Mercurial > repos > deepakjadmin > mayatool3_test2
comparison bin/AnalyzeTextFilesData.pl @ 0:4816e4a8ae95 draft default tip
Uploaded
| author | deepakjadmin |
|---|---|
| date | Wed, 20 Jan 2016 09:23:18 -0500 |
| parents | |
| children |
comparison
equal
deleted
inserted
replaced
| -1:000000000000 | 0:4816e4a8ae95 |
|---|---|
| 1 #!/usr/bin/perl -w | |
| 2 # | |
| 3 # $RCSfile: AnalyzeTextFilesData.pl,v $ | |
| 4 # $Date: 2015/02/28 20:46:04 $ | |
| 5 # $Revision: 1.36 $ | |
| 6 # | |
| 7 # Author: Manish Sud <msud@san.rr.com> | |
| 8 # | |
| 9 # Copyright (C) 2015 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 strict; | |
| 30 use FindBin; use lib "$FindBin::Bin/../lib"; | |
| 31 use Getopt::Long; | |
| 32 use File::Basename; | |
| 33 use Text::ParseWords; | |
| 34 use Benchmark; | |
| 35 use FileUtil; | |
| 36 use TextUtil; | |
| 37 use StatisticsUtil; | |
| 38 | |
| 39 my($ScriptName, %Options, $StartTime, $EndTime, $TotalTime); | |
| 40 | |
| 41 # Autoflush STDOUT | |
| 42 $| = 1; | |
| 43 | |
| 44 # Starting message... | |
| 45 $ScriptName = basename($0); | |
| 46 print "\n$ScriptName: Starting...\n\n"; | |
| 47 $StartTime = new Benchmark; | |
| 48 | |
| 49 # Get the options and setup script... | |
| 50 SetupScriptUsage(); | |
| 51 if ($Options{help} || @ARGV < 1) { | |
| 52 die GetUsageFromPod("$FindBin::Bin/$ScriptName"); | |
| 53 } | |
| 54 | |
| 55 my(@TextFilesList); | |
| 56 @TextFilesList = ExpandFileNames(\@ARGV, "csv tsv"); | |
| 57 | |
| 58 print "Processing options...\n"; | |
| 59 my(%OptionsInfo); | |
| 60 ProcessOptions(); | |
| 61 | |
| 62 # Collect column information for all the text files... | |
| 63 print "Checking input text file(s)...\n"; | |
| 64 my(%TextFilesInfo); | |
| 65 RetrieveTextFilesInfo(); | |
| 66 ProcessColumnsInfo(); | |
| 67 | |
| 68 # Generate output files... | |
| 69 my($FileIndex); | |
| 70 if (@TextFilesList > 1) { | |
| 71 print "\nProcessing text files...\n"; | |
| 72 } | |
| 73 for $FileIndex (0 .. $#TextFilesList) { | |
| 74 if ($TextFilesInfo{FileOkay}[$FileIndex]) { | |
| 75 print "\nProcessing file $TextFilesList[$FileIndex]...\n"; | |
| 76 AnalyzeTextFile($FileIndex); | |
| 77 } | |
| 78 } | |
| 79 print "\n$ScriptName:Done...\n\n"; | |
| 80 | |
| 81 $EndTime = new Benchmark; | |
| 82 $TotalTime = timediff ($EndTime, $StartTime); | |
| 83 print "Total time: ", timestr($TotalTime), "\n"; | |
| 84 | |
| 85 ############################################################################### | |
| 86 | |
| 87 # Analyze data... | |
| 88 sub AnalyzeTextFile { | |
| 89 my($Index) = @_; | |
| 90 my($TextFile, $Line, $InDelim, $ColNum, $Value, @LineWords, @ColNumsToAnalyze, %ColValuesToAnalyzeMap); | |
| 91 | |
| 92 $TextFile = $TextFilesList[$Index]; | |
| 93 $InDelim = $TextFilesInfo{InDelim}[$Index]; | |
| 94 @ColNumsToAnalyze = @{$TextFilesInfo{UniqueColNumsToAnalyze}[$Index]}; | |
| 95 %ColValuesToAnalyzeMap = (); | |
| 96 for $ColNum (@ColNumsToAnalyze) { | |
| 97 @{$ColValuesToAnalyzeMap{$ColNum}} = (); | |
| 98 } | |
| 99 | |
| 100 my($LineCount, $InvalidLineCount, @InvalidColLabels); | |
| 101 | |
| 102 open TEXTFILE, "$TextFile" or die "Error: Can't open $TextFile: $! \n"; | |
| 103 # Skip over column labels line in text file and collect appropriate column data | |
| 104 # for analysis... | |
| 105 $Line = GetTextLine(\*TEXTFILE); | |
| 106 $LineCount = 1; | |
| 107 $InvalidLineCount = 0; | |
| 108 while ($Line = GetTextLine(\*TEXTFILE)) { | |
| 109 $LineCount++; | |
| 110 @LineWords = quotewords($InDelim, 0, $Line); | |
| 111 @InvalidColLabels = (); | |
| 112 COLNUM: for $ColNum (@ColNumsToAnalyze) { | |
| 113 $Value = $LineWords[$ColNum]; | |
| 114 if ($OptionsInfo{CheckData}) { | |
| 115 if (!IsNumerical($Value)) { | |
| 116 push @InvalidColLabels, $TextFilesInfo{ColLabels}[$Index][$ColNum]; | |
| 117 next COLNUM; | |
| 118 } | |
| 119 } | |
| 120 push @{$ColValuesToAnalyzeMap{$ColNum}}, $Value; | |
| 121 } | |
| 122 if (@InvalidColLabels) { | |
| 123 $InvalidLineCount++; | |
| 124 if ($OptionsInfo{DetailLevel} >=4 ) { | |
| 125 print "Line number $LineCount contains ", scalar(@InvalidColLabels)," non-numerical or empty value(s) for column(s) - ", JoinWords(\@InvalidColLabels, ", ", 0)," - to be analyzed: $Line \n"; | |
| 126 } | |
| 127 elsif ($OptionsInfo{DetailLevel} >= 3) { | |
| 128 print "Line number $LineCount contains ", scalar(@InvalidColLabels)," non-numerical or empty value(s) for column(s) - ", JoinWords(\@InvalidColLabels, ", ", 0)," - to be analyzed...\n"; | |
| 129 } | |
| 130 elsif ($OptionsInfo{DetailLevel} >= 2) { | |
| 131 print "Line number $LineCount contains ", scalar(@InvalidColLabels)," non-numerical or empty value(s) for columns to be analyzed...\n"; | |
| 132 } | |
| 133 } | |
| 134 } | |
| 135 if ($InvalidLineCount && ($OptionsInfo{DetailLevel} >= 1)) { | |
| 136 print "Non-numerical or empty data present in $InvalidLineCount line(s)...\n"; | |
| 137 } | |
| 138 close TEXTFILE; | |
| 139 | |
| 140 # Perform the analysis... | |
| 141 my(@SpecifiedFunctionNames, $SpecifiedFunction); | |
| 142 @SpecifiedFunctionNames = (); | |
| 143 | |
| 144 for $SpecifiedFunction (@{$OptionsInfo{SpecifiedStatisticalFunctions}}) { | |
| 145 if ($SpecifiedFunction !~ /^(Covariance|Correlation|Frequency|Rsquare|StandardScores|StandardScoresN)$/i) { | |
| 146 push @SpecifiedFunctionNames, $OptionsInfo{SpecifiedStatisticalFunctionsMap}{lc($SpecifiedFunction)}; | |
| 147 } | |
| 148 } | |
| 149 if (@SpecifiedFunctionNames) { | |
| 150 PerformAnalysis($Index, \@SpecifiedFunctionNames, \%ColValuesToAnalyzeMap) | |
| 151 } | |
| 152 if (exists($OptionsInfo{SpecifiedStatisticalFunctionsMap}{covariance}) || exists($OptionsInfo{SpecifiedStatisticalFunctionsMap}{correlation}) || exists($OptionsInfo{SpecifiedStatisticalFunctionsMap}{rsquare})) { | |
| 153 if ($OptionsInfo{AllColumnPairs}) { | |
| 154 PerformMatrixAnalysis($Index, \%ColValuesToAnalyzeMap); | |
| 155 } | |
| 156 else { | |
| 157 # Perform pairwise analysis for specified columns and write out calculated values - correlation | |
| 158 # rsquare, or covariance - in the same file. | |
| 159 PerformColumnPairAnalysis($Index, \%ColValuesToAnalyzeMap); | |
| 160 } | |
| 161 } | |
| 162 if (exists($OptionsInfo{SpecifiedStatisticalFunctionsMap}{standardscores}) || exists($OptionsInfo{SpecifiedStatisticalFunctionsMap}{standardscoresn}) ) { | |
| 163 PerformStandardScoresAnalysis($Index, \%ColValuesToAnalyzeMap); | |
| 164 } | |
| 165 if (exists($OptionsInfo{SpecifiedStatisticalFunctionsMap}{frequency})) { | |
| 166 PerformFrequencyAnalysis($Index, \%ColValuesToAnalyzeMap); | |
| 167 } | |
| 168 } | |
| 169 | |
| 170 # Calculate values for various statistical functions... | |
| 171 sub PerformAnalysis { | |
| 172 my($Index, $SpecifiedFunctionNamesRef, $ColValuesToAnalyzeMapRef) = @_; | |
| 173 my($NewTextFile, $Line, $SpecifiedFunction, $Label, @ColLabels, @ColNumsToAnalyze); | |
| 174 | |
| 175 $NewTextFile = $TextFilesInfo{OutFileRoot}[$Index] . $OptionsInfo{FileNameMode} . "." . $TextFilesInfo{OutFileExt}[$Index]; | |
| 176 | |
| 177 print "Generating new text file $NewTextFile...\n"; | |
| 178 open NEWTEXTFILE, ">$NewTextFile" or die "Error: Can't open $NewTextFile: $! \n"; | |
| 179 | |
| 180 # Write out column labels... | |
| 181 @ColLabels = (); | |
| 182 push @ColLabels, "ColumnID"; | |
| 183 for $SpecifiedFunction (@{$SpecifiedFunctionNamesRef}) { | |
| 184 $Label = $SpecifiedFunction; | |
| 185 if ($SpecifiedFunction =~ /^(KLargest|KSmallest)$/i) { | |
| 186 my($KthValue); | |
| 187 $KthValue = ($SpecifiedFunction =~ /^KLargest$/i) ? $OptionsInfo{KLargest} : $OptionsInfo{KSmallest}; | |
| 188 $Label = AddNumberSuffix($KthValue) . "$SpecifiedFunction"; | |
| 189 $Label =~ s/K//g; | |
| 190 } | |
| 191 elsif ($SpecifiedFunction =~ /^TrimMean$/i) { | |
| 192 $Label = "${SpecifiedFunction}($OptionsInfo{TrimFraction})"; | |
| 193 } | |
| 194 push @ColLabels, $Label; | |
| 195 } | |
| 196 $Line = JoinWords(\@ColLabels, $OptionsInfo{OutDelim}, $OptionsInfo{OutQuote}); | |
| 197 print NEWTEXTFILE "$Line\n"; | |
| 198 | |
| 199 # Go over each column to be analyzed... | |
| 200 @ColNumsToAnalyze = @{$TextFilesInfo{ColNumsToAnalyze}[$Index]}; | |
| 201 | |
| 202 # Turn off "strict"; otherwise, invoking statistical functions using function name string | |
| 203 # is problematic. | |
| 204 no strict; | |
| 205 | |
| 206 my($ColValuesRef, $ColNum, $Value, @RowValues, %CalculatedValues); | |
| 207 %CalculatedValues = (); | |
| 208 for $ColNum (@ColNumsToAnalyze) { | |
| 209 @RowValues = (); | |
| 210 # Setup column id... | |
| 211 push @RowValues, $TextFilesInfo{ColLabels}[$Index][$ColNum]; | |
| 212 $ColValuesRef = \@{$ColValuesToAnalyzeMapRef->{$ColNum}}; | |
| 213 FUNCTIONNAME: for $SpecifiedFunction (@{$SpecifiedFunctionNamesRef}) { | |
| 214 $Value = ""; | |
| 215 if (!@{$ColValuesToAnalyzeMapRef->{$ColNum}}) { | |
| 216 # Invalid column values... | |
| 217 push @RowValues, $Value; | |
| 218 next FUNCTIONNAME; | |
| 219 } | |
| 220 if ($SpecifiedFunction =~ /^Count$/i) { | |
| 221 $Value = @{$ColValuesToAnalyzeMapRef->{$ColNum}}; | |
| 222 } | |
| 223 elsif ($SpecifiedFunction =~ /^KLargest$/i) { | |
| 224 $Value = &$SpecifiedFunction($ColValuesRef, $OptionsInfo{KLargest}); | |
| 225 } | |
| 226 elsif ($SpecifiedFunction =~ /^KSmallest$/i) { | |
| 227 $Value = &$SpecifiedFunction($ColValuesRef, $OptionsInfo{KSmallest}); | |
| 228 } | |
| 229 elsif ($SpecifiedFunction =~ /^StandardDeviation$/i) { | |
| 230 if (exists($CalculatedValues{$ColNum}{StandardDeviation})) { | |
| 231 $Value = $CalculatedValues{$ColNum}{StandardDeviation}; | |
| 232 } | |
| 233 else { | |
| 234 $Value = &$SpecifiedFunction($ColValuesRef); | |
| 235 $CalculatedValues{$ColNum}{StandardDeviation} = $Value; | |
| 236 } | |
| 237 } | |
| 238 elsif ($SpecifiedFunction =~ /^StandardError$/i) { | |
| 239 if (!exists($CalculatedValues{$ColNum}{StandardDeviation})) { | |
| 240 $Value = StandardDeviation($ColValuesRef); | |
| 241 $CalculatedValues{$ColNum}{StandardDeviation} = $Value; | |
| 242 } | |
| 243 if (defined $CalculatedValues{$ColNum}{StandardDeviation}) { | |
| 244 $Value = &$SpecifiedFunction($CalculatedValues{$ColNum}{StandardDeviation}, @{$ColValuesToAnalyzeMapRef->{$ColNum}}); | |
| 245 } | |
| 246 } | |
| 247 elsif ($SpecifiedFunction =~ /^TrimMean$/i) { | |
| 248 $Value = &$SpecifiedFunction($ColValuesRef, $OptionsInfo{TrimFraction}); | |
| 249 } | |
| 250 else { | |
| 251 $Value = &$SpecifiedFunction($ColValuesRef); | |
| 252 } | |
| 253 # Format the output value. And add zero to get rid of tariling zeros... | |
| 254 $Value = (defined($Value) && length($Value)) ? (sprintf("%.$OptionsInfo{Precision}f", $Value) + 0) : ""; | |
| 255 push @RowValues, $Value; | |
| 256 } | |
| 257 $Line = JoinWords(\@RowValues, $OptionsInfo{OutDelim}, $OptionsInfo{OutQuote}); | |
| 258 print NEWTEXTFILE "$Line\n"; | |
| 259 } | |
| 260 close NEWTEXTFILE; | |
| 261 } | |
| 262 | |
| 263 # Calculate covariance, correlation, rsquare for specified column pairs.... | |
| 264 sub PerformColumnPairAnalysis { | |
| 265 my($Index, $ColValuesToAnalyzeMapRef) = @_; | |
| 266 my($NewTextFile, @ColLabels, $Line, $CalculateCorrelation, $CalculateRSquare, $CalculateCovariance); | |
| 267 $CalculateCorrelation = exists($OptionsInfo{SpecifiedStatisticalFunctionsMap}{correlation}) ? 1 : 0; | |
| 268 $CalculateRSquare = exists($OptionsInfo{SpecifiedStatisticalFunctionsMap}{rsquare}) ? 1 : 0; | |
| 269 $CalculateCovariance = exists($OptionsInfo{SpecifiedStatisticalFunctionsMap}{covariance}) ? 1 : 0; | |
| 270 | |
| 271 $NewTextFile = $TextFilesInfo{OutFileRoot}[$Index] . "ColumnPairsAnalysis." . $TextFilesInfo{OutFileExt}[$Index]; | |
| 272 print "Generating new text file $NewTextFile...\n"; | |
| 273 open NEWTEXTFILE, ">$NewTextFile" or die "Error: Can't open $NewTextFile: $! \n"; | |
| 274 | |
| 275 # Write out the column labels... | |
| 276 @ColLabels = (); | |
| 277 push @ColLabels, ("ColumnID1", "ColumnID2"); | |
| 278 if ($CalculateCorrelation || $CalculateRSquare) { | |
| 279 push @ColLabels, "Correlation"; | |
| 280 if ($CalculateRSquare) { | |
| 281 push @ColLabels, "RSquare"; | |
| 282 } | |
| 283 } | |
| 284 if ($CalculateCovariance) { | |
| 285 push @ColLabels, "Covariance"; | |
| 286 } | |
| 287 $Line = JoinWords(\@ColLabels, $OptionsInfo{OutDelim}, $OptionsInfo{OutQuote}); | |
| 288 print NEWTEXTFILE "$Line\n"; | |
| 289 | |
| 290 # Go over each column pair... | |
| 291 my($CorrelationValue, $RSquareValue, $CovarianceValue, $ColIndex, $ColNum1, $ColNum2, $ColValuesRef1, $ColValuesRef2, @ColPairs1ToAnalyze, @ColPairs2ToAnalyze, @RowValues, $Value); | |
| 292 | |
| 293 @ColPairs1ToAnalyze = @{$TextFilesInfo{ColPairs1ToAnalyze}[$Index]}; | |
| 294 @ColPairs2ToAnalyze = @{$TextFilesInfo{ColPairs2ToAnalyze}[$Index]}; | |
| 295 for $ColIndex (0 .. $#ColPairs1ToAnalyze) { | |
| 296 @RowValues = (); | |
| 297 $ColNum1 = $ColPairs1ToAnalyze[$ColIndex]; | |
| 298 $ColNum2 = $ColPairs2ToAnalyze[$ColIndex]; | |
| 299 $ColValuesRef1 = \@{$ColValuesToAnalyzeMapRef->{$ColNum1}}; | |
| 300 $ColValuesRef2 = \@{$ColValuesToAnalyzeMapRef->{$ColNum2}}; | |
| 301 | |
| 302 # Setup column ids... | |
| 303 push @RowValues, $TextFilesInfo{ColLabels}[$Index][$ColNum1]; | |
| 304 push @RowValues, $TextFilesInfo{ColLabels}[$Index][$ColNum2]; | |
| 305 | |
| 306 if (@$ColValuesRef1 != @$ColValuesRef2) { | |
| 307 # Print a warning... | |
| 308 warn "Warning: Skipping analysis for column pair $TextFilesInfo{ColLabels}[$Index][$ColNum1], $TextFilesInfo{ColLabels}[$Index][$ColNum2]: Number of valid data values must be same.\n"; | |
| 309 if ($CalculateCorrelation || $CalculateRSquare) { | |
| 310 push @RowValues, ""; | |
| 311 if ($CalculateRSquare) { | |
| 312 push @RowValues, ""; | |
| 313 } | |
| 314 } | |
| 315 if ($CalculateCovariance) { | |
| 316 push @RowValues, ""; | |
| 317 } | |
| 318 } | |
| 319 else { | |
| 320 # Calculate appropriate value... | |
| 321 if ($CalculateCorrelation || $CalculateRSquare) { | |
| 322 $CorrelationValue = Correlation($ColValuesRef1, $ColValuesRef2); | |
| 323 $Value = (defined($CorrelationValue) && length($CorrelationValue)) ? (sprintf("%.$OptionsInfo{Precision}f", $CorrelationValue) + 0) : ""; | |
| 324 push @RowValues, $Value; | |
| 325 if ($CalculateRSquare) { | |
| 326 $RSquareValue = (defined($CorrelationValue) && length($CorrelationValue)) ? ($CorrelationValue ** 2) : ""; | |
| 327 $Value = (length($RSquareValue)) ? (sprintf("%.$OptionsInfo{Precision}f", $RSquareValue) + 0) : ""; | |
| 328 push @RowValues, $Value; | |
| 329 } | |
| 330 } | |
| 331 if ($CalculateCovariance) { | |
| 332 $CovarianceValue = Covariance($ColValuesRef1, $ColValuesRef2); | |
| 333 $Value = (defined($CovarianceValue) && length($CovarianceValue)) ? (sprintf("%.$OptionsInfo{Precision}f", $CovarianceValue) + 0) : ""; | |
| 334 push @RowValues, $Value; | |
| 335 } | |
| 336 } | |
| 337 $Line = JoinWords(\@RowValues, $OptionsInfo{OutDelim}, $OptionsInfo{OutQuote}); | |
| 338 print NEWTEXTFILE "$Line\n"; | |
| 339 } | |
| 340 close NEWTEXTFILE; | |
| 341 } | |
| 342 | |
| 343 # Generate histogram numbers... | |
| 344 sub PerformFrequencyAnalysis { | |
| 345 my($Index, $ColValuesToAnalyzeMapRef) = @_; | |
| 346 my($NewTextFile, $ColLabel, @ColLabels, @RowValues, $Line, $ColNum, @ColNumsToAnalyze, $ColValuesRef, $BinValue, $FrequencyValue, $Value, %FrequencyMap); | |
| 347 | |
| 348 @ColNumsToAnalyze = @{$TextFilesInfo{ColNumsToAnalyze}[$Index]}; | |
| 349 for $ColNum (@ColNumsToAnalyze) { | |
| 350 $NewTextFile = $TextFilesInfo{OutFileRoot}[$Index] . $TextFilesInfo{ColLabels}[$Index][$ColNum] . "FrequencyAnalysis." . $TextFilesInfo{OutFileExt}[$Index]; | |
| 351 print "Generating new text file $NewTextFile...\n"; | |
| 352 open NEWTEXTFILE, ">$NewTextFile" or die "Error: Can't open $NewTextFile: $! \n"; | |
| 353 | |
| 354 # Write out the column labels... | |
| 355 @ColLabels = (); | |
| 356 push @ColLabels , ("Bins", "Frequency"); | |
| 357 $Line = JoinWords(\@ColLabels, $OptionsInfo{OutDelim}, $OptionsInfo{OutQuote}); | |
| 358 print NEWTEXTFILE "$Line\n"; | |
| 359 | |
| 360 #Calculate and write out frequency values... | |
| 361 %FrequencyMap = (); | |
| 362 $ColValuesRef = \@{$ColValuesToAnalyzeMapRef->{$ColNum}}; | |
| 363 if (@$ColValuesRef) { | |
| 364 if (@{$OptionsInfo{BinRange}}) { | |
| 365 %FrequencyMap = Frequency($ColValuesRef, \@{$OptionsInfo{BinRange}}); | |
| 366 } | |
| 367 else { | |
| 368 %FrequencyMap = Frequency($ColValuesRef, $OptionsInfo{NumOfBins}); | |
| 369 } | |
| 370 } | |
| 371 for $BinValue (sort { $a <=> $b } keys %FrequencyMap) { | |
| 372 $FrequencyValue = $FrequencyMap{$BinValue}; | |
| 373 | |
| 374 @RowValues = (); | |
| 375 $Value = (length($BinValue)) ? (sprintf("%.$OptionsInfo{Precision}f", $BinValue) + 0) : ""; | |
| 376 push @RowValues, $Value; | |
| 377 $Value = (length($FrequencyValue)) ? (sprintf("%.$OptionsInfo{Precision}f", $FrequencyValue) + 0) : ""; | |
| 378 push @RowValues, $Value; | |
| 379 | |
| 380 $Line = JoinWords(\@RowValues, $OptionsInfo{OutDelim}, $OptionsInfo{OutQuote}); | |
| 381 print NEWTEXTFILE "$Line\n"; | |
| 382 } | |
| 383 close NEWTEXTFILE; | |
| 384 } | |
| 385 } | |
| 386 | |
| 387 # Calculate covariance, correlation/rsquare matrices.... | |
| 388 sub PerformMatrixAnalysis { | |
| 389 my($Index, $ColValuesToAnalyzeMapRef) = @_; | |
| 390 my($CorrelationTextFile, $CovarianceTextFile, $RSquareTextFile, $CalculateCorrelation, $CalculateRSquare, $CalculateCovariance); | |
| 391 | |
| 392 $CalculateCorrelation = exists($OptionsInfo{SpecifiedStatisticalFunctionsMap}{correlation}) ? 1 : 0; | |
| 393 $CalculateRSquare = exists($OptionsInfo{SpecifiedStatisticalFunctionsMap}{rsquare}) ? 1 : 0; | |
| 394 $CalculateCovariance = exists($OptionsInfo{SpecifiedStatisticalFunctionsMap}{covariance}) ? 1 : 0; | |
| 395 | |
| 396 $CorrelationTextFile = $TextFilesInfo{OutFileRoot}[$Index] . "CorrelationMatrix." . $TextFilesInfo{OutFileExt}[$Index]; | |
| 397 $RSquareTextFile = $TextFilesInfo{OutFileRoot}[$Index] . "RSquareMatrix." . $TextFilesInfo{OutFileExt}[$Index]; | |
| 398 $CovarianceTextFile = $TextFilesInfo{OutFileRoot}[$Index] . "CovarianceMatrix." . $TextFilesInfo{OutFileExt}[$Index]; | |
| 399 | |
| 400 my($TextFilesList, $Delimiter); | |
| 401 $TextFilesList = ""; | |
| 402 if ($CalculateCorrelation || $CalculateRSquare) { | |
| 403 $TextFilesList = $CorrelationTextFile; | |
| 404 if ($CalculateRSquare) { | |
| 405 $TextFilesList .= ", $CorrelationTextFile"; | |
| 406 } | |
| 407 } | |
| 408 $Delimiter = length($TextFilesList) ? "," : ""; | |
| 409 if ($CalculateCovariance) { | |
| 410 $TextFilesList .= "${Delimiter} ${CorrelationTextFile}"; | |
| 411 } | |
| 412 if ($TextFilesList =~ /\,/) { | |
| 413 print "Generating new text files $TextFilesList...\n" | |
| 414 } | |
| 415 else { | |
| 416 print "Generating new text file $TextFilesList...\n" | |
| 417 } | |
| 418 if ($CalculateCorrelation || $CalculateRSquare) { | |
| 419 open CORRELATIONTEXTFILE, ">$CorrelationTextFile" or die "Error: Can't open $CorrelationTextFile: $! \n"; | |
| 420 if ($CalculateRSquare) { | |
| 421 open RSQUARETEXTFILE, ">$RSquareTextFile" or die "Error: Can't open $RSquareTextFile: $! \n"; | |
| 422 } | |
| 423 } | |
| 424 if ($CalculateCovariance) { | |
| 425 open COVARIANCETEXTFILE, ">$CovarianceTextFile" or die "Error: Can't open $CovarianceTextFile: $! \n"; | |
| 426 } | |
| 427 | |
| 428 my($Line, $Value, $CorrelationValue, $RSquareValue, $CovarianceValue, $ColNum, $ColNum1, $ColNum2, $ColValuesRef1, $ColValuesRef2, @ColLabels, @CovarianceRowValues, @CorrelationRowValues, @RSquareRowValues); | |
| 429 | |
| 430 # Write out the column labels... | |
| 431 @ColLabels = (); | |
| 432 push @ColLabels, ""; | |
| 433 for $ColNum (0 .. ($TextFilesInfo{ColCount}[$Index] - 1)) { | |
| 434 push @ColLabels, $TextFilesInfo{ColLabels}[$Index][$ColNum]; | |
| 435 } | |
| 436 $Line = JoinWords(\@ColLabels, $OptionsInfo{OutDelim}, $OptionsInfo{OutQuote}); | |
| 437 if ($CalculateCorrelation || $CalculateRSquare) { | |
| 438 print CORRELATIONTEXTFILE "$Line\n"; | |
| 439 if ($CalculateRSquare) { | |
| 440 print RSQUARETEXTFILE "$Line\n"; | |
| 441 } | |
| 442 } | |
| 443 if ($CalculateCovariance) { | |
| 444 print COVARIANCETEXTFILE "$Line\n"; | |
| 445 } | |
| 446 | |
| 447 # Due to symmetric nature of these matrices, only one half needs to be | |
| 448 # calculated. So, just calculate the lower half and copy it to upper half... | |
| 449 my(%CorrelationMatrixMap, %RSquareMatrixMap, %CovarianceMatrixMap); | |
| 450 | |
| 451 %CorrelationMatrixMap = (); %RSquareMatrixMap = (); %CovarianceMatrixMap = (); | |
| 452 for $ColNum1 (0 .. ($TextFilesInfo{ColCount}[$Index] - 1)) { | |
| 453 for $ColNum2 (0 .. $ColNum1) { | |
| 454 $ColValuesRef1 = \@{$ColValuesToAnalyzeMapRef->{$ColNum1}}; | |
| 455 $ColValuesRef2 = \@{$ColValuesToAnalyzeMapRef->{$ColNum2}}; | |
| 456 if ($CalculateCorrelation || $CalculateRSquare) { | |
| 457 $CorrelationValue = Correlation($ColValuesRef1, $ColValuesRef2); | |
| 458 $CorrelationValue = (defined($CorrelationValue) && length($CorrelationValue)) ? (sprintf("%.$OptionsInfo{Precision}f", $CorrelationValue) + 0) : ""; | |
| 459 $CorrelationMatrixMap{$ColNum1}{$ColNum2} = $CorrelationValue; | |
| 460 if ($ColNum1 != $ColNum2) { | |
| 461 $CorrelationMatrixMap{$ColNum2}{$ColNum1} = $CorrelationValue; | |
| 462 } | |
| 463 if ($CalculateRSquare) { | |
| 464 $RSquareValue = (defined($CorrelationValue) && length($CorrelationValue)) ? ($CorrelationValue ** 2) : ""; | |
| 465 $RSquareValue = (length($RSquareValue)) ? (sprintf("%.$OptionsInfo{Precision}f", $RSquareValue) + 0) : ""; | |
| 466 $RSquareMatrixMap{$ColNum1}{$ColNum2} = $RSquareValue; | |
| 467 if ($ColNum1 != $ColNum2) { | |
| 468 $RSquareMatrixMap{$ColNum2}{$ColNum1} = $RSquareValue; | |
| 469 } | |
| 470 } | |
| 471 } | |
| 472 if ($CalculateCovariance) { | |
| 473 $CovarianceValue = Covariance($ColValuesRef1, $ColValuesRef2); | |
| 474 $CovarianceValue = (defined($CovarianceValue) && length($CovarianceValue)) ? (sprintf("%.$OptionsInfo{Precision}f", $CovarianceValue) + 0) : ""; | |
| 475 $CovarianceMatrixMap{$ColNum1}{$ColNum2} = $CovarianceValue; | |
| 476 if ($ColNum1 != $ColNum2) { | |
| 477 $CovarianceMatrixMap{$ColNum2}{$ColNum1} = $CovarianceValue; | |
| 478 } | |
| 479 } | |
| 480 } | |
| 481 } | |
| 482 | |
| 483 # Write out the matrices... | |
| 484 for $ColNum1 (0 .. ($TextFilesInfo{ColCount}[$Index] - 1)) { | |
| 485 @CorrelationRowValues = (); | |
| 486 @RSquareRowValues = (); | |
| 487 @CovarianceRowValues = (); | |
| 488 if ($CalculateCorrelation || $CalculateRSquare) { | |
| 489 push @CorrelationRowValues, $TextFilesInfo{ColLabels}[$Index][$ColNum1]; | |
| 490 if ($CalculateRSquare) { | |
| 491 push @RSquareRowValues, $TextFilesInfo{ColLabels}[$Index][$ColNum1]; | |
| 492 } | |
| 493 } | |
| 494 if ($CalculateCovariance) { | |
| 495 push @CovarianceRowValues, $TextFilesInfo{ColLabels}[$Index][$ColNum1]; | |
| 496 } | |
| 497 for $ColNum2 (0 .. ($TextFilesInfo{ColCount}[$Index] - 1)) { | |
| 498 if ($CalculateCorrelation || $CalculateRSquare) { | |
| 499 push @CorrelationRowValues, $CorrelationMatrixMap{$ColNum1}{$ColNum2}; | |
| 500 if ($CalculateRSquare) { | |
| 501 push @RSquareRowValues, $RSquareMatrixMap{$ColNum1}{$ColNum2}; | |
| 502 } | |
| 503 } | |
| 504 if ($CalculateCovariance) { | |
| 505 push @CovarianceRowValues, $CovarianceMatrixMap{$ColNum1}{$ColNum2}; | |
| 506 } | |
| 507 } | |
| 508 if ($CalculateCorrelation || $CalculateRSquare) { | |
| 509 $Line = JoinWords(\@CorrelationRowValues, $OptionsInfo{OutDelim}, $OptionsInfo{OutQuote}); | |
| 510 print CORRELATIONTEXTFILE "$Line\n"; | |
| 511 if ($CalculateRSquare) { | |
| 512 $Line = JoinWords(\@RSquareRowValues, $OptionsInfo{OutDelim}, $OptionsInfo{OutQuote}); | |
| 513 print RSQUARETEXTFILE "$Line\n"; | |
| 514 } | |
| 515 } | |
| 516 if ($CalculateCovariance) { | |
| 517 $Line = JoinWords(\@CovarianceRowValues, $OptionsInfo{OutDelim}, $OptionsInfo{OutQuote}); | |
| 518 print COVARIANCETEXTFILE "$Line\n"; | |
| 519 } | |
| 520 } | |
| 521 if ($CalculateCorrelation || $CalculateRSquare) { | |
| 522 close CORRELATIONTEXTFILE; | |
| 523 if ($CalculateRSquare) { | |
| 524 close RSQUARETEXTFILE; | |
| 525 } | |
| 526 } | |
| 527 if ($CalculateCovariance) { | |
| 528 close COVARIANCETEXTFILE; | |
| 529 } | |
| 530 } | |
| 531 | |
| 532 # Calculate standard scores... | |
| 533 sub PerformStandardScoresAnalysis { | |
| 534 my($Index, $ColValuesToAnalyzeMapRef) = @_; | |
| 535 my($StandardScores, $StandardScoresN, $NewTextFile, @ColLabels, $Label, $NewLine); | |
| 536 | |
| 537 $StandardScores = exists($OptionsInfo{SpecifiedStatisticalFunctionsMap}{standardscores}) ? 1 : 0; | |
| 538 $StandardScoresN = exists($OptionsInfo{SpecifiedStatisticalFunctionsMap}{standardscoresn}) ? 1 : 0; | |
| 539 | |
| 540 $NewTextFile = $TextFilesInfo{OutFileRoot}[$Index] . "StandardScores." . $TextFilesInfo{OutFileExt}[$Index]; | |
| 541 print "Generating new text file $NewTextFile...\n"; | |
| 542 open NEWTEXTFILE, ">$NewTextFile" or die "Error: Can't open $NewTextFile: $! \n"; | |
| 543 | |
| 544 my($ColValuesRef, $ColNum, @ColNumsToAnalyze); | |
| 545 # Write out column labels... | |
| 546 @ColLabels = (); | |
| 547 @ColNumsToAnalyze = @{$TextFilesInfo{ColNumsToAnalyze}[$Index]}; | |
| 548 for $ColNum (@ColNumsToAnalyze) { | |
| 549 $Label = $TextFilesInfo{ColLabels}[$Index][$ColNum]; | |
| 550 if ($StandardScores) { | |
| 551 push @ColLabels, "${Label}\(StandardScores)"; | |
| 552 } | |
| 553 if ($StandardScoresN) { | |
| 554 push @ColLabels, "${Label}\(StandardScoresN)"; | |
| 555 } | |
| 556 } | |
| 557 $NewLine = JoinWords(\@ColLabels, $OptionsInfo{OutDelim}, $OptionsInfo{OutQuote}); | |
| 558 print NEWTEXTFILE "$NewLine\n"; | |
| 559 | |
| 560 # Go over each column to be analyzed and calculate standard deviation | |
| 561 # and mean values... | |
| 562 my(%StandardDeviationMap, %StandardDeviationNMap, %MeanMap); | |
| 563 %StandardDeviationMap = (); | |
| 564 %StandardDeviationNMap = (); | |
| 565 %MeanMap = (); | |
| 566 for $ColNum (@ColNumsToAnalyze) { | |
| 567 $ColValuesRef = \@{$ColValuesToAnalyzeMapRef->{$ColNum}}; | |
| 568 if (!exists($MeanMap{$ColNum})) { | |
| 569 $MeanMap{$ColNum} = Mean($ColValuesRef); | |
| 570 } | |
| 571 if ($StandardScores) { | |
| 572 if (!exists($StandardDeviationMap{$ColNum})) { | |
| 573 $StandardDeviationMap{$ColNum} = StandardDeviation($ColValuesRef); | |
| 574 } | |
| 575 } | |
| 576 if ($StandardScoresN) { | |
| 577 if (!exists($StandardDeviationNMap{$ColNum})) { | |
| 578 $StandardDeviationNMap{$ColNum} = StandardDeviationN($ColValuesRef); | |
| 579 } | |
| 580 } | |
| 581 } | |
| 582 # | |
| 583 # Go over each row and calculate standard scores for each column | |
| 584 # using (x[i] - mean) / (n - 1) for StandardScores and (x[i] - mean) / n | |
| 585 # for StandardScoresN; write out the calculated values as well... | |
| 586 | |
| 587 my($TextFile, $InDelim, $Line, $Value, $ValueOkay, $ScoreValue, @RowValues, @LineWords); | |
| 588 $TextFile = $TextFilesList[$Index]; | |
| 589 $InDelim = $TextFilesInfo{InDelim}[$Index]; | |
| 590 | |
| 591 open TEXTFILE, "$TextFile" or die "Error: Can't open $TextFile: $! \n"; | |
| 592 $Line = GetTextLine(\*TEXTFILE); | |
| 593 while ($Line = GetTextLine(\*TEXTFILE)) { | |
| 594 @LineWords = quotewords($InDelim, 0, $Line); | |
| 595 @RowValues = (); | |
| 596 COLNUM: for $ColNum (@ColNumsToAnalyze) { | |
| 597 $Value = $LineWords[$ColNum]; | |
| 598 $ValueOkay = ($OptionsInfo{CheckData} && !IsNumerical($Value)) ? 0 : 1; | |
| 599 if ($StandardScores) { | |
| 600 $ScoreValue = $ValueOkay ? (($Value - $MeanMap{$ColNum})/$StandardDeviationMap{$ColNum}) : ""; | |
| 601 $ScoreValue = (defined($ScoreValue) && length($ScoreValue)) ? (sprintf("%.$OptionsInfo{Precision}f", $ScoreValue) + 0) : ""; | |
| 602 push @RowValues, $ScoreValue; | |
| 603 } | |
| 604 if ($StandardScoresN) { | |
| 605 $ScoreValue = $ValueOkay ? (($Value - $MeanMap{$ColNum})/$StandardDeviationNMap{$ColNum}) : ""; | |
| 606 $ScoreValue = (defined($ScoreValue) && length($ScoreValue)) ? (sprintf("%.$OptionsInfo{Precision}f", $ScoreValue) + 0) : ""; | |
| 607 push @RowValues, $ScoreValue; | |
| 608 } | |
| 609 } | |
| 610 $NewLine = JoinWords(\@RowValues, $OptionsInfo{OutDelim}, $OptionsInfo{OutQuote}); | |
| 611 print NEWTEXTFILE "$NewLine\n"; | |
| 612 } | |
| 613 close TEXTFILE; | |
| 614 close NEWTEXTFILE; | |
| 615 } | |
| 616 | |
| 617 # Make sure the specified columns exists in text files... | |
| 618 sub ProcessColumnsInfo { | |
| 619 my($Index, $TextFile, $ColNum, $NewColNum, $ColIndex, @ColNumsToAnalyze, %UniqueColNumsToAnalyzeMap); | |
| 620 | |
| 621 @{$TextFilesInfo{ColNumsToAnalyze}} = (); | |
| 622 @{$TextFilesInfo{ColPairs1ToAnalyze}} = (); | |
| 623 @{$TextFilesInfo{ColPairs2ToAnalyze}} = (); | |
| 624 @{$TextFilesInfo{UniqueColNumsToAnalyze}} = (); | |
| 625 | |
| 626 FILELIST: for $Index (0 .. $#TextFilesList) { | |
| 627 $TextFile = $TextFilesList[$Index]; | |
| 628 | |
| 629 @{$TextFilesInfo{ColNumsToAnalyze}[$Index]} = (); | |
| 630 @{$TextFilesInfo{ColPairs1ToAnalyze}[$Index]} = (); | |
| 631 @{$TextFilesInfo{ColPairs2ToAnalyze}[$Index]} = (); | |
| 632 @{$TextFilesInfo{UniqueColNumsToAnalyze}[$Index]} = (); | |
| 633 | |
| 634 %UniqueColNumsToAnalyzeMap = (); | |
| 635 | |
| 636 if ($TextFilesInfo{FileOkay}[$Index]) { | |
| 637 @ColNumsToAnalyze = (); | |
| 638 if (@{$OptionsInfo{SpecifiedColumns}}) { | |
| 639 if ($OptionsInfo{ColMode} =~ /^colnum$/i) { | |
| 640 for $ColNum (@{$OptionsInfo{SpecifiedColumns}}) { | |
| 641 if ($ColNum >=1 && $ColNum <= $TextFilesInfo{ColCount}[$Index]) { | |
| 642 $NewColNum = $ColNum -1; | |
| 643 push @ColNumsToAnalyze, $NewColNum; | |
| 644 } | |
| 645 } | |
| 646 } | |
| 647 else { | |
| 648 my($ColLabel); | |
| 649 for $ColLabel (@{$OptionsInfo{SpecifiedColumns}}) { | |
| 650 if (exists($TextFilesInfo{ColLabelToNumMap}[$Index]{$ColLabel})) { | |
| 651 push @ColNumsToAnalyze, $TextFilesInfo{ColLabelToNumMap}[$Index]{$ColLabel}; | |
| 652 } | |
| 653 } | |
| 654 } | |
| 655 } | |
| 656 elsif (defined $OptionsInfo{Columns} && $OptionsInfo{Columns} =~ /^All$/i) { | |
| 657 for $ColNum (0 .. ($TextFilesInfo{ColCount}[$Index] - 1)) { | |
| 658 push @ColNumsToAnalyze, $ColNum; | |
| 659 } | |
| 660 } | |
| 661 else { | |
| 662 push @ColNumsToAnalyze, 0; | |
| 663 } | |
| 664 if (@ColNumsToAnalyze) { | |
| 665 push @{$TextFilesInfo{ColNumsToAnalyze}[$Index]}, @ColNumsToAnalyze; | |
| 666 # Set up unique columns map as well... | |
| 667 for $ColNum (@ColNumsToAnalyze) { | |
| 668 if (!exists $UniqueColNumsToAnalyzeMap{$ColNum}) { | |
| 669 $UniqueColNumsToAnalyzeMap{$ColNum} = $ColNum; | |
| 670 } | |
| 671 } | |
| 672 } | |
| 673 else { | |
| 674 warn "Warning: Ignoring file $TextFile: None of the columns specified, @{$OptionsInfo{SpecifiedColumns}}, using \"--columns\" option exist.\n"; | |
| 675 $TextFilesInfo{FileOkay}[$Index] = 0; | |
| 676 next FILELIST; | |
| 677 } | |
| 678 if (!$OptionsInfo{Overwrite} && exists($OptionsInfo{SpecifiedStatisticalFunctionsMap}{frequency})) { | |
| 679 # Make sure specific frequency files don't exist... | |
| 680 my($FrequencyFile); | |
| 681 for $ColNum (@ColNumsToAnalyze) { | |
| 682 $FrequencyFile = $TextFilesInfo{OutFileRoot}[$Index] . $TextFilesInfo{ColLabels}[$Index][$ColNum] . "FrequencyAnalysis." . $TextFilesInfo{OutFileExt}[$Index]; | |
| 683 if (-e $FrequencyFile) { | |
| 684 warn "Warning: Ignoring file $TextFile: The file $FrequencyFile already exists.\n"; | |
| 685 $TextFilesInfo{FileOkay}[$Index] = 0; | |
| 686 next FILELIST; | |
| 687 } | |
| 688 } | |
| 689 } | |
| 690 # Setup specified column pairs... | |
| 691 if (exists $OptionsInfo{SpecifiedStatisticalFunctionsMap}{correlation} || exists $OptionsInfo{SpecifiedStatisticalFunctionsMap}{covariance} || exists $OptionsInfo{SpecifiedStatisticalFunctionsMap}{rsquare}) { | |
| 692 my(@ColPairsToAnalyze, $ColNum1, $ColNum2); | |
| 693 if (@{$OptionsInfo{SpecifiedColumnPairs}}) { | |
| 694 # Make sure both columns exist... | |
| 695 if ($OptionsInfo{ColMode} =~ /^colnum$/i) { | |
| 696 for ($ColIndex = 0; (($ColIndex + 1) < @{$OptionsInfo{SpecifiedColumnPairs}}); $ColIndex += 2 ) { | |
| 697 $ColNum1 = $OptionsInfo{SpecifiedColumnPairs}[$ColIndex]; | |
| 698 $ColNum2 = $OptionsInfo{SpecifiedColumnPairs}[$ColIndex + 1]; | |
| 699 if ($ColNum1 >=1 && $ColNum1 <= $TextFilesInfo{ColCount}[$Index] && $ColNum2 >=1 && $ColNum2 <= $TextFilesInfo{ColCount}[$Index]) { | |
| 700 $ColNum1 -= 1; | |
| 701 $ColNum2 -= 1; | |
| 702 push @ColPairsToAnalyze, ($ColNum1, $ColNum2); | |
| 703 } | |
| 704 } | |
| 705 } | |
| 706 else { | |
| 707 my($ColLabel1, $ColLabel2); | |
| 708 for ($ColIndex = 0; (($ColIndex + 1) < @{$OptionsInfo{SpecifiedColumnPairs}}); $ColIndex += 2 ) { | |
| 709 $ColLabel1 = $OptionsInfo{SpecifiedColumnPairs}[$ColIndex]; | |
| 710 $ColLabel2 = $OptionsInfo{SpecifiedColumnPairs}[$ColIndex + 1]; | |
| 711 if (exists($TextFilesInfo{ColLabelToNumMap}[$Index]{$ColLabel1}) && exists($TextFilesInfo{ColLabelToNumMap}[$Index]{$ColLabel2})) { | |
| 712 $ColNum1 = $TextFilesInfo{ColLabelToNumMap}[$Index]{$ColLabel1}; | |
| 713 $ColNum2 = $TextFilesInfo{ColLabelToNumMap}[$Index]{$ColLabel2}; | |
| 714 push @ColPairsToAnalyze, ($ColNum1, $ColNum2); | |
| 715 } | |
| 716 } | |
| 717 } | |
| 718 } | |
| 719 elsif ($OptionsInfo{AllColumnPairs}) { | |
| 720 for $ColNum1 (0 .. ($TextFilesInfo{ColCount}[$Index] - 1)) { | |
| 721 for $ColNum2 (0 .. ($TextFilesInfo{ColCount}[$Index] - 1)) { | |
| 722 push @ColPairsToAnalyze, ($ColNum1, $ColNum2); | |
| 723 } | |
| 724 } | |
| 725 } | |
| 726 else { | |
| 727 if ($TextFilesInfo{ColCount}[$Index] >= 2) { | |
| 728 push @ColPairsToAnalyze, (0,1); | |
| 729 } | |
| 730 } | |
| 731 if (@ColPairsToAnalyze) { | |
| 732 if (@ColPairsToAnalyze % 2) { | |
| 733 warn "Warning: Ignoring file $TextFile: Invalid number values specified using \"--columnpairs\" option: It must contain even number of valid values.\n"; | |
| 734 $TextFilesInfo{FileOkay}[$Index] = 0; | |
| 735 next FILELIST; | |
| 736 } | |
| 737 else { | |
| 738 for ($ColIndex = 0; $ColIndex < @ColPairsToAnalyze; $ColIndex += 2) { | |
| 739 push @{$TextFilesInfo{ColPairs1ToAnalyze}[$Index]}, $ColPairsToAnalyze[$ColIndex]; | |
| 740 push @{$TextFilesInfo{ColPairs2ToAnalyze}[$Index]}, $ColPairsToAnalyze[$ColIndex + 1]; | |
| 741 } | |
| 742 # Set up unique columns map as well... | |
| 743 for $ColNum (@ColPairsToAnalyze) { | |
| 744 if (!exists $UniqueColNumsToAnalyzeMap{$ColNum}) { | |
| 745 $UniqueColNumsToAnalyzeMap{$ColNum} = $ColNum; | |
| 746 } | |
| 747 } | |
| 748 } | |
| 749 } | |
| 750 } | |
| 751 # Setup uniques columns array... | |
| 752 push @{$TextFilesInfo{UniqueColNumsToAnalyze}[$Index]}, (sort keys %UniqueColNumsToAnalyzeMap); | |
| 753 } | |
| 754 } | |
| 755 } | |
| 756 | |
| 757 # Retrieve information about input text files... | |
| 758 sub RetrieveTextFilesInfo { | |
| 759 my($Index, $TextFile, $FileDir, $FileName, $FileExt, $InDelim, $Line, @ColLabels, $OutFileRoot, $OutFile, $OutFileExt, $ColNum, $ColLabel); | |
| 760 | |
| 761 %TextFilesInfo = (); | |
| 762 | |
| 763 @{$TextFilesInfo{FileOkay}} = (); | |
| 764 @{$TextFilesInfo{ColCount}} = (); | |
| 765 @{$TextFilesInfo{ColLabels}} = (); | |
| 766 @{$TextFilesInfo{ColLabelToNumMap}} = (); | |
| 767 @{$TextFilesInfo{InDelim}} = (); | |
| 768 @{$TextFilesInfo{OutFileRoot}} = (); | |
| 769 @{$TextFilesInfo{OutFileExt}} = (); | |
| 770 | |
| 771 FILELIST: for $Index (0 .. $#TextFilesList) { | |
| 772 $TextFile = $TextFilesList[$Index]; | |
| 773 | |
| 774 $TextFilesInfo{FileOkay}[$Index] = 0; | |
| 775 $TextFilesInfo{ColCount}[$Index] = 0; | |
| 776 $TextFilesInfo{InDelim}[$Index] = ""; | |
| 777 $TextFilesInfo{OutFileRoot}[$Index] = ""; | |
| 778 $TextFilesInfo{OutFileExt}[$Index] = ""; | |
| 779 | |
| 780 @{$TextFilesInfo{ColLabels}[$Index]} = (); | |
| 781 %{$TextFilesInfo{ColLabelToNumMap}[$Index]} = (); | |
| 782 | |
| 783 if (!(-e $TextFile)) { | |
| 784 warn "Warning: Ignoring file $TextFile: It doesn't exist\n"; | |
| 785 next FILELIST; | |
| 786 } | |
| 787 if (!CheckFileType($TextFile, "csv tsv")) { | |
| 788 warn "Warning: Ignoring file $TextFile: It's not a csv or tsv file\n"; | |
| 789 next FILELIST; | |
| 790 } | |
| 791 ($FileDir, $FileName, $FileExt) = ParseFileName($TextFile); | |
| 792 if ($FileExt =~ /^tsv$/i) { | |
| 793 $InDelim = "\t"; | |
| 794 } | |
| 795 else { | |
| 796 $InDelim = "\,"; | |
| 797 if ($Options{indelim} !~ /^(comma|semicolon)$/i) { | |
| 798 warn "Warning: Ignoring file $TextFile: The value specified, $Options{indelim}, for option \"--indelim\" is not valid for csv files\n"; | |
| 799 next FILELIST; | |
| 800 } | |
| 801 if ($Options{indelim} =~ /^semicolon$/i) { | |
| 802 $InDelim = "\;"; | |
| 803 } | |
| 804 } | |
| 805 | |
| 806 if (!open TEXTFILE, "$TextFile") { | |
| 807 warn "Warning: Ignoring file $TextFile: Couldn't open it: $! \n"; | |
| 808 next FILELIST; | |
| 809 } | |
| 810 | |
| 811 $Line = GetTextLine(\*TEXTFILE); | |
| 812 @ColLabels = quotewords($InDelim, 0, $Line); | |
| 813 close TEXTFILE; | |
| 814 | |
| 815 $FileDir = ""; $FileName = ""; $FileExt = ""; | |
| 816 ($FileDir, $FileName, $FileExt) = ParseFileName($TextFile); | |
| 817 $FileExt = "csv"; | |
| 818 if ($Options{outdelim} =~ /^tab$/i) { | |
| 819 $FileExt = "tsv"; | |
| 820 } | |
| 821 $OutFileExt = $FileExt; | |
| 822 if ($Options{root} && (@TextFilesList == 1)) { | |
| 823 my ($RootFileDir, $RootFileName, $RootFileExt) = ParseFileName($Options{root}); | |
| 824 if ($RootFileName && $RootFileExt) { | |
| 825 $FileName = $RootFileName; | |
| 826 } | |
| 827 else { | |
| 828 $FileName = $Options{root}; | |
| 829 } | |
| 830 $OutFileRoot = $FileName; | |
| 831 } | |
| 832 else { | |
| 833 $OutFileRoot = $FileName; | |
| 834 } | |
| 835 $OutFile = $OutFileRoot . $OptionsInfo{FileNameMode} . ".$OutFileExt"; | |
| 836 | |
| 837 if (lc($OutFile) eq lc($TextFile)) { | |
| 838 warn "Warning: Ignoring file $TextFile:Output file name, $OutFile, is same as input text file name, $TextFile\n"; | |
| 839 next FILELIST; | |
| 840 } | |
| 841 if (!$Options{overwrite}) { | |
| 842 if (-e $OutFile) { | |
| 843 warn "Warning: Ignoring file $TextFile: The file $OutFile already exists\n"; | |
| 844 next FILELIST; | |
| 845 } | |
| 846 if (exists($OptionsInfo{SpecifiedStatisticalFunctionsMap}{covariance}) || exists($OptionsInfo{SpecifiedStatisticalFunctionsMap}{correlation}) || exists($OptionsInfo{SpecifiedStatisticalFunctionsMap}{rsquare})) { | |
| 847 if ($OptionsInfo{AllColumnPairs}) { | |
| 848 if (exists($OptionsInfo{SpecifiedStatisticalFunctionsMap}{covariance}) && (-e "${OutFileRoot}CovarianceMatrix.${FileExt}")) { | |
| 849 warn "Warning: Ignoring file $TextFile: The file ${OutFileRoot}Covariance.${FileExt} already exists.\n"; | |
| 850 next FILELIST; | |
| 851 } | |
| 852 if (exists($OptionsInfo{SpecifiedStatisticalFunctionsMap}{correlation}) && (-e "${OutFileRoot}CorrelationMatrix.${FileExt}")) { | |
| 853 warn "Warning: Ignoring file $TextFile: The file ${OutFileRoot}CorrelationMatrix.${FileExt} already exists.\n"; | |
| 854 next FILELIST; | |
| 855 } | |
| 856 if (exists($OptionsInfo{SpecifiedStatisticalFunctionsMap}{rsquare}) && (-e "${OutFileRoot}RSquareMatrix.${FileExt}")) { | |
| 857 warn "Warning: Ignoring file $TextFile: The file ${OutFileRoot}RSquareMatrix.${FileExt} already exists.\n"; | |
| 858 next FILELIST; | |
| 859 } | |
| 860 } | |
| 861 else { | |
| 862 if (-e "${OutFileRoot}ColumnPairsAnalysis.${FileExt}") { | |
| 863 warn "Warning: Ignoring file $TextFile: The file ${OutFileRoot}ColumnPairsAnalysis.${FileExt} already exists.\n"; | |
| 864 next FILELIST; | |
| 865 } | |
| 866 } | |
| 867 } | |
| 868 if (exists($OptionsInfo{SpecifiedStatisticalFunctionsMap}{standardscores}) && (-e "${OutFileRoot}StandardScores.${FileExt}")) { | |
| 869 warn "Warning: Ignoring file $TextFile: The file ${OutFileRoot}StandardScores.${FileExt} already exists.\n"; | |
| 870 next FILELIST; | |
| 871 } | |
| 872 } | |
| 873 | |
| 874 $TextFilesInfo{FileOkay}[$Index] = 1; | |
| 875 $TextFilesInfo{InDelim}[$Index] = $InDelim; | |
| 876 $TextFilesInfo{OutFileRoot}[$Index] = "$OutFileRoot"; | |
| 877 $TextFilesInfo{OutFileExt}[$Index] = "$OutFileExt"; | |
| 878 | |
| 879 $TextFilesInfo{ColCount}[$Index] = @ColLabels; | |
| 880 push @{$TextFilesInfo{ColLabels}[$Index]}, @ColLabels; | |
| 881 for $ColNum (0 .. $#ColLabels) { | |
| 882 $ColLabel = $ColLabels[$ColNum]; | |
| 883 $TextFilesInfo{ColLabelToNumMap}[$Index]{$ColLabel} = $ColNum; | |
| 884 } | |
| 885 } | |
| 886 } | |
| 887 | |
| 888 # Process option values... | |
| 889 sub ProcessOptions { | |
| 890 %OptionsInfo = (); | |
| 891 | |
| 892 $OptionsInfo{Mode} = $Options{mode}; | |
| 893 | |
| 894 $OptionsInfo{DetailLevel} = $Options{detail}; | |
| 895 | |
| 896 # Setup supported statistical functions... | |
| 897 my($SupportedFunction, @SupportedStatisticaFunctions, %SupportedStatisticaFunctionsMap); | |
| 898 %SupportedStatisticaFunctionsMap = (); | |
| 899 @SupportedStatisticaFunctions = qw(Average AverageDeviation Correlation Count Covariance GeometricMean Frequency HarmonicMean KLargest KSmallest Kurtosis Maximum Minimum Mean Median Mode RSquare Skewness Sum SumOfSquares StandardDeviation StandardDeviationN StandardError StandardScores StandardScoresN TrimMean Variance VarianceN); | |
| 900 | |
| 901 for $SupportedFunction (@SupportedStatisticaFunctions) { | |
| 902 $SupportedStatisticaFunctionsMap{lc($SupportedFunction)} = $SupportedFunction; | |
| 903 } | |
| 904 | |
| 905 # Setup a list of functions to use for analysis... | |
| 906 my($SpecifiedFunction); | |
| 907 %{$OptionsInfo{SpecifiedStatisticalFunctionsMap}} = (); | |
| 908 @{$OptionsInfo{SpecifiedStatisticalFunctions}} = (); | |
| 909 # Check mode values... | |
| 910 if ($Options{mode} =~ /^DescriptiveStatisticsBasic$/i ) { | |
| 911 $OptionsInfo{FileNameMode} = "DescriptiveStatisticsBasic"; | |
| 912 @{$OptionsInfo{SpecifiedStatisticalFunctions}} = qw(Count Maximum Minimum Mean Median StandardDeviation StandardError Variance Sum); | |
| 913 } | |
| 914 elsif ($Options{mode} =~ /^DescriptiveStatisticsAll$/i ) { | |
| 915 $OptionsInfo{FileNameMode} = "DescriptiveStatisticsAll"; | |
| 916 @{$OptionsInfo{SpecifiedStatisticalFunctions}} = qw(Count Maximum Minimum Mean GeometricMean HarmonicMean TrimMean Median Mode StandardDeviation Kurtosis Skewness StandardError Variance RSquare Frequency KLargest KSmallest Sum); | |
| 917 } | |
| 918 elsif ($Options{mode} =~ /^All$/i ) { | |
| 919 $OptionsInfo{FileNameMode} = "AllStatistics"; | |
| 920 @{$OptionsInfo{SpecifiedStatisticalFunctions}} = @SupportedStatisticaFunctions; | |
| 921 } | |
| 922 else { | |
| 923 $OptionsInfo{FileNameMode} = "SpecifiedStatistics"; | |
| 924 # Comma delimited list of functions... | |
| 925 my($Mode, @SpecifiedFunctions, @UnsupportedSpecifiedFunctions); | |
| 926 $Mode = $Options{mode}; | |
| 927 $Mode =~ s/ //g; | |
| 928 @SpecifiedFunctions = split ",", $Mode; | |
| 929 @UnsupportedSpecifiedFunctions = (); | |
| 930 for $SpecifiedFunction (@SpecifiedFunctions) { | |
| 931 if (exists($SupportedStatisticaFunctionsMap{lc($SpecifiedFunction)})) { | |
| 932 push @{$OptionsInfo{SpecifiedStatisticalFunctions}}, $SpecifiedFunction; | |
| 933 } | |
| 934 else { | |
| 935 push @UnsupportedSpecifiedFunctions, $SpecifiedFunction; | |
| 936 } | |
| 937 } | |
| 938 if (@UnsupportedSpecifiedFunctions) { | |
| 939 if (@UnsupportedSpecifiedFunctions > 1) { | |
| 940 warn "Error: The values specified - ", JoinWords(\@UnsupportedSpecifiedFunctions, ", ", 0)," - for option \"-m --mode\" are not valid.\n"; | |
| 941 } | |
| 942 else { | |
| 943 warn "Error: The value specified, @UnsupportedSpecifiedFunctions , for option \"-m --mode\" is not valid.\n"; | |
| 944 } | |
| 945 die "Allowed values:", JoinWords(\@SupportedStatisticaFunctions, ", ", 0), "\n"; | |
| 946 } | |
| 947 } | |
| 948 FUNCTION: for $SpecifiedFunction (@{$OptionsInfo{SpecifiedStatisticalFunctions}}) { | |
| 949 if (exists $OptionsInfo{SpecifiedStatisticalFunctionsMap}{lc($SpecifiedFunction)} ) { | |
| 950 next FUNCTION; | |
| 951 } | |
| 952 $OptionsInfo{SpecifiedStatisticalFunctionsMap}{lc($SpecifiedFunction)} = $SupportedStatisticaFunctionsMap{lc($SpecifiedFunction)}; | |
| 953 } | |
| 954 | |
| 955 $OptionsInfo{OutDelim} = ($Options{outdelim} =~ /tab/i ) ? "\t" : (($Options{outdelim} =~ /semicolon/i) ? "\;" : "\,"); | |
| 956 $OptionsInfo{OutQuote} = ($Options{quote} =~ /yes/i ) ? 1 : 0; | |
| 957 | |
| 958 $OptionsInfo{Overwrite} = defined $Options{overwrite} ? $Options{overwrite} : undef; | |
| 959 $OptionsInfo{Root} = defined $Options{root} ? $Options{root} : undef; | |
| 960 | |
| 961 $OptionsInfo{CheckData} = $Options{fast} ? 0 : 1; | |
| 962 $OptionsInfo{Precision} = $Options{precision}; | |
| 963 | |
| 964 $OptionsInfo{KLargest} = $Options{klargest}; | |
| 965 $OptionsInfo{KSmallest} = $Options{ksmallest}; | |
| 966 | |
| 967 $OptionsInfo{TrimFraction} = $Options{trimfraction}; | |
| 968 | |
| 969 # Setup frequency bin values... | |
| 970 $OptionsInfo{NumOfBins} = 10; | |
| 971 @{$OptionsInfo{BinRange}} = (); | |
| 972 if ($Options{frequencybins} =~ /\,/) { | |
| 973 my($BinValue, @SpecifiedBinRange); | |
| 974 @SpecifiedBinRange = split /\,/, $Options{frequencybins}; | |
| 975 if (@SpecifiedBinRange < 2) { | |
| 976 die "Error: The value specified, $Options{frequencybins}, for option \"--frequencybins\" is not valid: Must contain at least two values. \n"; | |
| 977 } | |
| 978 for $BinValue (@SpecifiedBinRange) { | |
| 979 if (!IsNumerical($BinValue)) { | |
| 980 die "Error: The value specified, $Options{frequencybins}, for option \"--frequencybins\" is not valid: Contains non numeric values. \n"; | |
| 981 } | |
| 982 } | |
| 983 my($Index1, $Index2); | |
| 984 for $Index1 (0 .. $#SpecifiedBinRange) { | |
| 985 for $Index2 (($Index1 + 1) .. $#SpecifiedBinRange) { | |
| 986 if ($SpecifiedBinRange[$Index1] >= $SpecifiedBinRange[$Index2]) { | |
| 987 die "Error: The value specified, $Options{frequencybins}, for option \"--frequencybins\" is not valid: Must contain values in ascending order. \n"; | |
| 988 } | |
| 989 } | |
| 990 } | |
| 991 push @{$OptionsInfo{BinRange}}, @SpecifiedBinRange; | |
| 992 } | |
| 993 else { | |
| 994 $OptionsInfo{NumOfBins} = $Options{frequencybins}; | |
| 995 if (!IsPositiveInteger($OptionsInfo{NumOfBins})) { | |
| 996 die "Error: The value specified, $Options{frequencybins}, for option \"--frequencybins\" is not valid. Allowed values: positive integer or \"number,number,[number]...\". \n"; | |
| 997 } | |
| 998 } | |
| 999 | |
| 1000 # Setup specified columns... | |
| 1001 $OptionsInfo{ColMode} = $Options{colmode}; | |
| 1002 $OptionsInfo{Columns} = defined $Options{columns} ? $Options{columns} : undef; | |
| 1003 | |
| 1004 @{$OptionsInfo{SpecifiedColumns}} = (); | |
| 1005 if (defined $Options{columns} && $Options{columns} !~ /^All$/i) { | |
| 1006 my(@SpecifiedValues) = split ",", $Options{columns}; | |
| 1007 if ($Options{colmode} =~ /^colnum$/i) { | |
| 1008 my($ColValue); | |
| 1009 for $ColValue (@SpecifiedValues) { | |
| 1010 if (!IsPositiveInteger($ColValue)) { | |
| 1011 die "Error: Column value, $ColValue, specified using \"--columns\" is not valid: Allowed integer values: > 0.\n"; | |
| 1012 } | |
| 1013 } | |
| 1014 } | |
| 1015 push @{$OptionsInfo{SpecifiedColumns}}, @SpecifiedValues; | |
| 1016 } | |
| 1017 @{$OptionsInfo{SpecifiedColumnPairs}} = (); | |
| 1018 $OptionsInfo{AllColumnPairs} = (defined($Options{columnpairs}) && $Options{columnpairs} =~ /^AllPairs$/i) ? 1 : 0; | |
| 1019 if (defined($Options{columnpairs}) && !$OptionsInfo{AllColumnPairs}) { | |
| 1020 my(@SpecifiedValues) = split ",", $Options{columnpairs}; | |
| 1021 if (@SpecifiedValues % 2) { | |
| 1022 die "Error: Invalid number of values specified using \"--columnpairs\" option: It must contain even number of values.\n"; | |
| 1023 } | |
| 1024 if ($Options{colmode} =~ /^colnum$/i) { | |
| 1025 my($ColValue); | |
| 1026 for $ColValue (@SpecifiedValues) { | |
| 1027 if (!IsPositiveInteger($ColValue)) { | |
| 1028 die "Error: Column value, $ColValue, specified using \"--columnpairs\" is not valid: Allowed integer values: > 0.\n"; | |
| 1029 } | |
| 1030 } | |
| 1031 } | |
| 1032 push @{$OptionsInfo{SpecifiedColumnPairs}}, @SpecifiedValues; | |
| 1033 } | |
| 1034 | |
| 1035 } | |
| 1036 | |
| 1037 # Setup script usage and retrieve command line arguments specified using various options... | |
| 1038 sub SetupScriptUsage { | |
| 1039 | |
| 1040 # Retrieve all the options... | |
| 1041 %Options = (); | |
| 1042 $Options{colmode} = "colnum"; | |
| 1043 $Options{detail} = 1; | |
| 1044 $Options{indelim} = "comma"; | |
| 1045 $Options{frequencybins} = 10; | |
| 1046 $Options{klargest} = 2; | |
| 1047 $Options{ksmallest} = 2; | |
| 1048 $Options{mode} = "DescriptiveStatisticsBasic"; | |
| 1049 $Options{outdelim} = "comma"; | |
| 1050 $Options{precision} = 2; | |
| 1051 $Options{quote} = "yes"; | |
| 1052 $Options{trimfraction} = 0.1; | |
| 1053 | |
| 1054 if (!GetOptions(\%Options, "colmode|c=s", "columns=s", "columnpairs=s", "detail|d=i", "frequencybins=s", "fast|f", "help|h", "indelim=s", "klargest=i", "ksmallest=i", "mode|m=s", "outdelim=s", "overwrite|o", "precision|p=i", "quote|q=s", "root|r=s", "trimfraction=f", "workingdir|w=s")) { | |
| 1055 die "\nTo get a list of valid options and their values, use \"$ScriptName -h\" or\n\"perl -S $ScriptName -h\" command and try again...\n"; | |
| 1056 } | |
| 1057 if ($Options{workingdir}) { | |
| 1058 if (! -d $Options{workingdir}) { | |
| 1059 die "Error: The value specified, $Options{workingdir}, for option \"-w --workingdir\" is not a directory name.\n"; | |
| 1060 } | |
| 1061 chdir $Options{workingdir} or die "Error: Couldn't chdir $Options{workingdir}: $! \n"; | |
| 1062 } | |
| 1063 if ($Options{colmode} !~ /^(colnum|collabel)$/i) { | |
| 1064 die "Error: The value specified, $Options{colmode}, for option \"-c --colmode\" is not valid. Allowed values: colnum or collabel\n"; | |
| 1065 } | |
| 1066 if (!IsPositiveInteger($Options{detail})) { | |
| 1067 die "Error: The value specified, $Options{detail}, for option \"-d --detail\" is not valid. Allowed values: > 0\n"; | |
| 1068 } | |
| 1069 if ($Options{indelim} !~ /^(comma|semicolon)$/i) { | |
| 1070 die "Error: The value specified, $Options{indelim}, for option \"--indelim\" is not valid. Allowed values: comma or semicolon\n"; | |
| 1071 } | |
| 1072 if ($Options{outdelim} !~ /^(comma|semicolon|tab)$/i) { | |
| 1073 die "Error: The value specified, $Options{outdelim}, for option \"--outdelim\" is not valid. Allowed values: comma, tab, or semicolon\n"; | |
| 1074 } | |
| 1075 if ($Options{quote} !~ /^(yes|no)$/i) { | |
| 1076 die "Error: The value specified, $Options{quote}, for option \"-q --quote\" is not valid. Allowed values: yes or no\n"; | |
| 1077 } | |
| 1078 if (!IsPositiveInteger($Options{precision})) { | |
| 1079 die "Error: The value specified, $Options{precision}, for option \"-p --precision\" is not valid. Allowed values: > 0 \n"; | |
| 1080 } | |
| 1081 if (!IsPositiveInteger($Options{klargest})) { | |
| 1082 die "Error: The value specified, $Options{klargest}, for option \"--klargest\" is not valid. Allowed values: > 0 \n"; | |
| 1083 } | |
| 1084 if (!IsPositiveInteger($Options{ksmallest})) { | |
| 1085 die "Error: The value specified, $Options{ksmallest}, for option \"--ksmallest\" is not valid. Allowed values: > 0 \n"; | |
| 1086 } | |
| 1087 if (IsFloat($Options{trimfraction})) { | |
| 1088 if ($Options{trimfraction} <= 0 || $Options{trimfraction} >= 1.0) { | |
| 1089 die "Error: The value specified, $Options{trimfraction}, for option \"--trimfraction\" is not valid. Allowed values: > 0 and < 1.0\n"; | |
| 1090 } | |
| 1091 } | |
| 1092 else { | |
| 1093 die "Error: The value specified, $Options{trimfraction}, for option \"--trimfraction\" is not valid. Allowed values: > 0 and < 1.0\n"; | |
| 1094 } | |
| 1095 } | |
| 1096 | |
| 1097 __END__ | |
| 1098 | |
| 1099 =head1 NAME | |
| 1100 | |
| 1101 AnalyzeTextFilesData.pl - Analyze numerical coulmn data in TextFile(s) | |
| 1102 | |
| 1103 =head1 SYNOPSIS | |
| 1104 | |
| 1105 AnalyzeTextFilesData.pl TextFile(s)... | |
| 1106 | |
| 1107 AnalyzeTextFilesData.pl [B<-c, --colmode> colnum | collabel] [B<--columns> "colnum,[colnum,...]" | "collabel,[collabel,...]" | All] | |
| 1108 [B<--columnpairs> "colnum,colnum,[colnum,colnum]..." | "collabel,collabel,[collabel,collabel]..." | AllPairs] | |
| 1109 [B<-d, --detail> infolevel] [B<-f, --fast>] [B<--frequencybins> number | "number,number,[number,...]"] [B<-h, --help>] | |
| 1110 [B<--indelim> comma | semicolon] [B<--klargest> number] [B<--ksmallest> number] | |
| 1111 [B<-m, --mode> DescriptiveStatisticsBasic | DescriptiveStatisticsAll | All | "function1, [function2,...]"] | |
| 1112 [B<-o, --overwrite>] [B<--outdelim> comma | tab | semicolon] [B<-p, --precision> number] | |
| 1113 [B<-q, --quote> yes | no] [B<-r, --root> rootname] [B<--trimfraction> number] [B<-w, --workingdir> dirname] TextFiles(s)... | |
| 1114 | |
| 1115 =head1 DESCRIPTION | |
| 1116 | |
| 1117 Anaylze numerical column data in I<TextFile(s)> using a combination of various statistical | |
| 1118 functions; Non-numerical values are simply ignored. For I<Correlation, RSquare, and Covariance> | |
| 1119 analysis, the count of valid values in specifed column pair must be same; otherwise, column | |
| 1120 pair is ignored. The file names are separated by space. The valid file extensions are I<.csv> | |
| 1121 and I<.tsv> for comma/semicolon and tab delimited text files respectively. All other | |
| 1122 file names are ignored. All the text files in a current directory can be specified by | |
| 1123 I<*.csv>, I<*.tsv>, or the current directory name. The B<--indelim> option determines | |
| 1124 the format of I<TextFile(s)>. Any file which doesn't correspond to the format indicated | |
| 1125 by B<--indelim> option is ignored. | |
| 1126 | |
| 1127 =head1 OPTIONS | |
| 1128 | |
| 1129 =over 4 | |
| 1130 | |
| 1131 =item B<-c, --colmode> I<colnum | collabel> | |
| 1132 | |
| 1133 Specify how columns are identified in TextFile(s): using column number or column | |
| 1134 label. Possible values: I<colnum or collabel>. Default value: I<colnum>. | |
| 1135 | |
| 1136 =item B<--columns> I<"colnum,[colnum,...]" | "collabel,[collabel]..." | All> | |
| 1137 | |
| 1138 This value is mode specific. It's a list of comma delimited columns to use | |
| 1139 for data analysis. Default value: I<First column>. | |
| 1140 | |
| 1141 This value is ignored during I<Correlation/Pearson Correlation> and I<Covariance> | |
| 1142 data analysis; B<-coulmnparis> option is used instead. | |
| 1143 | |
| 1144 For I<colnum> value of B<-c, --colmode> option, input values format is: | |
| 1145 I<colnum,colnum,...>. Example: | |
| 1146 | |
| 1147 1,3,5 | |
| 1148 | |
| 1149 For I<collabel> value of B<-c, --colmode> option, input values format is: | |
| 1150 I<collabel,collabel,..>. Example: | |
| 1151 | |
| 1152 ALogP,MolWeight,EC50 | |
| 1153 | |
| 1154 =item B<--columnpairs> I<"colnum,colnum,[colnum,colnum,...]" | "collabel,collabel,[collabel,collabel,...]" | AllPairs> | |
| 1155 | |
| 1156 This value is mode specific and is only used for I<Correlation, PearsonCorrelation, or | |
| 1157 Covariance> value of B<-m, --mode> option. It is a comma delimited list of column pairs | |
| 1158 to use for data analysis during I<Correlation> and I<Covariance> calculations. Default value: | |
| 1159 I<First column, Second column>. | |
| 1160 | |
| 1161 For I<colnum> value of B<-c, --colmode> option, input values format is: | |
| 1162 I<colnum,colnum,[colnum,colnum]...>. Example: | |
| 1163 | |
| 1164 1,3,5,6,1,6 | |
| 1165 | |
| 1166 For I<collabel> value of B<-c, --colmode> option, input values format is: | |
| 1167 I<collabel,collabel,[collabel,collabel]..>. Example: | |
| 1168 | |
| 1169 MolWeight,EC50,NumN+O,PSA | |
| 1170 | |
| 1171 For I<AllPairs> value of B<--columnparis> option, all column pairs are used for I<Correlation> | |
| 1172 and I<Covariance> calculations. | |
| 1173 | |
| 1174 =item B<-d, --detail> I<infolevel> | |
| 1175 | |
| 1176 Level of information to print about column values being ignored. Default: I<1>. Possible values: | |
| 1177 1, 2, 3, or 4. | |
| 1178 | |
| 1179 =item B<-f, --fast> | |
| 1180 | |
| 1181 In this mode, all the columns specified for analysis are assumed to contain numerical | |
| 1182 data and no checking is performed before analysis. By default, only numerical data is | |
| 1183 used for analysis; other types of column data is ignored. | |
| 1184 | |
| 1185 =item B<--frequencybins> I<number | "number,number,[number,...]"> | |
| 1186 | |
| 1187 Specify number of bins or bin range to use for frequency analysis. Default value: I<10> | |
| 1188 | |
| 1189 Number of bins value along with the smallest and largest value for a column is used to | |
| 1190 group the column values into different groups. | |
| 1191 | |
| 1192 The bin range list is used to group values for a column into different groups; It must contain | |
| 1193 values in ascending order. Examples: | |
| 1194 | |
| 1195 10,20,30 | |
| 1196 0.1,0.2,0.3,0.4,0.5 | |
| 1197 | |
| 1198 The frequency value calculated for a specific bin corresponds to all the column values | |
| 1199 which are greater than the previous bin value and less than or equal to the current bin value. | |
| 1200 | |
| 1201 =item B<-h, --help> | |
| 1202 | |
| 1203 Print this help message. | |
| 1204 | |
| 1205 =item B<--indelim> I<comma | semicolon> | |
| 1206 | |
| 1207 Input delimiter for CSV I<TextFile(s)>. Possible values: I<comma or semicolon>. | |
| 1208 Default value: I<comma>. For TSV files, this option is ignored and I<tab> is used as a | |
| 1209 delimiter. | |
| 1210 | |
| 1211 =item B<--klargest> I<number> | |
| 1212 | |
| 1213 Kth largest value to find by I<KLargest> function. Default value: I<2> Valid values: positive | |
| 1214 integers. | |
| 1215 | |
| 1216 =item B<--ksmallest> I<number> | |
| 1217 | |
| 1218 Kth smallest value to find by I<KSmallest> function. Default value: I<2>. Valid values: positive | |
| 1219 integers. | |
| 1220 | |
| 1221 =item B<-m, --mode> I<DescriptiveStatisticsBasic | DescriptiveStatisticsAll | All | "function1, [function2,...]"> | |
| 1222 | |
| 1223 Specify how to analyze data in TextFile(s): calculate basic or all descriptive statistics; or | |
| 1224 use a comma delimited list of supported statistical functions. Possible values: | |
| 1225 I<DescriptiveStatisticsBasic | DescriptiveStatisticsAll | "function1,[function2]...">. Default | |
| 1226 value: I<DescriptiveStatisticsBasic> | |
| 1227 | |
| 1228 I<DescriptiveStatisticsBasic> includes these functions: I<Count, Maximum, Minimum, Mean, | |
| 1229 Median, Sum, StandardDeviation, StandardError, Variance>. | |
| 1230 | |
| 1231 I<DescriptiveStatisticsAll>, in addition to I<DescriptiveStatisticsBasic> functions, includes: | |
| 1232 I<GeometricMean, Frequency, HarmonicMean, KLargest, KSmallest, Kurtosis, Mode, RSquare, | |
| 1233 Skewness, TrimMean>. | |
| 1234 | |
| 1235 I<All> uses complete list of supported functions: I<Average, AverageDeviation, Correlation, | |
| 1236 Count, Covariance, GeometricMean, Frequency, HarmonicMean, KLargest, KSmallest, Kurtosis, | |
| 1237 Maximum, Minimum, Mean, Median, Mode, RSquare, Skewness, Sum, | |
| 1238 SumOfSquares, StandardDeviation, StandardDeviationN, StandardError, StandardScores, | |
| 1239 StandardScoresN, TrimMean, Variance, VarianceN>. The function names ending with N | |
| 1240 calculate corresponding values assuming an entire population instead of a population sample. | |
| 1241 | |
| 1242 Here are the formulas for these functions: | |
| 1243 | |
| 1244 Average: See Mean | |
| 1245 | |
| 1246 AverageDeviation: SUM( ABS(x[i] - Xmean) ) / n | |
| 1247 | |
| 1248 Correlation: See Pearson Correlation | |
| 1249 | |
| 1250 Covariance: SUM( (x[i] - Xmean)(y[i] - Ymean) ) / n | |
| 1251 | |
| 1252 GeometricMean: NthROOT( PRODUCT(x[i]) ) | |
| 1253 | |
| 1254 HarmonicMean: 1 / ( SUM(1/x[i]) / n ) | |
| 1255 | |
| 1256 Mean: SUM( x[i] ) / n | |
| 1257 | |
| 1258 Median: Xsorted[(n - 1)/2 + 1] for even values of n; (Xsorted[n/2] + Xsorted[n/2 + 1])/2 | |
| 1259 for odd values of n. | |
| 1260 | |
| 1261 Kurtosis: [ {n(n + 1)/(n - 1)(n - 2)(n - 3)} SUM{ ((x[i] - Xmean)/STDDEV)^4 } ] - | |
| 1262 {3((n - 1)^2)}/{(n - 2)(n-3)} | |
| 1263 | |
| 1264 PearsonCorrelation: SUM( (x[i] - Xmean)(y[i] - Ymean) ) / SQRT( SUM( (x[i] - Xmean)^2 ) | |
| 1265 (SUM( (y[i] - Ymean)^2 )) ) | |
| 1266 | |
| 1267 RSquare: PearsonCorrelation^2 | |
| 1268 | |
| 1269 Skewness: {n/(n - 1)(n - 2)} SUM{ ((x[i] - Xmean)/STDDEV)^3 } | |
| 1270 | |
| 1271 StandardDeviation: SQRT ( SUM( (x[i] - Mean)^2 ) / (n - 1) ) | |
| 1272 | |
| 1273 StandardDeviationN: SQRT ( SUM( (x[i] - Mean)^2 ) / n ) | |
| 1274 | |
| 1275 StandardError: StandardDeviation / SQRT( n ) | |
| 1276 | |
| 1277 StandardScore: (x[i] - Mean) / (n - 1) | |
| 1278 | |
| 1279 StandardScoreN: (x[i] - Mean) / n | |
| 1280 | |
| 1281 Variance: SUM( (x[i] - Xmean)^2 / (n - 1) ) | |
| 1282 | |
| 1283 VarianceN: SUM( (x[i] - Xmean)^2 / n ) | |
| 1284 | |
| 1285 =item B<-o, --overwrite> | |
| 1286 | |
| 1287 Overwrite existing files. | |
| 1288 | |
| 1289 =item B<--outdelim> I<comma | tab | semicolon> | |
| 1290 | |
| 1291 Output text file delimiter. Possible values: I<comma, tab, or semicolon> | |
| 1292 Default value: I<comma>. | |
| 1293 | |
| 1294 =item B<-p, --precision> I<number> | |
| 1295 | |
| 1296 Precision of calculated values in the output file. Default: up to I<2> decimal places. | |
| 1297 Valid values: positive integers. | |
| 1298 | |
| 1299 =item B<-q, --quote> I<yes | no> | |
| 1300 | |
| 1301 Put quotes around column values in output text file. Possible values: I<yes or | |
| 1302 no>. Default value: I<yes>. | |
| 1303 | |
| 1304 =item B<-r, --root> I<rootname> | |
| 1305 | |
| 1306 New text file name is generated using the root: <Root>.<Ext>. Default new file | |
| 1307 name: <InitialTextFileName><Mode>.<Ext>. Based on the specified analysis, | |
| 1308 <Mode> corresponds to one of these values: DescriptiveStatisticsBasic, | |
| 1309 DescriptiveStatisticsAll, AllStatistics, SpecifiedStatistics, Covariance, Correlation, | |
| 1310 Frequency, or StandardScores. The csv, and tsv <Ext> values are used for | |
| 1311 comma/semicolon, and tab delimited text files respectively. This option is ignored for | |
| 1312 multiple input files. | |
| 1313 | |
| 1314 =item B<--trimfraction> I<number> | |
| 1315 | |
| 1316 Fraction of data to exclude from the top and bottom of the data set during | |
| 1317 I<TrimMean> calculation. Default value: I<0.1>. Valid values: > 0 and < 1. | |
| 1318 | |
| 1319 =item B<-w --workingdir> I<text> | |
| 1320 | |
| 1321 Location of working directory. Default: current directory. | |
| 1322 | |
| 1323 =back | |
| 1324 | |
| 1325 =head1 EXAMPLES | |
| 1326 | |
| 1327 To calculate basic statistics for data in first column and generate a | |
| 1328 NewSample1DescriptiveStatisticsBasic.csv file, type: | |
| 1329 | |
| 1330 % AnalyzeTextFilesData.pl -o -r NewSample1 Sample1.csv | |
| 1331 | |
| 1332 To calculate basic statistics for data in third column and generate a | |
| 1333 NewSample1DescriptiveStatisticsBasic.csv file, type: | |
| 1334 | |
| 1335 % AnalyzeTextFilesData.pl --columns 3 -o -r NewSample1 Sample1.csv | |
| 1336 | |
| 1337 To calculate basic statistics for data in MolWeight column and generate a | |
| 1338 NewSample1DescriptiveStatisticsBasic.csv file, type: | |
| 1339 | |
| 1340 % AnalyzeTextFilesData.pl -colmode collabel --columns MolWeight -o | |
| 1341 -r NewSample1 Sample1.csv | |
| 1342 | |
| 1343 To calculate all available statistics for data in third column and all column pairs, | |
| 1344 and generate NewSample1DescriptiveStatisticsAll.csv, NewSample1CorrelationMatrix.csv, | |
| 1345 NewSample1CorrelationMatrix.csv, and NewSample1MolWeightFrequencyAnalysis.csv files, | |
| 1346 type: | |
| 1347 | |
| 1348 % AnalyzeTextFilesData.pl -m DescriptiveStatisticsAll --columns 3 -o | |
| 1349 --columnpairs AllPairs -r NewSample1 Sample1.csv | |
| 1350 | |
| 1351 To compute frequency distribution of data in third column into five bins and | |
| 1352 generate NewSample1MolWeightFrequencyAnalysis.csv, type: | |
| 1353 | |
| 1354 % AnalyzeTextFilesData.pl -m Frequency --frequencybins 5 --columns 3 | |
| 1355 -o -r NewSample1 Sample1.csv | |
| 1356 | |
| 1357 To compute frequency distribution of data in third column into specified bin range | |
| 1358 values, and generate NewSample1MolWeightFrequencyAnalysis.csv, type: | |
| 1359 | |
| 1360 % AnalyzeTextFilesData.pl -m Frequency --frequencybins "100,200,400" | |
| 1361 --columns 3 -o -r NewSample1 Sample1.csv | |
| 1362 | |
| 1363 To calculate all available statistics for data in all columns and column pairs, type: | |
| 1364 | |
| 1365 % AnalyzeTextFilesData.pl -m All --columns All --columnpairs | |
| 1366 AllPairs -o -r NewSample1 Sample1.csv | |
| 1367 | |
| 1368 =head1 AUTHOR | |
| 1369 | |
| 1370 Manish Sud <msud@san.rr.com> | |
| 1371 | |
| 1372 =head1 SEE ALSO | |
| 1373 | |
| 1374 JoinTextFiles.pl, MergeTextFilesWithSD.pl, ModifyTextFilesFormat.pl, SplitTextFiles.pl, TextFilesToHTML.pl | |
| 1375 | |
| 1376 =head1 COPYRIGHT | |
| 1377 | |
| 1378 Copyright (C) 2015 Manish Sud. All rights reserved. | |
| 1379 | |
| 1380 This file is part of MayaChemTools. | |
| 1381 | |
| 1382 MayaChemTools is free software; you can redistribute it and/or modify it under | |
| 1383 the terms of the GNU Lesser General Public License as published by the Free | |
| 1384 Software Foundation; either version 3 of the License, or (at your option) | |
| 1385 any later version. | |
| 1386 | |
| 1387 =cut |