Cluster Intervals By Gene Id In A Bed File

Question

[Rad]

Hello everyone,

I have a Bed file with 5 columns where the 4th is a unique ID and the 5th is a geneID

I was trying to play with bedtools to cluster this bed file by gene ID and output a single line for each gene with the range (chr start end) of the region. Basically I want to cluster intervals. Example

chr1 10  1000 ID1 GeneID1
chr1 20  1300 ID2 GeneID1
chr1 1400  1600 ID3 GeneID1

I'm trying to get an output like

chr1 10 1600 GeneID1

Can anyone tell me if playing with bedtools is the best way of doing this or is it possible just by awk ? any idea ?

Thank you

Answer 1

[Pierre Lindenbaum]

using awk and sqlite:

~$ echo -e "chr1\t10\t\t1000\tID1\tGeneID1\nchr1\t20\t\t1300\tID2\tGeneID1\nchr1\t1400\t\t1600\tID3\tGeneID1" |\
awk 'BEGIN{printf("create table t(chrom text,start int,end int, name text);\n");} {printf("insert into t(chrom,start,end,name) values(\"%s\",%s,%s,\"%s\");\n",$1,$2,$3,$5);} END {printf("select chrom,min(start),max(end),name from t group by chrom,name;\n");}' |\
sqlite3 tmp.sqlite

chr1|10|1600|GeneID1

[Rad]

Awesome, I added "drop table if exists t;" before table creation so that we can use it within a script Thx man !

[Pierre Lindenbaum]

you can also add some indexes on chrom and name if you have a large input...

[GouthamAtla]

I just could not believe this awesomeness.