3. environmental variables

Order of actions ( * = can be turned on/off using flags):

1. reading files
2. padding(*)
3. merging(*)/deduplication(*)
4. sorting
5. writing output

Arguments
<inputs> ...	Bed file path(s). If more than one is provided the files will be joined as if they were one file

Flags (with format and defaults)	Environmental variables	Description
-h --help		Show context-sensitive help.
-v --version		Print version information.
-c --config-file=CONFIG-FLAG	CONFIG_FILE	The path to configuration file (must be in key-value yaml format)
-o --output=STRING	OUTPUT_FILE	Path to the output file. If unset the output will be written to stdout
-f --fasta-idx=STRING	FASTA_IDX	Tab separated file containing at least two columns where the first column contains the chromosome and the second it's size. Compatible with fasta index files, but any text file can be used as long as the file conditions are met
input
--strand-col=INT	STRAND_COL	The column containing the strand information (1-based column index). If this option is set regions on the same strand will not be merged
--feat-col=INT	FEAT_COL	The column containing the feature (e.g. gene id, transcript id etc.) information (1-based column index). If this option is set regions on the same feature will not be merged
sorting
-s --sort-type="lex"	SORT_TYPE	How the bed file should be sorted. - lex = lexicographic sorting (chr: 1 < 10 < 2 < MT < X) - nat = natural sorting (chr: 1 < 2 < 10 < MT < X) - ccs = custom chromosome sorting (see --chr-order flag ) - fidx = use ordering from fasta index file (must be used together with --fasta-idx)
--chr-order=CHR-ORDER,...	CHR_ORDER	Comma separated custom chromosome order, to be used with custom chromosome sorting (--sort-type=ccs). Chromosomes not on the list will be sorted naturally after the ones in the list
-d --deduplicate	DEDUPLICATE	Remove duplicated lines
merging
--no-merge	NO_MERGE	Do not merge regions
--overlap=0	OVERLAP	Overlap between regions to be merged. Note that touching regions are merged (e.g. if two regions are on the same chr, and the overlap is they will be merged if one ends at 5 and the other starts at 6). If you don't want touching regions to be merged set overlap to -1
padding
-p --padding=INT	PADDING	Padding in bp. Note that padding is done before merging
--padding-type="safe"	PADDING_TYPE	Padding type. - safe = bedfusion will fail if it encounters a chromosome not in the fasta index file, -lax = will only pad regions in the fasta index file and give a warning about chromosomes not in the fasta index file, - force = will pad regardless, if --fasta-idx is set there will be given a warning about the chromosomes not in the fasta index file, if --fasta-idx is not set no warnings will be given
--first-base=0	FIRST_BASE	The start coordinate of the first base on each chromosome