diff rankfilterGCMS_tabular.xml @ 0:9d5f4f5f764b

Initial commit to toolshed

--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/rankfilterGCMS_tabular.xml	Thu Jan 16 13:10:00 2014 +0100
@@ -0,0 +1,77 @@
+<tool id="rankfilterGCMS_tabular" name="RIQC-RankFilter GC-MS from tabular file" version="1.0.2">
+  <description>Convert Retention Time to Retention Index</description>
+  <command interpreter="python">rankfilter_GCMS/rankfilter.py $input_file</command>
+  <inputs>
+    <param format="tabular" name="sample" type="data" label="Sample File" 
+	       help="Converted PDF file in tabular format" />
+	<!-- question: is this calibration file not column specific as it includes RT info?? -->
+    <param name="calibration"  type="select" label="Calibration File" 
+           help="Calibration file with reference masses (e.g. alkanes) with their RT and RI values"
+    		dynamic_options='get_directory_files("tool-data/shared/PRIMS-metabolomics/RankFilter_Calibration_Files")'/>
+
+    <param name="analysis_type" type="select" format="text" label="Analysis Type"
+    	   help="Select the type of analysis that has been used to generate the sample file">
+      <option value="NIST">NIST</option>
+      <option value="AMDIS">AMDIS</option>
+    </param>
+    <param name="model" type="select" format="text" label="Select a model to be used "
+    	   help="Both linear and (3rd degree) polynomial models are available ">
+      <option value="linear">Linear</option>
+      <option value="poly">Polynomial</option>
+    </param>
+    <param name="lib_data" type="select" label="Library" 
+	       help="Reference global lookup library file with CAS numbers and respective (previously calculated) RIsvr values" 
+           dynamic_options='get_directory_files("tool-data/shared/PRIMS-metabolomics/RankFilter_lookup_libraries")'/>	       
+	       
+    <param name="window" type="float" label="Window" value="10.56" />
+  </inputs>
+  <outputs>
+    <data format="tabular" label="${tool.name}" name="onefile" />
+  </outputs>
+  <!-- file with implementation of the function get_directory_files() used above  -->
+  <code file="match_library.py" />
+  <configfiles>
+    <configfile name="input_file">
+      sample = ${sample}
+      calibration = ${calibration}
+      lib_data = ${lib_data}
+      window = ${window}
+      analysis_type = ${analysis_type}
+      tabular = True
+      onefile = ${onefile}
+      model = ${model}
+    </configfile>
+  </configfiles>
+  <help>
+Basically estimates the experimental RI (RIexp) by building a RI(RT) function based on the
+given calibration file.   
+
+It also determines the estimated RI (RIsvr) by looking up for each entry of the given input file (Sample File), 
+based on its CAS number, its respective RIsvr value in the given global lookup library
+(this step is also called the "RankFilter analysis" -see reference below; Sample File may be either from NIST or AMDIS). 
+This generates an prediction of the RI for 
+a compound according to the "RankFilter procedure" (RIsvr). 
+
+Output is a tab separated file in which four columns are added:
+
+	- **Rank** Calculated rank
+	- **RIexp** Experimental Retention Index (RI)
+	- **RIsvr** Calculated RI based on support vector regression (SVR)
+	- **%rel.err** Relative RI error (%rel.error = 100 * (RISVR − RIexp) / RIexp)
+
+.. class:: infomark
+
+**Notes**
+
+	- The layout of the Calibration file should include the following columns: 'MW', 'R.T.' and 'RI'.
+	- Selecting 'Polynomial' in the model parameter will calculate a 3rd degree polynomial model that will
+	  be used to convert from XXXX to YYYY.
+
+-----
+
+**References**
+
+    - **RankFilter**: Mihaleva et. al. (2009) *Automated procedure for candidate compound selection in GC-MS 
+      metabolomics based on prediction of Kovats retention index*. Bioinformatics, 25 (2009), pp. 787–794
+  </help>
+</tool>