create windows10Kb.bed

[Sam64]

Hi,

I try to find a way to create a bedfile in order to compute the coverage of aligned sequences on 10 kilobase “windows” spanning the genome.

chr1 0 10000
chr1 10000 20000
..... .... ....

Go to http://code.google.com/p/bedtools/wi...ge#coverageBed and clik "coveragebed"

How create this 10Kb windows file please ? UCSC ? Table browser ?

Thank you,

Sam

[Sam64]

up up up

[dphansti]

I would like to know this answer too. I have always done this by writing a little script but it would be great if it could be done with a bedtools command. Basically, you just need the length of each chromosome which I get from one of the files generated by bwa index (e.g. hg19.dict I think).

Then I do something like this.

grep -P "^@SQ" hg19.dict | awk 'BEGIN {OFS = "\t";} {split($2,chrom,":");split($3,ln,":"); i =10000; j=0; while (i < ln[2]){print chrom[2],i-10000,i; j=i; i += 10000;}; print chrom[2],j,ln[2];}' > hg19.10kbwindows

I haven't double checked to see if this works correctly but if it doesn't something like this should do the trick.

[quinlana]

Have a look at the "bedtools makewindows" command in bedtools v2.15.0.

Examples:

 # Divide the human genome into windows of 1MB:
 $ bedtools makewindows -g hg19.txt -w 1000000
 chr1 0 1000000
 chr1 1000000 2000000
 chr1 2000000 3000000
 chr1 3000000 4000000
 chr1 4000000 5000000
 ...

 # Divide the human genome into sliding (=overlapping) windows of 1MB, with 500KB overlap:
 $ bedtools makewindows -g hg19.txt -w 1000000 -s 500000
 chr1 0 1000000
 chr1 500000 1500000
 chr1 1000000 2000000
 chr1 1500000 2500000
 chr1 2000000 3000000
 ...

 # Divide each chromosome in human genome to 1000 windows of equal size:
 $ bedtools makewindows -g hg19.txt -n 1000
 chr1 0 249251
 chr1 249251 498502
 chr1 498502 747753
 chr1 747753 997004
 chr1 997004 1246255
 ...

 # Divide each interval in the given BED file into 10 equal-sized windows:
 $ cat input.bed
 chr5 60000 70000
 chr5 73000 90000
 chr5 100000 101000
 $ bedtools makewindows -b input.bed -n 10
 chr5 60000 61000
 chr5 61000 62000
 chr5 62000 63000
 chr5 63000 64000
 chr5 64000 65000
 ...

 # Add a name column, based on the window number: 
 $ cat input.bed
 chr5  60000  70000 AAA
 chr5  73000  90000 BBB
 chr5 100000 101000 CCC
 $ bedtools makewindows -b input.bed -n 3 -i winnum
 chr5        60000   63334   1
 chr5        63334   66668   2
 chr5        66668   70000   3
 chr5        73000   78667   1
 chr5        78667   84334   2
 chr5        84334   90000   3
 chr5        100000  100334  1
 chr5        100334  100668  2
 chr5        100668  101000  3
 ...

 # Add a name column, based on the source ID + window number: 
 $ cat input.bed
 chr5  60000  70000 AAA
 chr5  73000  90000 BBB
 chr5 100000 101000 CCC
 $ bedtools makewindows -b input.bed -n 3 -i srcwinnum
 chr5        60000   63334   AAA_1
 chr5        63334   66668   AAA_2
 chr5        66668   70000   AAA_3
 chr5        73000   78667   BBB_1
 chr5        78667   84334   BBB_2
 chr5        84334   90000   BBB_3
 chr5        100000  100334  CCC_1
 chr5        100334  100668  CCC_2
 chr5        100668  101000  CCC_3
 ...

[dphansti]

Excellent. Thanks. And thanks for providing the version number (it's not part of BEDTools 2.14.3). Time to upgrade.

[robert-nci]

Bedtool makewindows

Hello there,

Part of this example doesn't work for me.
Basically, makewindows does not recognize any of the -b and -n options.

BWT, I checked the version of my bedtools and it is 2.15.0.

Thanks,
Robert

Have a look at the "bedtools makewindows" command in bedtools v2.15.0.

Examples:

 # Divide the human genome into windows of 1MB:
 $ bedtools makewindows -g hg19.txt -w 1000000
 chr1 0 1000000
 chr1 1000000 2000000
 chr1 2000000 3000000
 chr1 3000000 4000000
 chr1 4000000 5000000
 ...

 # Divide the human genome into sliding (=overlapping) windows of 1MB, with 500KB overlap:
 $ bedtools makewindows -g hg19.txt -w 1000000 -s 500000
 chr1 0 1000000
 chr1 500000 1500000
 chr1 1000000 2000000
 chr1 1500000 2500000
 chr1 2000000 3000000
 ...

 # Divide each chromosome in human genome to 1000 windows of equal size:
 $ bedtools makewindows -g hg19.txt -n 1000
 chr1 0 249251
 chr1 249251 498502
 chr1 498502 747753
 chr1 747753 997004
 chr1 997004 1246255
 ...

 # Divide each interval in the given BED file into 10 equal-sized windows:
 $ cat input.bed
 chr5 60000 70000
 chr5 73000 90000
 chr5 100000 101000
 $ bedtools makewindows -b input.bed -n 10
 chr5 60000 61000
 chr5 61000 62000
 chr5 62000 63000
 chr5 63000 64000
 chr5 64000 65000
 ...

 # Add a name column, based on the window number: 
 $ cat input.bed
 chr5  60000  70000 AAA
 chr5  73000  90000 BBB
 chr5 100000 101000 CCC
 $ bedtools makewindows -b input.bed -n 3 -i winnum
 chr5        60000   63334   1
 chr5        63334   66668   2
 chr5        66668   70000   3
 chr5        73000   78667   1
 chr5        78667   84334   2
 chr5        84334   90000   3
 chr5        100000  100334  1
 chr5        100334  100668  2
 chr5        100668  101000  3
 ...

 # Add a name column, based on the source ID + window number: 
 $ cat input.bed
 chr5  60000  70000 AAA
 chr5  73000  90000 BBB
 chr5 100000 101000 CCC
 $ bedtools makewindows -b input.bed -n 3 -i srcwinnum
 chr5        60000   63334   AAA_1
 chr5        63334   66668   AAA_2
 chr5        66668   70000   AAA_3
 chr5        73000   78667   BBB_1
 chr5        78667   84334   BBB_2
 chr5        84334   90000   BBB_3
 chr5        100000  100334  CCC_1
 chr5        100334  100668  CCC_2
 chr5        100668  101000  CCC_3
 ...