comparison cgatools/tools/cgatools_1.5/join.xml @ 0:182426b32995 draft default tip

Uploaded

0:182426b32995

<tool id="cg_join" name="join(beta) 1.5" version="1.0.0">
<!--
This tool creates a GUI for the join function of cgatools from Complete Genomics, Inc.
written 6-18-2012 by [e-mail]
-->

  <description>two tsv files based on equal fields or overlapping regions.</description> <!--adds description in toolbar-->

  <requirements>
  	<requirement type="binary">cgatools</requirement>
  </requirements>

  <command> <!--run executable-->
		cgatools | head -1;
		cgatools join --beta 
		--input $inputA 
		--input $inputB 
		--output $output 
		--output-mode $outmode 
		$dump 
		--select $col
		#for $m in $matches <!--get all matched columns-->
			--match ${m.match}
		#end for
		#if $range_overlap.range == 'yes'
			#for $o in $range_overlap.overlaps <!--get all matched columns-->
				--overlap ${o.overlap}
			#end for
			--overlap-mode $range_overlap.overlapmode
			--overlap-fraction-A $range_overlap.fractionA
			--boundary-uncertainty-A $range_overlap.boundaryA
			--overlap-fraction-B $range_overlap.fractionB
			--boundary-uncertainty-B $range_overlap.boundaryB
		#end if
  </command>

  <outputs>
		<data format="tabular" name="output" />
  </outputs>
  
  <inputs>
   	<!--form field to select input file A-->
    <param name="inputA" type="data" format="tabular" label="Select input file A ">
      <validator type="unspecified_build" />
			<validator type="dataset_metadata_in_file" filename="cg_crr_files.loc"
				metadata_name="dbkey" metadata_column="0"
				message="cgatools is not currently available for this build."/>
    </param>
    
  	<!--form field to select input file B-->
    <param name="inputB" type="data" format="tabular" label="Select input file B ">
      <validator type="unspecified_build" />
			<validator type="dataset_metadata_in_file" filename="cg_crr_files.loc"
				metadata_name="dbkey" metadata_column="0"
				message="cgatools is not currently available for this build."/>
    </param>
    
  	<!--form field to specify columns to print-->
    <param name="col" type="text" value="A.*,B.*" size="40" label="Specify columns for output" help="The default value A.*,B.* prints all columns from both files, other selections enter in the format A.col_name1,A.col_name3,B.col_name1" />

  	<!--form field to select output-mode-->
		<param name="outmode" type="select" label="Select output mode">
			<option value="full" selected="true">full (1 line for each match of records in A and B)</option>
			<option value="compact">compact (1 line for each record in A, joining multiple records in B by semicolon)</option>
			<option value="compact-pct">compact-pct (same as compact, annotated with % overlap)</option>
		</param>

		<!--form field to select columns to match-->
		<param name="dump" type="select" label="Select records to print">
			<option value="--always-dump" selected="true">print all records of A even if not matched in B</option>
			<option value="">print only records of A that are matched in B</option>
		</param>

  	<!--form field to specify columns to match-->
    <repeat name="matches" title="Exact match column">
      <param name="match" type="text" size="40" label="Enter column:column" help="Enter column_from_A:column_from_B, e.g. chromosome:chromosome"/>
    </repeat>
    
    <conditional name="range_overlap">
    	<param name="range" type="select" label="Do you want to match columns by overlapping range?">
    		<option value="no">no</option>
    		<option value="yes">yes</option>
    	</param>
    	
    	<when value="yes">
				<!--form field to specify columns to overlap-->
				<repeat name="overlaps" title="Range column">
					<param name="overlap" type="text" size="40" label="Enter column&#91;,column&#93;:column&#91;,column&#93;" help="Enter range_start_from_A&#91;,range_stop_from_A&#93;:range_start_from_B&#91;,range_stop_from_B&#93;, e.g. begin,end:begin,end (overlapping range of positions) or begin,end:position"/>
				</repeat>

				<!--form field to select overlap-mode-->
				<param name="overlapmode" type="select" label="Select overlap mode">
					<option value="strict" selected="true">strict (overlap if A.begin&lt;B.end and B.begin&gt;A.end)</option>
					<option value="allow-abutting-points">allow-abutting-points (overlap if A.begin&lt;B.end and B.begin&gt;A.end, or if A.begin&lt;=B.end and B.begin&lt;=A.end and either A or B has zero length.)</option>
				</param>

				<!--form fields to overlap options-->
				<param name="fractionA" type="integer" value="0" label="Minimum fraction of A region overlap " />
				<param name="boundaryA" type="integer" value="0" label="Boundary uncertainty for A for overlap filtering " help="Records failing the following boundary-uncertainty calculation are not included in the output: overlap length >= overlap-fraction-A * (A-range-length - boundary-uncertainty-A)"/>
				
				<param name="fractionB" type="integer" value="0" label="Minimum fraction of B region overlap " />
				<param name="boundaryB" type="integer" value="0" label="Boundary uncertainty for overlap filtering "  help="Records failing the following boundary-uncertainty calculation are not included in the output: overlap length >= overlap-fraction-B * (B-range-length - boundary-uncertainty-B)"/>
    	</when>
		</conditional>
  </inputs>
  
  <help>
  
**What it does**

This tool joins two tab-delimited files based on equal fields or overlapping regions.

**cgatools 1.5.0 Documentation**

Userguide: http://cgatools.sourceforge.net/docs/1.5.0/cgatools-user-guide.pdf

Release notes: http://cgatools.sourceforge.net/docs/1.5.0/cgatools-release-notes.pdf

**Command line reference**::

		COMMAND NAME
		  join - Joins two tab-delimited files based on equal fields or overlapping regions.
		
		DESCRIPTION
		  Joins two tab-delimited files based on equal fields or overlapping regions.
		  By default, an output record is produced for each match found between file 
		  A and file B, but output format can be controlled by the --output-mode 
		  parameter.
		
		OPTIONS
		  -h [ --help ] 
		      Print this help message.
		
		  --beta 
		      This is a beta command. To run this command, you must pass the --beta 
		      flag.
		
		  --input arg
		      File name to use as input (may be passed in as arguments at the end of 
		      the command), or omitted for stdin). There must be exactly two input 
		      files to join. If only one file is specified by name, file A is taken 
		      to be stdin and file B is the named file. File B is read fully into 
		      memory, and file A is streamed. File A's columns appear first in the 
		      output.
		
		  --output arg (=STDOUT)
		      The output file name (may be omitted for stdout).
		
		  --match arg
		      A match specification, which is a column from A and a column from B 
		      separated by a colon.
		
		  --overlap arg
		      Overlap specification. An overlap specification consists of a range 
		      definition for files A and B, separated by a colon. A range definition 
		      may be two columns, in which case they are interpreted as the beginning
		      and end of the range. Or it may be one column, in which case the range 
		      is defined as the 1-base range starting at the given value. The records
		      from the two files must overlap in order to be considered for output. 
		      Two ranges are considered to overlap if the overlap is at least one 
		      base long, or if one of the ranges is length 0 and the ranges overlap 
		      or abut. For example, "begin,end:offset" will match wherever end-begin 
		      &gt; 0, begin&lt;offset+1, and end&gt;offset, or wherever end-begin = 0, 
		      begin&lt;=offset+1, and end&gt;=offset.


		  -m [ --output-mode ] arg (=full)
		      Output mode, one of the following:
		        full        Print an output record for each match found between 
		                    file A and file B.
		        compact     Print at most one record for each record of file A, 
		                    joining the file B values by a semicolon and 
		                    suppressing repeated B values and empty B values.
		        compact-pct Same as compact, but for each distinct B value, 
		                    annotate with the percentage of the A record that is 
		                    overlapped by B records with that B value. Percentage 
		                    is rounded up to nearest integer.
		
		  --overlap-mode arg (=strict)
		      Overlap mode, one of the following:
		        strict                Range A and B overlap if A.begin &lt; B.end and 
		                              B.begin &lt; A.end.
		        allow-abutting-points Range A and B overlap they meet the strict 
		                              requirements, or if A.begin &lt;= B.end and 
		                              B.begin &lt;= A.end and either A or B has zero 
		                              length.

		  --select arg (=A.*,B.*)
		      Set of fields to select for output.
		
		  -a [ --always-dump ] 
		      Dump every record of A, even if there are no matches with file B.
		
		  --overlap-fraction-A arg (=0)
		      Minimum fraction of A region overlap for filtering output.
		
		  --boundary-uncertainty-A arg (=0)
		      Boundary uncertainty for overlap filtering. Specifically, records 
		      failing the following predicate are filtered away: overlap &gt;= 
		      overlap-fraction-A * ( A-range-length - boundary-uncertainty-A )
		
		  --overlap-fraction-B arg (=0)
		      Minimum fraction of B region overlap for filtering output.
		
		  --boundary-uncertainty-B arg (=0)
		      Boundary uncertainty for overlap filtering. Specifically, records 
		      failing the following predicate are filtered away: overlap &gt;= 
		      overlap-fraction-B * ( B-range-length - boundary-uncertainty-B )

		SUPPORTED FORMAT_VERSION
		  Any
  </help>
</tool>