[2] | 1 | #!/usr/bin/env python |
---|
| 2 | """ |
---|
| 3 | Join two sets of intervals using their overlap as the key. |
---|
| 4 | |
---|
| 5 | usage: %prog bed_file_1 bed_file_2 out_file |
---|
| 6 | -1, --cols1=N,N,N,N: Columns for start, end, strand in first file |
---|
| 7 | -2, --cols2=N,N,N,N: Columns for start, end, strand in second file |
---|
| 8 | -m, --mincols=N: Require this much overlap (default 1bp) |
---|
| 9 | -f, --fill=N: none, right, left, both |
---|
| 10 | """ |
---|
| 11 | from galaxy import eggs |
---|
| 12 | import pkg_resources |
---|
| 13 | pkg_resources.require( "bx-python" ) |
---|
| 14 | import sys, traceback, fileinput |
---|
| 15 | from warnings import warn |
---|
| 16 | from bx.intervals import * |
---|
| 17 | from bx.intervals.io import * |
---|
| 18 | from bx.intervals.operations.join import * |
---|
| 19 | from bx.cookbook import doc_optparse |
---|
| 20 | from galaxy.tools.util.galaxyops import * |
---|
| 21 | |
---|
| 22 | assert sys.version_info[:2] >= ( 2, 4 ) |
---|
| 23 | |
---|
| 24 | def main(): |
---|
| 25 | mincols = 1 |
---|
| 26 | upstream_pad = 0 |
---|
| 27 | downstream_pad = 0 |
---|
| 28 | leftfill = False |
---|
| 29 | rightfill = False |
---|
| 30 | |
---|
| 31 | options, args = doc_optparse.parse( __doc__ ) |
---|
| 32 | try: |
---|
| 33 | chr_col_1, start_col_1, end_col_1, strand_col_1 = parse_cols_arg( options.cols1 ) |
---|
| 34 | chr_col_2, start_col_2, end_col_2, strand_col_2 = parse_cols_arg( options.cols2 ) |
---|
| 35 | if options.mincols: mincols = int( options.mincols ) |
---|
| 36 | if options.fill: |
---|
| 37 | if options.fill == "both": |
---|
| 38 | rightfill = leftfill = True |
---|
| 39 | else: |
---|
| 40 | rightfill = options.fill == "right" |
---|
| 41 | leftfill = options.fill == "left" |
---|
| 42 | in_fname, in2_fname, out_fname = args |
---|
| 43 | except: |
---|
| 44 | doc_optparse.exception() |
---|
| 45 | |
---|
| 46 | g1 = NiceReaderWrapper( fileinput.FileInput( in_fname ), |
---|
| 47 | chrom_col=chr_col_1, |
---|
| 48 | start_col=start_col_1, |
---|
| 49 | end_col=end_col_1, |
---|
| 50 | strand_col=strand_col_1, |
---|
| 51 | fix_strand=True ) |
---|
| 52 | g2 = NiceReaderWrapper( fileinput.FileInput( in2_fname ), |
---|
| 53 | chrom_col=chr_col_2, |
---|
| 54 | start_col=start_col_2, |
---|
| 55 | end_col=end_col_2, |
---|
| 56 | strand_col=strand_col_2, |
---|
| 57 | fix_strand=True ) |
---|
| 58 | |
---|
| 59 | out_file = open( out_fname, "w" ) |
---|
| 60 | |
---|
| 61 | try: |
---|
| 62 | for outfields in join(g1, g2, mincols=mincols, rightfill=rightfill, leftfill=leftfill): |
---|
| 63 | if type( outfields ) is list: |
---|
| 64 | out_file.write( "%s\n" % "\t".join( outfields ) ) |
---|
| 65 | else: |
---|
| 66 | out_file.write( "%s\n" % outfields ) |
---|
| 67 | except ParseError, exc: |
---|
| 68 | out_file.close() |
---|
| 69 | fail( "Invalid file format: %s" % str( exc ) ) |
---|
| 70 | except MemoryError: |
---|
| 71 | out_file.close() |
---|
| 72 | fail( "Input datasets were too large to complete the join operation." ) |
---|
| 73 | |
---|
| 74 | out_file.close() |
---|
| 75 | |
---|
| 76 | if g1.skipped > 0: |
---|
| 77 | print skipped( g1, filedesc=" of 1st dataset" ) |
---|
| 78 | if g2.skipped > 0: |
---|
| 79 | print skipped( g2, filedesc=" of 2nd dataset" ) |
---|
| 80 | |
---|
| 81 | if __name__ == "__main__": |
---|
| 82 | main() |
---|