|
NCBI Home IEB Home C++ Toolkit docs C Toolkit source browser C Toolkit source browser (2) |
NCBI C++ Toolkit Cross ReferenceC++/src/algo/ms/omssa/omssa2pepXML.cpp |
source navigation diff markup identifier search freetext search file search |
1 /* $Id: omssa2pepXML.cpp 139701 2008-09-08 16:08:42Z slottad $
2 * ===========================================================================
3 *
4 * PUBLIC DOMAIN NOTICE
5 * National Center for Biotechnology Information
6 *
7 * This software/database is a "United States Government Work" under the
8 * terms of the United States Copyright Act. It was written as part of
9 * the author's official duties as a United States Government employee and
10 * thus cannot be copyrighted. This software/database is freely available
11 * to the public for use. The National Library of Medicine and the U.S.
12 * Government have not placed any restriction on its use or reproduction.
13 *
14 * Although all reasonable efforts have been taken to ensure the accuracy
15 * and reliability of the software and data, the NLM and the U.S.
16 * Government do not and cannot warrant the performance or results that
17 * may be obtained by using this software or data. The NLM and the U.S.
18 * Government disclaim all warranties, express or implied, including
19 * warranties of performance, merchantability or fitness for any particular
20 * purpose.
21 *
22 * Please cite the author in any work or product based on this material.
23 *
24 * ===========================================================================
25 *
26 * Authors: Douglas Slotta
27 *
28 * File Description:
29 * Command line utility to convert OMSSA output to the PepXML format
30 *
31 */
32
33 #include <ncbi_pch.hpp>
34 #include <corelib/ncbiapp.hpp>
35 #include <corelib/ncbienv.hpp>
36 #include <corelib/ncbiargs.hpp>
37 #include <corelib/ncbifile.hpp>
38 #include <serial/serial.hpp>
39 #include <serial/objistr.hpp>
40 #include <serial/objostr.hpp>
41 #include <objects/omssa/omssa__.hpp>
42 #include <util/compress/bzip2.hpp>
43
44 #include "omssa.hpp"
45 #include "pepxml.hpp"
46
47 USING_SCOPE(ncbi);
48 USING_SCOPE(objects);
49 USING_SCOPE(omssa);
50
51 // Helper function for debugging, might need this again in the future
52 // void PrintModInfo(CRef <CMSModSpecSet> Modset) {
53 // for (unsigned int modNum = 0; modNum < Modset->Get().size(); modNum++) {
54 // cout << MSSCALE2DBL(Modset->GetModMass(modNum)) << "\t";
55 // cout << MSSCALE2DBL(Modset->GetNeutralLoss(modNum)) << "\t";
56 // cout << Modset->GetModNumChars(modNum) << "\t";
57 // for (int i=0; i< Modset->GetModNumChars(modNum); i++) {
58 // cout << ConvertAA(Modset->GetModChar(modNum, i)) << " ";
59 // cout << MonoMass[static_cast <int> (Modset->GetModChar(modNum,i))] << " ";
60 // }
61 // cout << "\t" << Modset->GetModType(modNum) << "\t";
62 // cout << Modset->GetModName(modNum) << endl;
63 // }
64 // }
65
66 /////////////////////////////////////////////////////////////////////////////
67 // COmssa2pepxmlApplication::
68
69
70 class COmssa2pepxmlApplication : public CNcbiApplication
71 {
72 public:
73 COmssa2pepxmlApplication();
74 private:
75 virtual void Init(void);
76 virtual int Run(void);
77 virtual void Exit(void);
78 };
79
80
81 COmssa2pepxmlApplication::COmssa2pepxmlApplication() {
82 SetVersion(CVersionInfo(2, 1, 4));
83 }
84
85 /////////////////////////////////////////////////////////////////////////////
86 // Init test for all different types of arguments
87 void COmssa2pepxmlApplication::Init(void)
88 {
89 // Create command-line argument descriptions class
90 auto_ptr<CArgDescriptions> arg_desc(new CArgDescriptions);
91 arg_desc->PrintUsageIfNoArgs();
92
93 // Specify USAGE context
94 arg_desc->SetUsageContext(GetArguments().GetProgramBasename(),
95 "Convert OMSSA ASN.1 and XML output files to PepXML");
96
97 // Describe the expected command-line arguments
98 arg_desc->AddFlag("xml", "Input file is XML");
99 arg_desc->AddFlag("bz2", "Input file is bzipped XML");
100 arg_desc->AddFlag("asn", "Input file is ASN.1");
101 arg_desc->AddFlag("asntext", "Input file is ASN.1 text");
102
103 arg_desc->AddOptionalKey("o", "outfile",
104 "filename for pepXML formatted search results",
105 CArgDescriptions::eString);
106
107 arg_desc->AddPositional
108 ("filename",
109 "The name of the XML file to load",
110 CArgDescriptions::eString);
111
112 // Setup arg.descriptions for this application
113 SetupArgDescriptions(arg_desc.release());
114 }
115
116
117
118 /////////////////////////////////////////////////////////////////////////////
119 // Run test (printout arguments obtained from command-line)
120 int COmssa2pepxmlApplication::Run ( void )
121 {
122 CMSSearch inOMSSA;
123 CPepXML outPepXML;
124 ESerialDataFormat format = eSerial_Xml; // assume xml
125
126 // Get arguments
127 const CArgs& args = GetArgs();
128
129 string filename = args["filename"].AsString();
130 CFile file(filename);
131
132 string fullpath, path, base, ext;
133 if (CFile::IsAbsolutePath(file.GetPath())) {
134 fullpath = file.GetPath();
135 } else {
136 fullpath = CFile::CreateAbsolutePath(file.GetPath());
137 }
138 CFile::SplitPath(fullpath, &path, &base, &ext);
139 string basename = path + base;
140 string newname;
141 if (args["o"].HasValue()) {
142 newname = args["o"].AsString();
143 } else {
144 newname = basename + ".pep.xml";
145 }
146
147 // figure out input file type
148 bool notSet = true;
149 if (args["xml"] || args["bz2"]) {
150 format = eSerial_Xml;
151 notSet = false;
152 } else if (args["asn"]) {
153 format = eSerial_AsnBinary;
154 notSet = false;
155 } else if (args["asntext"]) {
156 format = eSerial_AsnText;
157 notSet = false;
158 }
159
160 if (notSet) { // Not explict, maybe extension gives us a clue?
161 if (ext == ".oms") {
162 format = eSerial_AsnBinary;
163 } else if (ext == ".omx") {
164 format = eSerial_Xml;
165 } else if (ext == ".omt") {
166 format = eSerial_AsnText;
167 }
168 }
169 cout << "Reading " << filename << " as ";
170 switch (format) {
171 case eSerial_AsnBinary:
172 cout << "ASN" << endl; break;
173 case eSerial_Xml:
174 cout << "XML" << endl; break;
175 case eSerial_AsnText:
176 cout << "ASN text" << endl; break;
177 default:
178 cout << "Unable to determine type of file" << endl;
179 return 0;
180 }
181
182 CSearchHelper::ReadCompleteSearch(filename, format, args["bz2"], inOMSSA);
183
184 if (!inOMSSA.CanGetRequest()) {
185 cout << "Sorry, this file cannot be converted." << endl;
186 cout << "The original search needs to have been executed with the '-w' flag set." << endl;
187 cout << "The search settings are not availiable in this file. Aborting" << endl;
188 return 0;
189 }
190
191 CRef <CMSModSpecSet> Modset(new CMSModSpecSet);
192 CSearchHelper::ReadModFiles("mods.xml","usermods.xml",GetProgramExecutablePath(), Modset);
193 Modset->CreateArrays();
194 //PrintModInfo(Modset);
195
196 outPepXML.ConvertFromOMSSA(inOMSSA, Modset, basename, newname);
197
198 //CNcbiOfstream out(newname.c_str());
199 //out << MSerial_Xml << outPepXML;
200
201 auto_ptr<CObjectOStream> oStream(CObjectOStream::Open(newname, eSerial_Xml));
202 CObjectOStreamXml *xml_out = dynamic_cast <CObjectOStreamXml *> (oStream.get());
203 xml_out->SetDefaultSchemaNamespace("http://regis-web.systemsbiology.net/pepXML");
204 xml_out->SetReferenceSchema();
205 *xml_out << outPepXML;
206
207 return 0;
208 }
209
210
211 /////////////////////////////////////////////////////////////////////////////
212 // Cleanup
213
214
215 void COmssa2pepxmlApplication::Exit ( void )
216 {
217 SetDiagStream ( 0 );
218 }
219
220
221 /////////////////////////////////////////////////////////////////////////////
222 // MAIN
223
224
225 int main ( int argc, const char* argv[] )
226 {
227 // Execute main application function
228 return COmssa2pepxmlApplication().AppMain ( argc, argv, 0, eDS_Default, 0 );
229 }
230 |
This page was automatically generated by the
LXR engine.
Visit the LXR main site for more information. |