OpenMS  2.5.0
IdXMLFile.h
Go to the documentation of this file.
1 // --------------------------------------------------------------------------
2 // OpenMS -- Open-Source Mass Spectrometry
3 // --------------------------------------------------------------------------
4 // Copyright The OpenMS Team -- Eberhard Karls University Tuebingen,
5 // ETH Zurich, and Freie Universitaet Berlin 2002-2020.
6 //
7 // This software is released under a three-clause BSD license:
8 // * Redistributions of source code must retain the above copyright
9 // notice, this list of conditions and the following disclaimer.
10 // * Redistributions in binary form must reproduce the above copyright
11 // notice, this list of conditions and the following disclaimer in the
12 // documentation and/or other materials provided with the distribution.
13 // * Neither the name of any author or any participating institution
14 // may be used to endorse or promote products derived from this software
15 // without specific prior written permission.
16 // For a full list of authors, refer to the file AUTHORS.
17 // --------------------------------------------------------------------------
18 // THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
19 // AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
20 // IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
21 // ARE DISCLAIMED. IN NO EVENT SHALL ANY OF THE AUTHORS OR THE CONTRIBUTING
22 // INSTITUTIONS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
23 // EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
24 // PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS;
25 // OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY,
26 // WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR
27 // OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF
28 // ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
29 //
30 // --------------------------------------------------------------------------
31 // $Maintainer: Timo Sachsenberg $
32 // $Authors: Marc Sturm $
33 // --------------------------------------------------------------------------
34 
35 #pragma once
36 
41 #include <OpenMS/FORMAT/XMLFile.h>
42 
43 #include <vector>
44 
45 namespace OpenMS
46 {
63  class OPENMS_DLLAPI IdXMLFile :
64  protected Internal::XMLHandler,
65  public Internal::XMLFile,
66  public ProgressLogger
67  {
68 public:
69  // both ConsensusXMLFile and FeatureXMLFile use some protected IdXML helper functions to parse identifications without code duplication
70  friend class ConsensusXMLFile;
71  friend class FeatureXMLFile;
72 
74  IdXMLFile();
75 
85  void load(const String& filename, std::vector<ProteinIdentification>& protein_ids, std::vector<PeptideIdentification>& peptide_ids);
86 
96  void load(const String& filename, std::vector<ProteinIdentification>& protein_ids, std::vector<PeptideIdentification>& peptide_ids, String& document_id);
97 
105  void store(const String& filename, const std::vector<ProteinIdentification>& protein_ids, const std::vector<PeptideIdentification>& peptide_ids, const String& document_id = "");
106 
107 
108 protected:
109  // Docu in base class
110  void endElement(const XMLCh* const /*uri*/, const XMLCh* const /*local_name*/, const XMLCh* const qname) override;
111 
112  // Docu in base class
113  void startElement(const XMLCh* const /*uri*/, const XMLCh* const /*local_name*/, const XMLCh* const qname, const xercesc::Attributes& attributes) override;
114 
118  void addProteinGroups_(MetaInfoInterface& meta, const std::vector<ProteinIdentification::ProteinGroup>& groups,
119  const String& group_name, const std::unordered_map<std::string, UInt>& accession_to_id, XMLHandler::ActionMode mode);
120 
122  void getProteinGroups_(std::vector<ProteinIdentification::ProteinGroup>& groups, const String& group_name);
123 
128  static std::ostream& createFlankingAAXMLString_(const std::vector<PeptideEvidence> & pes, std::ostream& os);
129 
134  static std::ostream& createPositionXMLString_(const std::vector<PeptideEvidence> & pes, std::ostream& os);
135 
136 
140  static void writeFragmentAnnotations_(const String & tag_name, std::ostream & os,
141  const std::vector<PeptideHit::PeakAnnotation>& annotations, UInt indent);
142 
146  static void parseFragmentAnnotation_(const String& s, std::vector<PeptideHit::PeakAnnotation> & annotations);
147 
148 
150 
151  std::vector<ProteinIdentification>* prot_ids_;
154  std::vector<PeptideIdentification>* pep_ids_;
158  std::map<String, ProteinIdentification::SearchParameters> parameters_;
174  std::vector<PeptideEvidence> peptide_evidences_;
176  std::unordered_map<std::string, String> proteinid_to_accession_;
182  };
183 
184 } // namespace OpenMS
185 
LogStream.h
OpenMS::PeptideHit::getCharge
Int getCharge() const
returns the charge of the peptide
ConsensusXMLFile.h
OpenMS::TOPPBase
Base class for TOPP applications.
Definition: TOPPBase.h:144
OpenMS::IdXMLFile::prot_id_in_run_
bool prot_id_in_run_
true if a prot id is contained in the current run
Definition: IdXMLFile.h:180
FileHandler.h
OpenMS::IdXMLFile::current_analysis_result_
PeptideHit::PepXMLAnalysisResult current_analysis_result_
Temporary analysis result instance.
Definition: IdXMLFile.h:172
XMLFile.h
OpenMS::ConsensusXMLFile::store
void store(const String &filename, const ConsensusMap &consensus_map)
Stores a consensus map to file.
OpenMS::IdXMLFile::store
void store(const String &filename, const std::vector< ProteinIdentification > &protein_ids, const std::vector< PeptideIdentification > &peptide_ids, const String &document_id="")
Stores the data in an idXML file.
OpenMS::PeptideHit::getSequence
const AASequence & getSequence() const
returns the peptide sequence without trailing or following spaces
OpenMS::Internal::XMLHandler
Base class for XML handlers.
Definition: XMLHandler.h:151
OpenMS::FeatureMap::Iterator
Base::iterator Iterator
Definition: FeatureMap.h:139
OpenMS::Exception::InvalidValue
Invalid value exception.
Definition: Exception.h:335
OpenMS::MzMLFile::store
void store(const String &filename, const PeakMap &map) const
Stores a map in an MzML file.
OpenMS::FileTypes::MZML
MzML file (.mzML)
Definition: FileTypes.h:72
OpenMS::PeptideHit::PepXMLAnalysisResult
Analysis Result (containing search engine / prophet results)
Definition: PeptideHit.h:210
OpenMS::MzMLFile
File adapter for MzML files.
Definition: MzMLFile.h:55
OpenMS::IdXMLFile::peptide_evidences_
std::vector< PeptideEvidence > peptide_evidences_
Temporary peptide evidences.
Definition: IdXMLFile.h:174
OpenMS::String
A more convenient string class.
Definition: String.h:58
OpenMS::MSExperiment::begin
Iterator begin()
Definition: MSExperiment.h:157
OpenMS::DataValue::isEmpty
bool isEmpty() const
Test if the value is empty.
Definition: DataValue.h:375
MzMLFile.h
OpenMS::MSExperiment
In-Memory representation of a mass spectrometry experiment.
Definition: MSExperiment.h:77
OpenMS::Size
size_t Size
Size type e.g. used as variable which can hold result of size()
Definition: Types.h:127
IdXMLFile.h
FeatureXMLFile.h
OpenMS::ProteinHit
Representation of a protein hit.
Definition: ProteinHit.h:57
OpenMS::DataValue::toString
String toString(bool full_precision=true) const
Conversion to String full_precision Controls number of fractional digits for all double types or list...
OpenMS::AASequence::toString
String toString() const
returns the peptide as string with modifications embedded in brackets
XMLHandler.h
OpenMS::ProteinIdentification
Representation of a protein identification run.
Definition: ProteinIdentification.h:71
OpenMS::ExperimentalSettings::getProteinIdentifications
const std::vector< ProteinIdentification > & getProteinIdentifications() const
returns a const reference to the protein ProteinIdentification vector
OpenMS::IdXMLFile::document_id_
String * document_id_
Document identifier.
Definition: IdXMLFile.h:178
OpenMS::IdXMLFile::load
void load(const String &filename, std::vector< ProteinIdentification > &protein_ids, std::vector< PeptideIdentification > &peptide_ids)
Loads the identifications of an idXML file without identifier.
OpenMS::IdXMLFile::param_
ProteinIdentification::SearchParameters param_
Temporary search parameters variable.
Definition: IdXMLFile.h:160
OpenMS::ProteinHit::getAccession
const String & getAccession() const
returns the accession of the protein
OpenMS::IdXMLFile::id_
String id_
Temporary id.
Definition: IdXMLFile.h:162
OpenMS
Main OpenMS namespace.
Definition: FeatureDeconvolution.h:46
OpenMS::MSExperiment::Iterator
std::vector< SpectrumType >::iterator Iterator
Mutable iterator.
Definition: MSExperiment.h:111
OpenMS::IdXMLFile::prot_id_
ProteinIdentification prot_id_
Temporary protein ProteinIdentification.
Definition: IdXMLFile.h:164
OpenMS::IdXMLFile::last_meta_
MetaInfoInterface * last_meta_
Pointer to last read object with MetaInfoInterface.
Definition: IdXMLFile.h:156
OpenMS::FileTypes::FEATUREXML
OpenMS feature file (.featureXML)
Definition: FileTypes.h:65
OpenMS::ProgressLogger
Base class for all classes that want to report their progress.
Definition: ProgressLogger.h:54
OpenMS::DataValue
Class to hold strings, numeric values, lists of strings and lists of numeric values.
Definition: DataValue.h:56
OpenMS::IdXMLFile::parameters_
std::map< String, ProteinIdentification::SearchParameters > parameters_
Search parameters map (key is the "id")
Definition: IdXMLFile.h:158
ProgressLogger.h
OpenMS::MzMLFile::load
void load(const String &filename, PeakMap &map)
Loads a map from a MzML file. Spectra and chromatograms are sorted by default (this can be disabled u...
OpenMS::MetaInfoInterface
Interface for classes that can store arbitrary meta information (Type-Name-Value tuples).
Definition: MetaInfoInterface.h:60
OpenMS::ConsensusMap::getProteinIdentifications
const std::vector< ProteinIdentification > & getProteinIdentifications() const
non-mutable access to the protein identifications
OpenMS::FeatureXMLFile::load
void load(const String &filename, FeatureMap &feature_map)
loads the file with name filename into map and calls updateRanges().
FeatureMap.h
ProteinIdentification.h
OpenMS::FileTypes::Type
Type
Actual file types enum.
Definition: FileTypes.h:58
OpenMS::FileHandler::getType
static FileTypes::Type getType(const String &filename)
Tries to determine the file type (by name or content)
OpenMS::ConsensusMap::getUnassignedPeptideIdentifications
const std::vector< PeptideIdentification > & getUnassignedPeptideIdentifications() const
non-mutable access to the unassigned peptide identifications
OpenMS::ConsensusMap
A container for consensus elements.
Definition: ConsensusMap.h:79
OpenMS::FeatureMap::getUnassignedPeptideIdentifications
const std::vector< PeptideIdentification > & getUnassignedPeptideIdentifications() const
non-mutable access to the unassigned peptide identifications
OpenMS::FeatureXMLFile::store
void store(const String &filename, const FeatureMap &feature_map)
stores the map feature_map in file with name filename.
OpenMS::ConsensusXMLFile::load
void load(const String &filename, ConsensusMap &map)
Loads a consensus map from file and calls updateRanges.
OpenMS::UInt
unsigned int UInt
Unsigned integer type.
Definition: Types.h:94
OpenMS::IdXMLFile::proteinid_to_accession_
std::unordered_map< std::string, String > proteinid_to_accession_
Map from protein id to accession.
Definition: IdXMLFile.h:176
main
int main(int argc, const char **argv)
Definition: INIFileEditor.cpp:73
OpenMS::Exception::RequiredParameterNotGiven
A required parameter was not given.
Definition: TOPPBase.h:111
OpenMS::FeatureMap
A container for features.
Definition: FeatureMap.h:95
OpenMS::PeptideIdentification
Represents the peptide hits for a spectrum.
Definition: PeptideIdentification.h:62
OpenMS::Internal::XMLHandler::ActionMode
ActionMode
Action to set the current mode (for error messages)
Definition: XMLHandler.h:169
OpenMS::MSExperiment::end
Iterator end()
Definition: MSExperiment.h:167
OpenMS::PeptideHit::getScore
double getScore() const
returns the PSM score
OpenMS::ConsensusMap::Iterator
std::vector< ConsensusFeature >::iterator Iterator
Mutable iterator.
Definition: ConsensusMap.h:160
OpenMS::IdXMLFile::pep_ids_
std::vector< PeptideIdentification > * pep_ids_
Pointer to fill in peptide identifications.
Definition: IdXMLFile.h:154
OpenMS::FeatureXMLFile
This class provides Input/Output functionality for feature maps.
Definition: FeatureXMLFile.h:68
OpenMS::IdXMLFile::pep_hit_
PeptideHit pep_hit_
Temporary peptide hit.
Definition: IdXMLFile.h:170
OpenMS::Internal::XMLFile
Base class for loading/storing XML files that have a handler derived from XMLHandler.
Definition: XMLFile.h:48
OpenMS::IdXMLFile::prot_hit_
ProteinHit prot_hit_
Temporary protein hit.
Definition: IdXMLFile.h:168
OpenMS::Exception::MissingInformation
Not all required information provided.
Definition: Exception.h:195
OPENMS_LOG_INFO
#define OPENMS_LOG_INFO
Macro if a information, e.g. a status should be reported.
Definition: LogStream.h:465
OpenMS::ProteinHit::getScore
double getScore() const
returns the score of the protein hit
PeptideIdentification.h
OpenMS::ProteinIdentification::SearchParameters
Search parameters of the DB search.
Definition: ProteinIdentification.h:221
OpenMS::DataProcessing::FILTERING
Data filtering or extraction.
Definition: DataProcessing.h:71
OpenMS::IdXMLFile::pep_id_
PeptideIdentification pep_id_
Temporary peptide ProteinIdentification.
Definition: IdXMLFile.h:166
TOPPBase.h
OpenMS::FeatureMap::getProteinIdentifications
const std::vector< ProteinIdentification > & getProteinIdentifications() const
non-mutable access to the protein identifications
OpenMS::ConsensusXMLFile
This class provides Input functionality for ConsensusMaps and Output functionality for alignments and...
Definition: ConsensusXMLFile.h:61
OpenMS::IdXMLFile
Used to load and store idXML files.
Definition: IdXMLFile.h:63
OpenMS::PeptideHit
Representation of a peptide hit.
Definition: PeptideHit.h:54