I have a bed file with regions of interest, and I would like to split it into a minimum number of bed files each of which will have all regions separated to each other by N bases or more. Any ideas what tool I could use?
E.g. input.bed
chr1 1000 1050
chr1 1080 1130
chr1 2000 2050
Would be split by:
split_by_distance -n 150 -i input.bed
And would produce:
input0001.bed
chr1 1000 1050
input0002.bed
chr1 1080 1130
chr1 2000 2050
Explained graphically:
Original file has 3 entries
Minimum distance:
[xxxxxxxxx]
First and second are too close:
[xxxxxxxxx]
##########
##########
##########
Output is:
file1: first
file2: second and third
Another example:
Minimum distance: [xxxxxx]
[xxxxxx] [xxxxxx] [xxxxxx] [xxxxxx]
AAAAAAAAAA BBBBBBBBBB CCCCCCCCCC DDDDDDDDD EEEEEEEEEE
Output files:
File1:
AAAAAAAAAA CCCCCCCCCC EEEEEEEEEE
File2:
BBBBBBBBBB DDDDDDDDD
Thx
•
written
by
14134125465346445
I quickly wrote something
java -jar dist/biostar178713.jar -d 100000 -o out.zip in1.bed in2.bed
[main] INFO jvarkit - sorting 1524
[main] INFO jvarkit - creating zip jeter.zip
[main] INFO jvarkit - creating bed001.bed
[main] INFO jvarkit - closing bed001.bed N=186
[main] INFO jvarkit - creating bed002.bed
[main] INFO jvarkit - closing bed002.bed N=143
[main] INFO jvarkit - creating bed003.bed
[main] INFO jvarkit - closing bed003.bed N=122
$ unzip -t out.zip
Archive: jeter.zip
testing: bed001.bed OK
testing: bed002.bed OK
testing: bed003.bed OK
No errors detected in compressed data of jeter.zip.
I think you can use bedtools spacing command:
bedtools spacing -i in.bed | awk '$4 > 150' > out1.bed
bedtools spacing -i in.bed | awk '$4 < 150' > out2.bed
Assuming here your input bed as only 3 columns.
I could not find the spacing command in bedtools or bedtools2... Any ideas?
What version of bedtools do you have? The spacing command has been included starting from release 2.23. Here's the help:
bedtools spacing
Tool: bedtools spacing
Version: v2.25.0
Summary: Report (last col.) the gap lengths between intervals in a file.
Usage: bedtools spacing [OPTIONS] -i <bed/gff/vcf/bam>
Notes:
(1) Input must be sorted by chrom,start (sort -k1,1 -k2,2n for BED).
(2) The 1st element for each chrom will have NULL distance. (".").
(3) Distance for overlapping intervaks is -1 and bookended is 0.
Example:
$ cat test.bed
chr1 0 10
chr1 10 20
chr1 21 30
chr1 35 45
chr1 100 200
$ bedtools spacing -i test.bed
chr1 0 10 .
chr1 10 20 0
chr1 21 30 1
chr1 35 45 5
chr1 100 200 55
-bed If using BAM input, write output as BED.
-header Print the header from the A file prior to results.
-nobuf Disable buffered output. Using this option will cause each line
of output to be printed as it is generated, rather than saved
in a buffer. This will make printing large output files
noticeably slower, but can be useful in conjunction with
other software tools and scripts that need to process one
line of bedtools output at a time.
-iobuf Specify amount of memory to use for input buffer.
Takes an integer argument. Optional suffixes K/M/G supported.
Note: currently has no effect with compressed files.
Log in to answer this question.
I don't think this completely solves the problem. But bedtools makewindows might be something to look into. It does not output the results into multiple bed files unfortunately, but it's a start.
So row1 is in its own file because row2 is less than 150bp away, but row3 is with row2 because its further than 150bp away?
wat.
I think he might have added the 1 in -n 150 by accident. We shouldn't assume, but it seems that he wants to be able to split his input bed file into multiple bedfiles based on the -n number. So every n bases the input would be split into a different file. I'm not sure WHY exactly.
I could understand that - but i think he actually wants to split the file so that subfile represent a "lone inverval", or a cluster of intervals. Perhaps from a peak caller. Also, not exactly sure why :P