# HG changeset patch -- Bitbucket.org # Project galaxy-dist # URL http://bitbucket.org/galaxy/galaxy-dist/overview # User jeremy goecks <jeremy.goecks@emory.edu> # Date 1277306218 14400 # Node ID 752cb3e325374381bdd0db6b9163b12ba88b7151 # Parent 03eb69c6e92ea5472b7627bd9990327f7546a8a7 Improvements to GOPS intersect: (a) preserve metadata for interval inputs; (b) allow arbitrary mix of interval and GFF inputs; and (c) functional tests updated to test new functionality. --- a/tools/new_operations/intersect.xml +++ b/tools/new_operations/intersect.xml @@ -1,131 +1,110 @@ <tool id="gops_intersect_1" name="Intersect"><description>the intervals of two queries</description><command interpreter="python">gops_intersect.py - #if $inputs.type == "Interval": - $inputs.interval_input1 $inputs.interval_input2 $output - -1 ${inputs.interval_input1.metadata.chromCol},${inputs.interval_input1.metadata.startCol},${inputs.interval_input1.metadata.endCol},${inputs.interval_input1.metadata.strandCol} - -2 ${inputs.interval_input2.metadata.chromCol},${inputs.interval_input2.metadata.startCol},${inputs.interval_input2.metadata.endCol},${inputs.interval_input2.metadata.strandCol} - $inputs.interval_returntype - #else - $inputs.gff_input1 $inputs.gff_input2 $output - ## TODO: can we use metadata like above to set these columns rather than hardcode them? - -1 1,4,5,7 - -2 1,4,5,7 - --gff - $inputs.gff_returntype + $input1 $input2 $output + + #if isinstance( $input1.datatype, $__app__.datatypes_registry.get_datatype_by_extension('gff').__class__): + -1 1,4,5,7 --gff1 + #else: + -1 ${input1.metadata.chromCol},${input1.metadata.startCol},${input1.metadata.endCol},${input1.metadata.strandCol} #end if - -m $min + + #if isinstance( $input2.datatype, $__app__.datatypes_registry.get_datatype_by_extension('gff').__class__): + -2 1,4,5,7 --gff2 + #else: + -2 ${input2.metadata.chromCol},${input2.metadata.startCol},${input2.metadata.endCol},${input2.metadata.strandCol} + #end if + + -m $min $returntype </command><inputs> - <conditional name="inputs"> - <param name="type" type="select" label="File Format to Use"> - <option value="Interval">Interval</option> - <option value="GFF">GFF</option> - </param> - <when value="Interval"> - <param name="interval_returntype" type="select" label="Return" help="(see figure below)"> - <option value="">Overlapping Intervals</option> - <option value="-p">Overlapping pieces of Intervals</option> - </param> - <param format="interval" name="interval_input1" type="data" help="First query"> - <label>of</label> - </param> - <param format="interval" name="interval_input2" type="data" help="Second query"> - <label>that intersect</label> - </param> - </when> - <when value="GFF"> - <param name="gff_returntype" type="select" label="Return" help="(see figure below)"> - <option value="">Overlapping Intervals</option> - <option value="-p">Overlapping pieces of Intervals</option> - </param> - <param format="gff" name="gff_input1" type="data" help="First query"> - <label>of</label> - </param> - <param format="gff" name="gff_input2" type="data" help="Second query"> - <label>that intersect</label> - </param> - </when> - </conditional> + <param name="returntype" type="select" label="Return" help="(see figure below)"> + <option value="">Overlapping Intervals</option> + <option value="-p">Overlapping pieces of Intervals</option> + </param> + <param format="interval,gff" name="input1" type="data" help="First query"> + <label>of</label> + </param> + <param format="interval,gff" name="input2" type="data" help="Second query"> + <label>that intersect</label> + </param><param name="min" size="4" type="integer" value="1" help="(bp)"><label>for at least</label></param></inputs><outputs> - <data format="input" name="output"> - #if inputs.type == "Interval": - metadata_source="inputs.interval_input1" - #else: - metadata_source="inputs.gff_input1" - #end if - </data> + <data format="input" name="output" metadata_source="input1"/></outputs><code file="operation_filter.py"/><tests><test><param name="type" value="Interval"/> - <param name="interval_input1" value="1.bed" /> - <param name="interval_input2" value="2.bed" /> + <param name="input1" value="1.bed" /> + <param name="input2" value="2.bed" /><param name="min" value="1" /> - <param name="interval_returntype" value="" /> + <param name="returntype" value="" /><output name="output" file="gops_intersect_out.bed" /></test><test><param name="type" value="Interval"/> - <param name="interval_input1" value="1.bed" /> - <param name="interval_input2" value="2_mod.bed" ftype="interval"/> + <param name="input1" value="1.bed" /> + <param name="input2" value="2_mod.bed" ftype="interval"/><param name="min" value="1" /> - <param name="interval_returntype" value="" /> + <param name="returntype" value="" /><output name="output" file="gops_intersect_diffCols.bed" /></test><test> - <param name="type" value="Interval"/> - <param name="interval_input1" value="1.bed" /> - <param name="interval_input2" value="2_mod.bed" ftype="interval"/> + <param name="input1" value="1.bed" /> + <param name="input2" value="2_mod.bed" ftype="interval"/><param name="min" value="1" /> - <param name="interval_returntype" value="Overlapping pieces of Intervals" /> + <param name="returntype" value="Overlapping pieces of Intervals" /><output name="output" file="gops_intersect_p_diffCols.bed" /></test><test> - <param name="type" value="Interval"/> - <param name="interval_input1" value="1.bed" /> - <param name="interval_input2" value="2.bed" /> + <param name="input1" value="1.bed" /> + <param name="input2" value="2.bed" /><param name="min" value="10" /> - <param name="interval_returntype" value="Overlapping pieces of Intervals" /> + <param name="returntype" value="Overlapping pieces of Intervals" /><output name="output" file="gops_intersect_p_out.bed" /></test><test> - <param name="type" value="Interval"/> - <param name="interval_input1" value="gops_bigint.interval" ftype="interval" /> - <param name="interval_input2" value="gops_bigint2.interval" ftype="interval" /> + <param name="input1" value="gops_bigint.interval" ftype="interval" /> + <param name="input2" value="gops_bigint2.interval" ftype="interval" /><param name="min" value="1" /> - <param name="interval_returntype" value="" /> + <param name="returntype" value="" /><output name="output" file="gops_intersect_bigint_out.interval" /></test><test> - <param name="type" value="Interval"/> - <param name="interval_input1" value="gops_bigint2.interval" ftype="interval" /> - <param name="interval_input2" value="gops_bigint.interval" ftype="interval" /> + <param name="input1" value="gops_bigint2.interval" ftype="interval" /> + <param name="input2" value="gops_bigint.interval" ftype="interval" /><param name="min" value="1" /> - <param name="interval_returntype" value="" /> + <param name="returntype" value="" /><output name="output" file="gops_intersect_bigint_out.interval" /></test><test> - <param name="type" value="Interval"/> - <param name="interval_input1" value="12.bed" ftype="bed" /> - <param name="interval_input2" value="1.bed" ftype="bed" /> + <param name="input1" value="12.bed" ftype="bed" /> + <param name="input2" value="1.bed" ftype="bed" /><param name="min" value="1" /> - <param name="interval_returntype" value="" /> + <param name="returntype" value="" /><output name="output" file="gops_intersect_no_strand_out.bed" /></test> + <!-- Intersect two GFF files. --><test> - <param name="type" value="GFF"/> - <param name="gff_input1" value="gops_subtract_in1.gff" /> - <param name="gff_input2" value="gops_subtract_in2.gff" /> + <param name="input1" value="gops_subtract_in1.gff" /> + <param name="input2" value="gops_subtract_in2.gff" /><param name="min" value="1" /> - <param name="gff_returntype" value="" /> + <param name="returntype" value="" /><output name="output" file="gops_intersect_out2.gff" /></test> + <!-- Intersect GFF file and bed file. --> + <test> + <param name="input1" value="gops_subtract_in1.gff" /> + <param name="input2" value="gops_subtract_in2.bed" /> + <param name="min" value="1" /> + <param name="returntype" value="" /> + <output name="output" file="gops_intersect_out2.gff" /> + </test> + </tests><help> --- a/tools/new_operations/subtract.xml +++ b/tools/new_operations/subtract.xml @@ -3,7 +3,6 @@ <command interpreter="python">gops_subtract.py $input1 $input2 $output - ##if $input1.ext in ['gff','gtf','gff3']: #if isinstance( $input1.datatype, $__app__.datatypes_registry.get_datatype_by_extension('gff').__class__): -1 1,4,5,7 --gff1 #else: --- a/tools/new_operations/gops_intersect.py +++ b/tools/new_operations/gops_intersect.py @@ -1,13 +1,14 @@ #!/usr/bin/env python """ -Find regions of first bed file that overlap regions in a second bed file +Find regions of first interval/GFF file that overlap regions in a second interval/GFF file usage: %prog bed_file_1 bed_file_2 out_file -1, --cols1=N,N,N,N: Columns for start, end, strand in first file -2, --cols2=N,N,N,N: Columns for start, end, strand in second file -m, --mincols=N: Require this much overlap (default 1bp) -p, --pieces: just print pieces of second set (after padding) - -G, --gff: inputs are GFF format, meaning start and end coordinates are 1-based, closed interval + -G, --gff1: input 1 is GFF format, meaning start and end coordinates are 1-based, closed interval + -H, --gff2: input 2 is GFF format, meaning start and end coordinates are 1-based, closed interval """ from galaxy import eggs import pkg_resources @@ -34,24 +35,29 @@ def main(): chr_col_2, start_col_2, end_col_2, strand_col_2 = parse_cols_arg( options.cols2 ) if options.mincols: mincols = int( options.mincols ) pieces = bool( options.pieces ) - gff_format = bool( options.gff ) + in1_gff_format = bool( options.gff1 ) + in2_gff_format = bool( options.gff2 ) in_fname, in2_fname, out_fname = args except: doc_optparse.exception() - # Set reader to handle either GFF or default format. - if gff_format: - reader_wrapper = GFFReaderWrapper + # Set readers to handle either GFF or default format. + if in1_gff_format: + in1_reader_wrapper = GFFReaderWrapper else: - reader_wrapper = NiceReaderWrapper + in1_reader_wrapper = NiceReaderWrapper + if in2_gff_format: + in2_reader_wrapper = GFFReaderWrapper + else: + in2_reader_wrapper = NiceReaderWrapper - g1 = reader_wrapper( fileinput.FileInput( in_fname ), + g1 = in1_reader_wrapper( fileinput.FileInput( in_fname ), chrom_col=chr_col_1, start_col=start_col_1, end_col=end_col_1, strand_col=strand_col_1, fix_strand=True ) - g2 = reader_wrapper( fileinput.FileInput( in2_fname ), + g2 = in2_reader_wrapper( fileinput.FileInput( in2_fname ), chrom_col=chr_col_2, start_col=start_col_2, end_col=end_col_2, @@ -63,7 +69,7 @@ def main(): try: for line in intersect( [g1,g2], pieces=pieces, mincols=mincols ): if type( line ) == GenomicInterval: - if gff_format: + if in1_gff_format: line = convert_to_gff_coordinates( line ) out_file.write( "%s\n" % "\t".join( line.fields ) ) else: