intersection of two files

Question

[a.rex]

Is there an efficient way of joining two files in the following way - i.e. using BedTools?

For example, FILE 1:

scaffold1          0        206        transcript_loc.00001      exon    
scaffold1         262      749       transcript_loc.00001      exon    
scaffold1         1391    1549     transcript_loc.00001      exon

FILE2:

scaffold1        517     540     Simple_repeat   
scaffold1        1063    1162    LTR/Gypsy        
scaffold1        1400     1498   LTR

Resultant file:

 scaffold1          0        206        transcript_loc.00001      exon    
 scaffold1         262      749       transcript_loc.00001      exon          517     540     Simple_repeat   
 scaffold1         1391    1549     transcript_loc.00001      exon          1400     1498   LTR

In this way, wherever the two files intersect, the intersection is appended to a new column on the first file. The non-intersect is discarded.

Many thanks.

[venu]

This can be done with intersectBed option from BedTools. Check wa, wb, wo options. Before using bedtools change your file formats to actual BED format.

Answer 1

[Alex Reynolds]

Sure, you can use BEDOPS bedmap to map overlaps in one file to elements in another file:

$ bedmap --echo --echo-map --delim '\t' file1.bed file2.bed > answer.bed

So long as your files are sorted, they are BED and can be used as-is. If not:

$ sort-bed file1.unsorted.bed > file1.bed
$ sort-bed file2.unsorted.bed > file2.bed

Then use bedmap, as described.

If you want to discard rows without overlaps between the first and second inputs, add the --skip-unmapped option:

$ bedmap --echo --echo-map --delim '\t' --skip-unmapped file1.bed file2.bed > answer.bed