Mercurial > repos > pieterlukasse > prims_proteomics
annotate msfilt.xml @ 28:6107b74eeb11 default tip
improved documentation
| author | pieter.lukasse@wur.nl |
|---|---|
| date | Fri, 30 Jan 2015 14:53:10 +0100 |
| parents | ad911e9aaf33 |
| children |
| rev | line source |
|---|---|
|
17
40ec8770780d
* Added support for pepxml (and more specifically for
pieter.lukasse@wur.nl
parents:
10
diff
changeset
|
1 <tool name="MsFilt" id="msfilt" version="1.0.4"> |
| 0 | 2 <description>Filters annotations based MS/MS peptide identification and annotation quality measures</description> |
| 3 <!-- | |
| 4 For remote debugging start you listener on port 8000 and use the following as command interpreter: | |
| 5 java -jar -Xdebug -Xrunjdwp:transport=dt_socket,address=D0100564.wurnet.nl:8000 | |
| 6 ////////////////////////// | |
| 7 --> | |
| 8 <command interpreter="java -jar "> | |
| 9 MsFilt.jar | |
| 10 -apmlFile $apmlFile | |
| 11 -datasetCode $apmlFile.metadata.base_name | |
| 12 -rankingMetadataFile $rankingMetadataFile | |
| 13 -statisticalMeasuresConfigFile $statisticalMeasuresConfigFile | |
| 14 -annotationSourceConfigFile $annotationSourceConfigFile | |
| 15 -outApml $outputApml | |
| 16 -outNewIdsApml $outNewIdsApml | |
| 17 -outFullCSV $outputCSV | |
| 18 -outRankingTable $outRankingTable | |
| 19 -outProteinCoverageCSV $outProteinCoverageCSV | |
| 20 -fpCriteriaExpression "$fpCriteriaExpression" | |
| 21 -filterOutFPAnnotations $filterOutFPAnnotations | |
| 22 -fpCriteriaExpressionForIds "$fpCriteriaExpressionForIds" | |
| 23 -filterOutFPIds $filterOutFPIds | |
| 24 -filterOutUnannotatedAlignments $filterOutUnannotatedAlignments | |
| 25 -addRawRankingInfo $addRawRankingInfo | |
| 26 -addScaledIntensityInfo $addScaledIntensityInfo | |
| 27 -addRawIntensityInfo $addRawIntensityInfo | |
| 28 -outReport $htmlReportFile | |
| 29 -outReportPicturesPath $htmlReportFile.files_path | |
|
17
40ec8770780d
* Added support for pepxml (and more specifically for
pieter.lukasse@wur.nl
parents:
10
diff
changeset
|
30 #if $containsPepxml.pepxmlInSet == True |
|
40ec8770780d
* Added support for pepxml (and more specifically for
pieter.lukasse@wur.nl
parents:
10
diff
changeset
|
31 -pepxmlDataType $containsPepxml.pepxmlDataType |
|
40ec8770780d
* Added support for pepxml (and more specifically for
pieter.lukasse@wur.nl
parents:
10
diff
changeset
|
32 -pepxmlGeneratedBy $containsPepxml.pepxmlGeneratedBy |
|
40ec8770780d
* Added support for pepxml (and more specifically for
pieter.lukasse@wur.nl
parents:
10
diff
changeset
|
33 #end if |
| 0 | 34 </command> |
| 35 | |
| 36 <inputs> | |
| 37 | |
| 38 <param name="apmlFile" type="data" format="apml" optional="true" | |
| 39 label="(Optional) Peptide quantification file (APML)" | |
| 40 help="The APML contents as aligned and annotated feature lists. E.g. produced by | |
| 41 SEDMAT or Quantiline tools." /> | |
| 42 | |
| 43 <repeat name="annotationSourceFiles" title="(Optional) Peptide identification files" help="Full set of MS/MS peptide identification files, including peptides that could not be quantified."> | |
|
17
40ec8770780d
* Added support for pepxml (and more specifically for
pieter.lukasse@wur.nl
parents:
10
diff
changeset
|
44 <param name="identificationsFile" type="data" format="apml,pepxml,mzidentml,prims.fileset.zip" label="Identifications file (APML, pepxml, MZIDENTML or MZIDENTML fileSet)" /> |
| 0 | 45 </repeat> |
| 46 | |
|
17
40ec8770780d
* Added support for pepxml (and more specifically for
pieter.lukasse@wur.nl
parents:
10
diff
changeset
|
47 <!-- ================== PEPXML specific ================== --> |
|
40ec8770780d
* Added support for pepxml (and more specifically for
pieter.lukasse@wur.nl
parents:
10
diff
changeset
|
48 <conditional name="containsPepxml"> |
|
40ec8770780d
* Added support for pepxml (and more specifically for
pieter.lukasse@wur.nl
parents:
10
diff
changeset
|
49 <param name="pepxmlInSet" type="boolean" truevalue="Yes" falsevalue="No" checked="false" |
|
40ec8770780d
* Added support for pepxml (and more specifically for
pieter.lukasse@wur.nl
parents:
10
diff
changeset
|
50 label="Identifications set contains one or more files in pepxml format" |
|
40ec8770780d
* Added support for pepxml (and more specifically for
pieter.lukasse@wur.nl
parents:
10
diff
changeset
|
51 help="Indicate whether one or more (Optional) Peptide identification files is in pepxml format. Support for pepxml is still considered 'beta'."/> |
|
40ec8770780d
* Added support for pepxml (and more specifically for
pieter.lukasse@wur.nl
parents:
10
diff
changeset
|
52 <when value="Yes"> |
|
40ec8770780d
* Added support for pepxml (and more specifically for
pieter.lukasse@wur.nl
parents:
10
diff
changeset
|
53 <param name="pepxmlDataType" type="select" label=">> Type of data stored in the pepxml" |
|
40ec8770780d
* Added support for pepxml (and more specifically for
pieter.lukasse@wur.nl
parents:
10
diff
changeset
|
54 help="Options marked with (*) are ProteomeDiscoverer specific scenarios"> |
|
40ec8770780d
* Added support for pepxml (and more specifically for
pieter.lukasse@wur.nl
parents:
10
diff
changeset
|
55 <option value="" selected="true">--Please select--</option> |
|
40ec8770780d
* Added support for pepxml (and more specifically for
pieter.lukasse@wur.nl
parents:
10
diff
changeset
|
56 <option value="single_2d" >2D LC-MS runs, one per msms_run_summary</option> |
|
40ec8770780d
* Added support for pepxml (and more specifically for
pieter.lukasse@wur.nl
parents:
10
diff
changeset
|
57 <option value="multi_2d">(*) 2D LC-MS runs, multiple runs (e.g. rx.F1 to rx.FN) merged as a 'single' msms_run_summary</option> |
|
40ec8770780d
* Added support for pepxml (and more specifically for
pieter.lukasse@wur.nl
parents:
10
diff
changeset
|
58 <option value="single_1d">1D LC-MS runs, one per msms_run_summary</option> |
|
40ec8770780d
* Added support for pepxml (and more specifically for
pieter.lukasse@wur.nl
parents:
10
diff
changeset
|
59 </param> |
|
40ec8770780d
* Added support for pepxml (and more specifically for
pieter.lukasse@wur.nl
parents:
10
diff
changeset
|
60 <param name="pepxmlGeneratedBy" type="select" label=">> pepxml generated by" |
|
40ec8770780d
* Added support for pepxml (and more specifically for
pieter.lukasse@wur.nl
parents:
10
diff
changeset
|
61 help="Some tools, like ProteomeDiscoverer 1.4, have specific issues in their pepxml generation logic. Correctly indicating the tool used here will ensure known issues are taken |
|
40ec8770780d
* Added support for pepxml (and more specifically for
pieter.lukasse@wur.nl
parents:
10
diff
changeset
|
62 into consideration when the file is parsed." > |
|
40ec8770780d
* Added support for pepxml (and more specifically for
pieter.lukasse@wur.nl
parents:
10
diff
changeset
|
63 <option value="" selected="true">--Please select--</option> |
|
40ec8770780d
* Added support for pepxml (and more specifically for
pieter.lukasse@wur.nl
parents:
10
diff
changeset
|
64 <option value="proteome_discoverer_v1.4">ProteomeDiscoverer 1.4</option> |
|
40ec8770780d
* Added support for pepxml (and more specifically for
pieter.lukasse@wur.nl
parents:
10
diff
changeset
|
65 <option value="other">Other</option> |
|
40ec8770780d
* Added support for pepxml (and more specifically for
pieter.lukasse@wur.nl
parents:
10
diff
changeset
|
66 </param> |
|
40ec8770780d
* Added support for pepxml (and more specifically for
pieter.lukasse@wur.nl
parents:
10
diff
changeset
|
67 </when> |
|
40ec8770780d
* Added support for pepxml (and more specifically for
pieter.lukasse@wur.nl
parents:
10
diff
changeset
|
68 <when value="No"> |
|
40ec8770780d
* Added support for pepxml (and more specifically for
pieter.lukasse@wur.nl
parents:
10
diff
changeset
|
69 </when> |
|
40ec8770780d
* Added support for pepxml (and more specifically for
pieter.lukasse@wur.nl
parents:
10
diff
changeset
|
70 </conditional> |
|
40ec8770780d
* Added support for pepxml (and more specifically for
pieter.lukasse@wur.nl
parents:
10
diff
changeset
|
71 <!-- ================== END - PEPXML specific ================== --> |
|
40ec8770780d
* Added support for pepxml (and more specifically for
pieter.lukasse@wur.nl
parents:
10
diff
changeset
|
72 |
| 0 | 73 <!-- |
| 74 <param name="maxNrRankings" type="integer" size="10" value="0" label="Maximum nr. of items to leave in the final ranking (set=0 for no limit) " /> | |
| 75 --> | |
| 76 <!-- TODO add info somewhere that deltaRt is 'corrected deltaRt' --> | |
| 77 <param name="rankingWeightConfig" type="text" area="true" size="13x70" label="Quality Measures (qm's) and ranking weights configuration" | |
| 78 help="Here you may specify a weight for each of the Quality Measures (QMs). These are used for the final QM score and possibly for ranking (e.g. in case of label-free data | |
| 79 processed by SEDMAT). The format is: QM alias => QM name,weight. " | |
| 80 value="qmDRT => delta rt (standard score),1 | |
| 81 
qmDMA => delta mass annotation (standard score),1 | |
| 82 
qmDMP => delta mass psm (standard score),1 | |
| 83 
qmBSCR => best peptide score (standard score),1 | |
| 84 
qmALCV => alignment coverage (fraction),1 | |
| 85 
qmSTCV => score type coverage (fraction),1 | |
| 86 
qmPACV => peptide's best proteinAnnotCoverage (standard score),1 | |
| 87 
qmPICV => peptide's best proteinIdentifCoverage (standard score),1 | |
| 88 
qmANS => annotation sources (count),1 | |
| 89 
qmCSEV => charge states evidence (count),0.2 | |
| 90 
qmBCSP=> best correlation with source or product peptide (correl),1 | |
| 91 
qmBCCS => best correlation with other charge state (correl),1 | |
| 92 
qmBCOS => best correlation with other sibling peptide (correl),1 | |
| 93 "/> | |
| 94 | |
|
17
40ec8770780d
* Added support for pepxml (and more specifically for
pieter.lukasse@wur.nl
parents:
10
diff
changeset
|
95 <param name="statisticalMeasuresConfig" type="text" area="true" size="8x70" label="Statistical measures configuration" |
| 0 | 96 help="Here you may specify the statistical measures that are found in the ms/ms results (e.g. p or e-values). |
| 97 The format is: SM alias => SM name,type,mode[min/max]. " | |
| 98 value="smXTD => MS:1001330,XSLASH!Tandem:expect,min | |
| 99 
pvCSVEX => p_value,CSV_EXPORT,min | |
| 100 
smAUTO_LIKELIHOOD => AUTOMOD_LOGLIKELIHOOD,PLGS/Auto-mod,max | |
| 101 
smLIKELIHOOD => LOGLIKELIHOOD,PLGS/Databank-search,max | |
|
17
40ec8770780d
* Added support for pepxml (and more specifically for
pieter.lukasse@wur.nl
parents:
10
diff
changeset
|
102 
smPercoProb => Percolator: probability,Percolator probability,max |
|
40ec8770780d
* Added support for pepxml (and more specifically for
pieter.lukasse@wur.nl
parents:
10
diff
changeset
|
103 
smPercoPEP => Percolator: PEP,Percolator PEP,min |
|
40ec8770780d
* Added support for pepxml (and more specifically for
pieter.lukasse@wur.nl
parents:
10
diff
changeset
|
104 
smPercoQval => Percolator: q-Value,Percolator q-Value,max |
| 0 | 105 "/> |
| 106 | |
| 107 <param name="filterOutUnannotatedAlignments" type="boolean" checked="true" | |
| 108 label="Filter out unannotated alignments" | |
| 109 help="This helps decrease the output file size (features with no annotation are then not reported anymore)"/> | |
| 110 | |
| 111 <param name="filterOutFPAnnotations" type="boolean" checked="true" | |
| 112 label="Filter out False Positive (FP) annotations" /> | |
| 113 | |
| 114 <param name="fpCriteriaExpression" type="text" size="120" label="False Positive (FP) criteria for annotations" | |
| 115 help="Criteria (in standard score measures) for classifying an annotation as False Positive (FP). | |
| 116 You can build logical rules using the QM aliases above, the keywords 'and', 'or' and parenthesis. | |
| 117 Comparisons can be made with '==,<,><=,>='" | |
| 118 value="qmDRT <0 or qmDMA <-0.5 or (qmDMP <-0.5 and qmBSCR<-0.5) or (!isNaN(smXTD) and smXTD >0.01)"/> | |
| 119 | |
| 120 | |
| 121 <param name="filterOutFPIds" type="boolean" checked="true" | |
| 122 label="Filter out False Positive (FP) peptide identifications" /> | |
| 123 | |
| 124 <param name="fpCriteriaExpressionForIds" type="text" size="120" | |
| 125 label="False Positive (FP) criteria for identifications" | |
| 126 help="Criteria (in standard score measures) for classifying a peptide identification as False Positive (FP). | |
| 127 Here you can use a subset of the quality measures (qmDMP, qmBSCR, qmSTCV, qmPICV, qmCSEV) and all statistical measures." | |
| 128 value="(qmDMP <-0.5 and qmBSCR<-0.5) or (!isNaN(smXTD) and smXTD >0.01)"/> | |
| 129 | |
| 130 | |
| 131 <param name="addRawRankingInfo" type="boolean" checked="false" | |
| 132 label="Include the raw scores/values of the ranking attributes in the CSV output" | |
| 133 help="This will result in one extra column per ranking attribute, each column holding the original data for this attribute (before normalization)."/> | |
| 134 | |
| 135 <param name="addScaledIntensityInfo" type="boolean" checked="false" | |
| 136 label="Include computed scaled intensity values in the CSV output" | |
| 137 help="The autoscaled and 'z-score'scaled (aka 'standard-score'scaled) intensity values are then added to the full CSV output file"/> | |
| 138 | |
| 139 <param name="addRawIntensityInfo" type="boolean" checked="false" | |
| 140 label="Include the raw intensity values in the CSV output" | |
| 141 help="The original intensity values (as found in the input file) are then added to the full CSV output file"/> | |
| 142 | |
| 143 | |
| 144 </inputs> | |
| 145 <configfiles> | |
| 146 <configfile name="rankingMetadataFile">${rankingWeightConfig}</configfile> | |
| 147 <configfile name="statisticalMeasuresConfigFile">${statisticalMeasuresConfig}</configfile> | |
| 148 <configfile name="annotationSourceConfigFile">## start comment | |
| 149 ## iterate over the selected files and store their names in the config file | |
| 150 #for $i, $s in enumerate( $annotationSourceFiles ) | |
| 10 | 151 ${s.identificationsFile} |
| 0 | 152 ## also print out the datatype in the next line, based on previously configured datatype |
|
17
40ec8770780d
* Added support for pepxml (and more specifically for
pieter.lukasse@wur.nl
parents:
10
diff
changeset
|
153 #if isinstance( $s.identificationsFile.datatype, $__app__.datatypes_registry.get_datatype_by_extension('pepxml').__class__): |
|
40ec8770780d
* Added support for pepxml (and more specifically for
pieter.lukasse@wur.nl
parents:
10
diff
changeset
|
154 pepxml |
|
40ec8770780d
* Added support for pepxml (and more specifically for
pieter.lukasse@wur.nl
parents:
10
diff
changeset
|
155 #elif isinstance( $s.identificationsFile.datatype, $__app__.datatypes_registry.get_datatype_by_extension('apml').__class__): |
| 0 | 156 apml |
| 157 #else: | |
| 158 mzid | |
| 159 #end if | |
| 160 #end for | |
| 161 ## end comment</configfile> | |
| 162 </configfiles> | |
| 163 <outputs> | |
| 164 <data name="outputApml" format="apml" label="${apmlFile.metadata.base_name} - ${tool.name} on ${on_string}: quantifications (filtered APML)" metadata_source="apmlFile"> | |
| 165 <!-- If the expression is false, the file is not created --> | |
| 166 <filter>( apmlFile != None )</filter> | |
| 167 </data> | |
| 168 <data name="outNewIdsApml" format="apml" label="${tool.name} on ${on_string}: identifications (filtered APML)" > | |
| 169 <filter>( filterOutFPIds == True )</filter> | |
| 170 </data> | |
| 171 <data name="outputCSV" format="csv" label="${apmlFile.metadata.base_name} - ${tool.name} on ${on_string}: Full CSV" metadata_source="apmlFile"> | |
| 172 <filter>( apmlFile != None )</filter> | |
| 173 </data> | |
| 174 <data name="outRankingTable" format="csv" label="${apmlFile.metadata.base_name} - ${tool.name} on ${on_string}: Ranking table (CSV)" metadata_source="apmlFile"> | |
| 175 <filter>( apmlFile != None )</filter> | |
| 176 </data> | |
| 177 <data name="outProteinCoverageCSV" format="csv" label="${tool.name} on ${on_string}: Protein coverage details (CSV)"> | |
| 178 <!-- If the expression is false, the file is not created --> | |
| 179 <filter>( len(list(enumerate(annotationSourceFiles))) > 0 )</filter> | |
| 180 </data> | |
| 181 <data name="htmlReportFile" format="html" label="${tool.name} on ${on_string} - HTML report"/> | |
| 182 </outputs> | |
| 183 <tests> | |
| 184 </tests> | |
| 185 <help> | |
| 186 | |
| 187 .. class:: infomark | |
| 188 | |
| 189 This tool takes in peptide quantification results (e.g. either by SEDMAT for label-free data or by Quantiline for labeled data) | |
| 190 and calculates a number of quality measures that can help in assessing the correctness of the quantification assignment and of the MS/MS peptide | |
| 191 identification itself. The user can use any combination of quality measures (qm's) and statistical measures (sm's) to filter out | |
| 192 low scoring entries. | |
| 193 | |
| 194 .. class:: infomark | |
| 195 | |
| 196 In the label-free data processed by SEDMAT it is possible that a feature quantification gets assigned to different peptides. This means | |
| 197 we have an ambiguous assignment. In such a case | |
| 198 this tool also does a ranking of the different assignments according to their quality measures so that the best scoring assignment | |
| 199 gets ranked as first. | |
| 200 | |
| 201 ----- | |
| 202 | |
| 203 **List of abbreviations** | |
| 204 | |
| 205 QM: Quality Measure | |
| 206 | |
| 207 SM: Statistical Measure (e.g. p-value, e-value from MS/MS identification) | |
| 208 | |
| 209 PSM: "Peptide to Spectrum Match" (aka peptide identification) | |
| 210 | |
| 211 FP: False Positive | |
| 212 | |
| 213 ----- | |
| 214 | |
| 215 **Filtering options details** | |
| 216 | |
| 217 The FP criteria will be applied to an annotation even if the corresponding quality measures involved | |
| 218 in the expression can NOT ALL be determined. QMs that cannot be determined, get the value 0 (zero) which is | |
| 219 equal to giving it the average value. | |
| 220 | |
| 221 The output report shows some plots that visualize the filtering done. This can help in fine-tuning the right filtering | |
| 222 criteria. | |
| 223 | |
| 18 | 224 |
| 225 *Filter criteria examples for filtering identifications* | |
| 226 | |
| 227 ================================== =========================================================================== | |
| 228 Data type Example filter criteria | |
| 229 ---------------------------------- --------------------------------------------------------------------------- | |
| 230 QExactive/ProteomeDiscoverer data * (qmBSCR<-1.0) or (!isNaN(smPercoProb) and 18-20*smPercoProb>qmBSCR) | |
| 231 X!Tandem identifications data * (qmDMP <-0.5 and qmBSCR<-0.5) or (!isNaN(smXTD) and smXTD>0.01) | |
| 232 ================================== =========================================================================== | |
| 233 | |
| 234 | |
| 0 | 235 ----- |
| 236 | |
| 237 **Output details** | |
| 238 | |
| 239 *APML output* | |
| 240 | |
| 241 This tools returns the given APML alignment file further annotated at the alignment level with the best ranking | |
| 242 peptides of each respective alignment. This APML can be used in subsequent Galaxy tools like the proteomics tools | |
| 243 from NBIC. | |
| 244 | |
| 245 The APML output can also be used for the Protein Inference step (see Quantifere tool). | |
| 246 | |
| 247 *CSV output* | |
| 248 | |
| 249 It also returns a CSV format output with the full quality measures and scoring and ranking details. The user could use | |
| 250 this to manually determine new weights for some of the quality measures by techniques such as | |
| 251 linear regression. In other words, this CSV can then be used to fine-tune the weights in a next run. | |
| 252 | |
| 253 Many of the quality measures (QMs) are normalized to their Standard Score (aka z-score). | |
| 254 `See Standard Score for more details...`__ | |
| 255 | |
| 256 Next to giving insight into how the ranking was established, a more complete version of this CSV file is also | |
| 257 generated for tools that cannot or won't process the APML output format. | |
| 258 | |
| 259 Below an brief overview of the CSV and an illustration of the ranking done in case of ambiguous peptides to feature assignments | |
| 260 (explained above, can happen in case of label-free data processing by SEDMAT). | |
| 261 | |
| 262 | |
| 263 .. image:: $PATH_TO_IMAGES/msfilt_csv_out.png | |
| 264 | |
| 265 | |
| 266 | |
| 267 .. __: javascript:window.open('http://en.wikipedia.org/wiki/Standard_score','popUpWindow','height=700,width=800,left=10,top=10,resizable=yes,scrollbars=yes,toolbar=yes,menubar=no,location=no,directories=no,status=yes') | |
| 268 | |
| 269 | |
| 270 | |
| 271 | |
| 272 </help> | |
| 273 </tool> |