1 #!/usr/bin/perl -w 2 # 3 # $RCSfile: ElementalAnalysisSDFiles.pl,v $ 4 # $Date: 2008/01/30 21:44:45 $ 5 # $Revision: 1.9 $ 6 # 7 # Author: Manish Sud <msud@san.rr.com> 8 # 9 # Copyright (C) 2004-2008 Manish Sud. All rights reserved. 10 # 11 # This file is part of MayaChemTools. 12 # 13 # MayaChemTools is free software; you can redistribute it and/or modify it under 14 # the terms of the GNU Lesser General Public License as published by the Free 15 # Software Foundation; either version 3 of the License, or (at your option) any 16 # later version. 17 # 18 # MayaChemTools is distributed in the hope that it will be useful, but without 19 # any warranty; without even the implied warranty of merchantability of fitness 20 # for a particular purpose. See the GNU Lesser General Public License for more 21 # details. 22 # 23 # You should have received a copy of the GNU Lesser General Public License 24 # along with MayaChemTools; if not, see <http://www.gnu.org/licenses/> or 25 # write to the Free Software Foundation Inc., 59 Temple Place, Suite 330, 26 # Boston, MA, 02111-1307, USA. 27 # 28 29 use 5.006; 30 use strict; 31 use FindBin; use lib "$FindBin::Bin/../lib"; 32 use Getopt::Long; 33 use File::Basename; 34 use Text::ParseWords; 35 use Benchmark; 36 use FileUtil; 37 use SDFileUtil; 38 use TextUtil; 39 use MolecularFormula; 40 41 my($ScriptName, %Options, $StartTime, $EndTime, $TotalTime); 42 43 # Autoflush STDOUT 44 $| = 1; 45 46 # Starting message... 47 $ScriptName = basename($0); 48 print "\n$ScriptName: Starting...\n\n"; 49 $StartTime = new Benchmark; 50 51 # Get the options and setup script... 52 SetupScriptUsage(); 53 if ($Options{help} || @ARGV < 1) { 54 die GetUsageFromPod("$FindBin::Bin/$ScriptName"); 55 } 56 57 my(@SDFilesList); 58 @SDFilesList = ExpandFileNames(\@ARGV, "sdf sd"); 59 60 my($SpecifiedFormulaFieldName, $CheckFormula, $Precision, $DetailLevel, @SpecifiedCalculations, @SpecifiedValueFieldNames); 61 ProcessOptions(); 62 63 print "Checking input SD file(s)...\n"; 64 my(@SDFilesOkay, @SDFilesOutFile, @SDFilesFormulaFieldName, @SDFilesValueFieldNamesMap); 65 RetrieveSDFilesInfo(); 66 67 # Generate output files... 68 my($Index, $SDFile); 69 if (@SDFilesList > 1) { 70 print "Processing SD file(s)...\n"; 71 } 72 for $Index (0 .. $#SDFilesList) { 73 if ($SDFilesOkay[$Index]) { 74 $SDFile = $SDFilesList[$Index]; 75 if (@SDFilesList > 1) { 76 print "\nProcessing file $SDFile...\n"; 77 } 78 else { 79 print "Processing file $SDFile...\n" 80 } 81 PerformElementalAnalysis($Index); 82 } 83 } 84 85 print "$ScriptName:Done...\n\n"; 86 87 $EndTime = new Benchmark; 88 $TotalTime = timediff ($EndTime, $StartTime); 89 print "Total time: ", timestr($TotalTime), "\n"; 90 91 ############################################################################### 92 93 # Perform analysis... 94 sub PerformElementalAnalysis { 95 my($Index) = @_; 96 my($SDFile, $NewSDFile, $KeyDataFieldName); 97 98 $SDFile = $SDFilesList[$Index]; 99 $NewSDFile = $SDFilesOutFile[$Index]; 100 101 print "Generating new SD file $NewSDFile...\n"; 102 open NEWSDFILE, ">$NewSDFile" or die "Error: Couldn't open $NewSDFile: $! \n"; 103 open SDFILE, "$SDFile" or die "Error: Can't open $SDFile: $! \n"; 104 105 # Go over all compound records and store 'em using key value as hash... 106 my($CmpdCount, $FormulaFieldName, $FormulaFieldValue, $CmpdString, $Value, $CalculationType, $CalculatedValue, $ErrorMsg, $Status, $ElementsRef, $ElementCompositionRef, @CalculatedValues, @CmpdLines, %DataFieldValuesMap); 107 $CmpdCount = 0; 108 $FormulaFieldName = $SDFilesFormulaFieldName[$Index]; 109 110 COMPOUND: while ($CmpdString = ReadCmpdString(\*SDFILE)) { 111 $CmpdCount++; 112 @CmpdLines = split "\n", $CmpdString; 113 %DataFieldValuesMap = GetCmpdDataHeaderLabelsAndValues(\@CmpdLines); 114 115 @CalculatedValues = (); 116 for $Value (@SpecifiedCalculations) { 117 push @CalculatedValues, ''; 118 } 119 120 if (!exists $DataFieldValuesMap{$FormulaFieldName}) { 121 $ErrorMsg = "Ignoring compound record $CmpdCount: Formula field $FormulaFieldName not found"; 122 PrintErrorMsg($CmpdString, $ErrorMsg); 123 WriteNewCompoundRecord($Index, \*NEWSDFILE, \@CmpdLines, \@CalculatedValues); 124 next COMPOUND; 125 } 126 127 # Make sure it's a valid molecular formula... 128 $FormulaFieldValue = $DataFieldValuesMap{$FormulaFieldName}; 129 if ($CheckFormula) { 130 ($Status, $ErrorMsg) = MolecularFormula::IsMolecularFormula($FormulaFieldValue); 131 if (!$Status) { 132 $ErrorMsg = "Ignoring compound record $CmpdCount: Formula field value $FormulaFieldValue is not valid: $ErrorMsg"; 133 PrintErrorMsg($CmpdString, $ErrorMsg); 134 WriteNewCompoundRecord($Index, \*NEWSDFILE, \@CmpdLines, \@CalculatedValues); 135 next COMPOUND; 136 } 137 } 138 # Calculate appropriate values and write 'em out... 139 @CalculatedValues = (); 140 for $CalculationType (@SpecifiedCalculations) { 141 if ($CalculationType =~ /^ElementalAnalysis$/i) { 142 ($ElementsRef, $ElementCompositionRef) = MolecularFormula::CalculateElementalComposition($FormulaFieldValue); 143 $CalculatedValue = (defined($ElementsRef) && defined($ElementCompositionRef)) ? MolecularFormula::FormatCompositionInfomation($ElementsRef, $ElementCompositionRef, $Precision) : ''; 144 } 145 elsif ($CalculationType =~ /^MolecularWeight$/i) { 146 $CalculatedValue = MolecularFormula::CalculateMolecularWeight($FormulaFieldValue); 147 $CalculatedValue = (defined($CalculatedValue) && length($CalculatedValue)) ? (sprintf("%.${Precision}f", $CalculatedValue)) : ""; 148 } 149 elsif ($CalculationType =~ /^ExactMass$/i) { 150 $CalculatedValue = MolecularFormula::CalculateExactMass($FormulaFieldValue); 151 $CalculatedValue = (defined($CalculatedValue) && length($CalculatedValue)) ? (sprintf("%.${Precision}f", $CalculatedValue)) : ""; 152 } 153 else { 154 $CalculatedValue = ''; 155 } 156 push @CalculatedValues, $CalculatedValue; 157 } 158 WriteNewCompoundRecord($Index, \*NEWSDFILE, \@CmpdLines, \@CalculatedValues); 159 } 160 close NEWSDFILE; 161 close SDFILE; 162 } 163 164 # Write out compound record with calculated values... 165 sub WriteNewCompoundRecord { 166 my($Index, $SDFileRef, $CmpdLinesRef, $CalculatedValuesRef) = @_; 167 168 # Write out compound lines except the last line which contains $$$$... 169 my($LineIndex); 170 for $LineIndex (0 .. ($#{$CmpdLinesRef} - 1)) { 171 print $SDFileRef "$CmpdLinesRef->[$LineIndex]\n"; 172 } 173 174 # Write out calculated values... 175 my($CalcIndex, $FieldName, $FieldValue); 176 for $CalcIndex (0 .. $#SpecifiedCalculations) { 177 $FieldName = $SDFilesValueFieldNamesMap[$Index]{$SpecifiedCalculations[$CalcIndex]}; 178 $FieldValue = $CalculatedValuesRef->[$CalcIndex]; 179 print $SDFileRef "> <$FieldName>\n$FieldValue\n\n"; 180 } 181 print $SDFileRef "\$\$\$\$\n"; 182 } 183 184 # Print out error message... 185 sub PrintErrorMsg { 186 my($CmpdString, $ErrorMsg) = @_; 187 188 if ($DetailLevel >= 2 ) { 189 print "$ErrorMsg:\n$CmpdString\n"; 190 } 191 elsif ($DetailLevel >= 1) { 192 print "$ErrorMsg\n"; 193 } 194 } 195 196 # Process option values... 197 sub ProcessOptions { 198 $DetailLevel = $Options{detail}; 199 $CheckFormula = $Options{fast} ? 0 : 1; 200 $Precision = $Options{precision}; 201 202 $SpecifiedFormulaFieldName = ""; 203 if (defined $Options{formulafield}) { 204 $SpecifiedFormulaFieldName = $Options{formulafield}; 205 } 206 207 # Setup what to calculate... 208 @SpecifiedCalculations = (); 209 if ($Options{mode} =~ /^All$/i) { 210 @SpecifiedCalculations = qw(ElementalAnalysis MolecularWeight ExactMass); 211 } 212 else { 213 my($Mode, $ModeValue, @SpecifiedModeValues); 214 $Mode = $Options{mode}; 215 $Mode =~ s/ //g; 216 @SpecifiedModeValues = split /\,/, $Mode; 217 for $ModeValue (@SpecifiedModeValues) { 218 if ($ModeValue !~ /^(ElementalAnalysis|MolecularWeight|ExactMass)$/i) { 219 if ($ModeValue =~ /^All$/i) { 220 die "Error: All value for option \"-m --mode\" is not allowed with other valid values.\n"; 221 } 222 else { 223 die "Error: The value specified, $ModeValue, for option \"-m --mode\" is not valid. Allowed values: ElementalAnalysis, MolecularWeight, or ExactMass\n"; 224 } 225 } 226 push @SpecifiedCalculations, $ModeValue; 227 } 228 } 229 230 @SpecifiedValueFieldNames = (); 231 if ($Options{valuefieldnames}) { 232 my($Value, $Label, @ValueLabels); 233 @ValueLabels = split /\,/, $Options{valuefieldnames}; 234 if (@ValueLabels % 2) { 235 die "Error: The value specified, $Options{valuefieldnames}, for option \"-v --valuefieldnames\" is not valid: It must contain even number of comma delimited values\n"; 236 } 237 for ($Index = 0; $Index < @ValueLabels; $Index +=2) { 238 $Value = $ValueLabels[$Index]; 239 $Value =~ s/ //g; 240 $Label = $ValueLabels[$Index + 1]; 241 if ($Value !~ /^(ElementalAnalysis|MolecularWeight|ExactMass)$/i) { 242 die "Error: The value specified, $Value, using option \"-v --valuefieldnames\" is not valid. Allowed values: ElementalAnalysis, MolecularWeight, or ExactMass\n"; 243 } 244 push @SpecifiedValueFieldNames, ($Value, $Label); 245 } 246 } 247 } 248 249 # Retrieve information about input SD files... 250 sub RetrieveSDFilesInfo { 251 my($Index, $SDFile, $FileDir, $FileName, $FileExt, $OutFileRoot, $OutFile, $FormulaFieldName, $Value, $FieldName, $NewFieldName, $Count); 252 253 my(%NewValueFieldNames) = (ElementalAnalysis => 'ElementalAnalysis', MolecularWeight => 'MolecularWeight', ExactMass => 'ExactMass'); 254 if (@SpecifiedValueFieldNames) { 255 for ($Index = 0; $Index < @SpecifiedValueFieldNames; $Index +=2) { 256 $Value = $SpecifiedValueFieldNames[$Index]; 257 $FieldName = $SpecifiedValueFieldNames[$Index + 1]; 258 if (exists $NewValueFieldNames{$Value}) { 259 $NewValueFieldNames{$Value} = $FieldName; 260 } 261 } 262 } 263 264 @SDFilesOkay = (); 265 @SDFilesOutFile = (); 266 @SDFilesFormulaFieldName = (); 267 @SDFilesValueFieldNamesMap = (); 268 269 FILELIST: for $Index (0 .. $#SDFilesList) { 270 $SDFile = $SDFilesList[$Index]; 271 $SDFilesOkay[$Index] = 0; 272 $SDFilesOutFile[$Index] = ''; 273 $SDFilesFormulaFieldName[$Index] = ''; 274 %{$SDFilesValueFieldNamesMap[$Index]} = (); 275 276 if (!(-e $SDFile)) { 277 warn "Warning: Ignoring file $SDFile: It doesn't exist\n"; 278 next FILELIST; 279 } 280 if (!CheckFileType($SDFile, "sd sdf")) { 281 warn "Warning: Ignoring file $SDFile: It's not a SD file\n"; 282 next FILELIST; 283 } 284 $FileDir = ""; $FileName = ""; $FileExt = ""; 285 ($FileDir, $FileName, $FileExt) = ParseFileName($SDFile); 286 if ($Options{root} && (@SDFilesList == 1)) { 287 my ($RootFileDir, $RootFileName, $RootFileExt) = ParseFileName($Options{root}); 288 if ($RootFileName && $RootFileExt) { 289 $FileName = $RootFileName; 290 } 291 else { 292 $FileName = $Options{root}; 293 } 294 $OutFileRoot = $FileName; 295 } 296 else { 297 $OutFileRoot = $FileName . "ElementalAnalysis"; 298 } 299 300 $OutFile = $OutFileRoot . ".$FileExt"; 301 if (lc($OutFile) eq lc($SDFile)) { 302 warn "Warning: Ignoring file $SDFile:Output file name, $OutFile, is same as input SD file name, $SDFile\n"; 303 next FILELIST; 304 } 305 if (!$Options{overwrite}) { 306 if (-e $OutFile) { 307 warn "Warning: Ignoring file $SDFile: The file $OutFile already exists\n"; 308 next FILELIST; 309 } 310 } 311 # Get data field names and values... 312 my($CmpdString, $FieldName, @CmpdLines, @DataFieldNames, %DataFieldNamesMap); 313 @DataFieldNames = (); 314 if (!open(SDFILE, "$SDFile")) { 315 warn "Warning: Ignoring file $SDFile: Couldn't open it: $! \n"; 316 next FILELIST; 317 } 318 $CmpdString = ReadCmpdString(\*SDFILE); 319 close SDFILE; 320 321 @CmpdLines = split "\n", $CmpdString; 322 @DataFieldNames = GetCmpdDataHeaderLabels(\@CmpdLines); 323 %DataFieldNamesMap = GetCmpdDataHeaderLabelsAndValues(\@CmpdLines); 324 325 # Setup formula field name... 326 $FormulaFieldName = ''; 327 if ($SpecifiedFormulaFieldName) { 328 $FormulaFieldName = $SpecifiedFormulaFieldName; 329 } 330 else { 331 FIELDNAME: for $FieldName (@DataFieldNames) { 332 if ($FieldName =~ /Formula/i) { 333 $FormulaFieldName = $FieldName; 334 last FIELDNAME; 335 } 336 } 337 if (!$FormulaFieldName) { 338 warn "Warning: Ignoring file $SDFile: Data field label containing the word Formula doesn't exist\n"; 339 next FILELIST; 340 } 341 } 342 $SDFilesOkay[$Index] = 1; 343 $SDFilesOutFile[$Index] = $OutFile; 344 $SDFilesFormulaFieldName[$Index] = $FormulaFieldName; 345 346 # Setup value data field names for calculated values... 347 for $Value (keys %NewValueFieldNames) { 348 $FieldName = $NewValueFieldNames{$Value}; 349 350 # Make sure it doesn't already exists... 351 $Count = 1; 352 $NewFieldName = $FieldName; 353 while (exists $DataFieldNamesMap{$NewFieldName}) { 354 $Count++; 355 $NewFieldName = $FieldName . $Count; 356 } 357 $SDFilesValueFieldNamesMap[$Index]{$Value} = $NewFieldName; 358 } 359 } 360 } 361 362 # Setup script usage and retrieve command line arguments specified using various options... 363 sub SetupScriptUsage { 364 365 # Retrieve all the options... 366 %Options = (); 367 $Options{detail} = 1; 368 $Options{mode} = "All"; 369 $Options{precision} = 2; 370 371 if (!GetOptions(\%Options, "detail|d=i", "fast", "formulafield|f=s", "mode|m=s", "help|h", "overwrite|o", "precision|p=i", "root|r=s", "valuefieldnames|v=s", "workingdir|w=s")) { 372 die "\nTo get a list of valid options and their values, use \"$ScriptName -h\" or\n\"perl -S $ScriptName -h\" command and try again...\n"; 373 } 374 if ($Options{workingdir}) { 375 if (! -d $Options{workingdir}) { 376 die "Error: The value specified, $Options{workingdir}, for option \"-w --workingdir\" is not a directory name.\n"; 377 } 378 chdir $Options{workingdir} or die "Error: Couldn't chdir $Options{workingdir}: $! \n"; 379 } 380 if (!IsPositiveInteger($Options{detail})) { 381 die "Error: The value specified, $Options{detail}, for option \"-d --detail\" is not valid. Allowed values: > 0\n"; 382 } 383 if (!IsPositiveInteger($Options{precision})) { 384 die "Error: The value specified, $Options{precision}, for option \"-p --precision\" is not valid. Allowed values: > 0 \n"; 385 } 386 } 387