[Bio] / Sprout / SproutLoad.pm Repository:
ViewVC logotype

Diff of /Sprout/SproutLoad.pm

Parent Directory Parent Directory | Revision Log Revision Log | View Patch Patch

revision 1.76, Fri Nov 3 00:43:22 2006 UTC revision 1.88, Mon Nov 5 22:52:06 2007 UTC
# Line 7  Line 7 
7      use PageBuilder;      use PageBuilder;
8      use ERDBLoad;      use ERDBLoad;
9      use FIG;      use FIG;
10        use FIGRules;
11      use Sprout;      use Sprout;
12      use Stats;      use Stats;
13      use BasicLocation;      use BasicLocation;
14      use HTML;      use HTML;
15        use AliasAnalysis;
16    
17  =head1 Sprout Load Methods  =head1 Sprout Load Methods
18    
# Line 101  Line 103 
103              # Here we want all the complete genomes and an access code of 1.              # Here we want all the complete genomes and an access code of 1.
104              my @genomeList = $fig->genomes(1);              my @genomeList = $fig->genomes(1);
105              %genomes = map { $_ => 1 } @genomeList;              %genomes = map { $_ => 1 } @genomeList;
106                Trace(scalar(keys %genomes) . " genomes found.") if T(3);
107          } else {          } else {
108              my $type = ref $genomeFile;              my $type = ref $genomeFile;
109              Trace("Genome file parameter type is \"$type\".") if T(3);              Trace("Genome file parameter type is \"$type\".") if T(3);
# Line 167  Line 170 
170          for my $subsystem (keys %subsystems) {          for my $subsystem (keys %subsystems) {
171              my $name = $subsystem;              my $name = $subsystem;
172              $name =~ s/_/ /g;              $name =~ s/_/ /g;
173              my $classes = $fig->subsystem_classification($subsystem);  #            my $classes = $fig->subsystem_classification($subsystem);
174              $name .= " " . join(" ", @{$classes});  #            $name .= " " . join(" ", @{$classes});
175              $subsystems{$subsystem} = $name;              $subsystems{$subsystem} = $name;
176          }          }
177      }      }
178        # Get the list of NMPDR-oriented attribute keys.
179        my @propKeys = $fig->get_group_keys("NMPDR");
180      # Get the data directory from the Sprout object.      # Get the data directory from the Sprout object.
181      my ($directory) = $sprout->LoadInfo();      my ($directory) = $sprout->LoadInfo();
182      # Create the Sprout load object.      # Create the Sprout load object.
# Line 183  Line 188 
188                    loadDirectory => $directory,                    loadDirectory => $directory,
189                    erdb => $sprout,                    erdb => $sprout,
190                    loaders => [],                    loaders => [],
191                    options => $options                    options => $options,
192                      propKeys => \@propKeys,
193                   };                   };
194      # Bless and return it.      # Bless and return it.
195      bless $retVal, $class;      bless $retVal, $class;
# Line 203  Line 209 
209      return $self->{options}->{loadOnly};      return $self->{options}->{loadOnly};
210  }  }
211    
 =head3 PrimaryOnly  
   
 C<< my $flag = $spl->PrimaryOnly; >>  
   
 Return TRUE if only the main entity is to be loaded, else FALSE.  
   
 =cut  
   
 sub PrimaryOnly {  
     my ($self) = @_;  
     return $self->{options}->{primaryOnly};  
 }  
212    
213  =head3 LoadGenomeData  =head3 LoadGenomeData
214    
# Line 255  Line 249 
249      my $genomeCount = (keys %{$genomeHash});      my $genomeCount = (keys %{$genomeHash});
250      # Create load objects for each of the tables we're loading.      # Create load objects for each of the tables we're loading.
251      my $loadGenome = $self->_TableLoader('Genome');      my $loadGenome = $self->_TableLoader('Genome');
252      my $loadHasContig = $self->_TableLoader('HasContig', $self->PrimaryOnly);      my $loadHasContig = $self->_TableLoader('HasContig');
253      my $loadContig = $self->_TableLoader('Contig', $self->PrimaryOnly);      my $loadContig = $self->_TableLoader('Contig');
254      my $loadIsMadeUpOf = $self->_TableLoader('IsMadeUpOf', $self->PrimaryOnly);      my $loadIsMadeUpOf = $self->_TableLoader('IsMadeUpOf');
255      my $loadSequence = $self->_TableLoader('Sequence', $self->PrimaryOnly);      my $loadSequence = $self->_TableLoader('Sequence');
256      if ($self->{options}->{loadOnly}) {      if ($self->{options}->{loadOnly}) {
257          Trace("Loading from existing files.") if T(2);          Trace("Loading from existing files.") if T(2);
258      } else {      } else {
# Line 274  Line 268 
268              my $extra = join " ", @extraData;              my $extra = join " ", @extraData;
269              # Get the full taxonomy.              # Get the full taxonomy.
270              my $taxonomy = $fig->taxonomy_of($genomeID);              my $taxonomy = $fig->taxonomy_of($genomeID);
271                # Get the version. If no version is specified, we default to the genome ID by itself.
272                my $version = $fig->genome_version($genomeID);
273                if (! defined($version)) {
274                    $version = $genomeID;
275                }
276                # Get the DNA size.
277                my $dnaSize = $fig->genome_szdna($genomeID);
278              # Open the NMPDR group file for this genome.              # Open the NMPDR group file for this genome.
279              my $group;              my $group;
280              if (open(TMP, "<$FIG_Config::organisms/$genomeID/NMPDR") &&              if (open(TMP, "<$FIG_Config::organisms/$genomeID/NMPDR") &&
# Line 286  Line 287 
287              }              }
288              close TMP;              close TMP;
289              # Output the genome record.              # Output the genome record.
290              $loadGenome->Put($genomeID, $accessCode, $fig->is_complete($genomeID), $genus,              $loadGenome->Put($genomeID, $accessCode, $fig->is_complete($genomeID),
291                               $group, $species, $extra, $taxonomy);                               $dnaSize, $genus, $group, $species, $extra, $version, $taxonomy);
292              # Now we loop through each of the genome's contigs.              # Now we loop through each of the genome's contigs.
293              my @contigs = $fig->all_contigs($genomeID);              my @contigs = $fig->all_contigs($genomeID);
294              for my $contigID (@contigs) {              for my $contigID (@contigs) {
# Line 325  Line 326 
326      return $retVal;      return $retVal;
327  }  }
328    
 =head3 LoadCouplingData  
   
 C<< my $stats = $spl->LoadCouplingData(); >>  
   
 Load the coupling and evidence data from FIG into Sprout.  
   
 The coupling data specifies which genome features are functionally coupled. The  
 evidence data explains why the coupling is functional.  
   
 The following relations are loaded by this method.  
   
     Coupling  
     IsEvidencedBy  
     PCH  
     ParticipatesInCoupling  
     UsesAsEvidence  
   
 =over 4  
   
 =item RETURNS  
   
 Returns a statistics object for the loads.  
   
 =back  
   
 =cut  
 #: Return Type $%;  
 sub LoadCouplingData {  
     # Get this object instance.  
     my ($self) = @_;  
     # Get the FIG object.  
     my $fig = $self->{fig};  
     # Get the genome hash.  
     my $genomeFilter = $self->{genomes};  
     # Set up an ID counter for the PCHs.  
     my $pchID = 0;  
     # Start the loads.  
     my $loadCoupling = $self->_TableLoader('Coupling');  
     my $loadIsEvidencedBy = $self->_TableLoader('IsEvidencedBy', $self->PrimaryOnly);  
     my $loadPCH = $self->_TableLoader('PCH', $self->PrimaryOnly);  
     my $loadParticipatesInCoupling = $self->_TableLoader('ParticipatesInCoupling', $self->PrimaryOnly);  
     my $loadUsesAsEvidence = $self->_TableLoader('UsesAsEvidence', $self->PrimaryOnly);  
     if ($self->{options}->{loadOnly}) {  
         Trace("Loading from existing files.") if T(2);  
     } else {  
         Trace("Generating coupling data.") if T(2);  
         # Loop through the genomes found.  
         for my $genome (sort keys %{$genomeFilter}) {  
             Trace("Generating coupling data for $genome.") if T(3);  
             $loadCoupling->Add("genomeIn");  
             # Create a hash table for holding coupled pairs. We use this to prevent  
             # duplicates. For example, if A is coupled to B, we don't want to also  
             # assert that B is coupled to A, because we already know it. Fortunately,  
             # all couplings occur within a genome, so we can keep the hash table  
             # size reasonably small.  
             my %dupHash = ();  
             # Get all of the genome's PEGs.  
             my @pegs = $fig->pegs_of($genome);  
             # Loop through the PEGs.  
             for my $peg1 (@pegs) {  
                 $loadCoupling->Add("pegIn");  
                 Trace("Processing PEG $peg1 for $genome.") if T(4);  
                 # Get a list of the coupled PEGs.  
                 my @couplings = $fig->coupled_to($peg1);  
                 # For each coupled PEG, we need to verify that a coupling already  
                 # exists. If not, we have to create one.  
                 for my $coupleData (@couplings) {  
                     my ($peg2, $score) = @{$coupleData};  
                     # Compute the coupling ID.  
                     my $coupleID = $self->{erdb}->CouplingID($peg1, $peg2);  
                     if (! exists $dupHash{$coupleID}) {  
                         $loadCoupling->Add("couplingIn");  
                         # Here we have a new coupling to store in the load files.  
                         Trace("Storing coupling ($coupleID) with score $score.") if T(4);  
                         # Ensure we don't do this again.  
                         $dupHash{$coupleID} = $score;  
                         # Write the coupling record.  
                         $loadCoupling->Put($coupleID, $score);  
                         # Connect it to the coupled PEGs.  
                         $loadParticipatesInCoupling->Put($peg1, $coupleID, 1);  
                         $loadParticipatesInCoupling->Put($peg2, $coupleID, 2);  
                         # Get the evidence for this coupling.  
                         my @evidence = $fig->coupling_evidence($peg1, $peg2);  
                         # Organize the evidence into a hash table.  
                         my %evidenceMap = ();  
                         # Process each evidence item.  
                         for my $evidenceData (@evidence) {  
                             $loadPCH->Add("evidenceIn");  
                             my ($peg3, $peg4, $usage) = @{$evidenceData};  
                             # Only proceed if the evidence is from a Sprout  
                             # genome.  
                             if ($genomeFilter->{$fig->genome_of($peg3)}) {  
                                 $loadUsesAsEvidence->Add("evidenceChosen");  
                                 my $evidenceKey = "$coupleID $peg3 $peg4";  
                                 # We store this evidence in the hash if the usage  
                                 # is nonzero or no prior evidence has been found. This  
                                 # insures that if there is duplicate evidence, we  
                                 # at least keep the meaningful ones. Only evidence in  
                                 # the hash makes it to the output.  
                                 if ($usage || ! exists $evidenceMap{$evidenceKey}) {  
                                     $evidenceMap{$evidenceKey} = $evidenceData;  
                                 }  
                             }  
                         }  
                         for my $evidenceID (keys %evidenceMap) {  
                             # Get the ID for this evidence.  
                             $pchID++;  
                             # Create the evidence record.  
                             my ($peg3, $peg4, $usage) = @{$evidenceMap{$evidenceID}};  
                             $loadPCH->Put($pchID, $usage);  
                             # Connect it to the coupling.  
                             $loadIsEvidencedBy->Put($coupleID, $pchID);  
                             # Connect it to the features.  
                             $loadUsesAsEvidence->Put($pchID, $peg3, 1);  
                             $loadUsesAsEvidence->Put($pchID, $peg4, 2);  
                         }  
                     }  
                 }  
             }  
         }  
     }  
     # All done. Finish the load.  
     my $retVal = $self->_FinishAll();  
     return $retVal;  
 }  
   
329  =head3 LoadFeatureData  =head3 LoadFeatureData
330    
331  C<< my $stats = $spl->LoadFeatureData(); >>  C<< my $stats = $spl->LoadFeatureData(); >>
# Line 463  Line 338 
338    
339      Feature      Feature
340      FeatureAlias      FeatureAlias
341        IsAliasOf
342      FeatureLink      FeatureLink
343      FeatureTranslation      FeatureTranslation
344      FeatureUpstream      FeatureUpstream
# Line 472  Line 348 
348      FeatureEssential      FeatureEssential
349      FeatureVirulent      FeatureVirulent
350      FeatureIEDB      FeatureIEDB
351        CDD
352        IsPresentOnProteinOf
353    
354  =over 4  =over 4
355    
# Line 493  Line 371 
371      my $genomeHash = $self->{genomes};      my $genomeHash = $self->{genomes};
372      # Create load objects for each of the tables we're loading.      # Create load objects for each of the tables we're loading.
373      my $loadFeature = $self->_TableLoader('Feature');      my $loadFeature = $self->_TableLoader('Feature');
374      my $loadIsLocatedIn = $self->_TableLoader('IsLocatedIn', $self->PrimaryOnly);      my $loadIsLocatedIn = $self->_TableLoader('IsLocatedIn');
375      my $loadFeatureAlias = $self->_TableLoader('FeatureAlias');      my $loadFeatureAlias = $self->_TableLoader('FeatureAlias');
376        my $loadIsAliasOf = $self->_TableLoader('IsAliasOf');
377      my $loadFeatureLink = $self->_TableLoader('FeatureLink');      my $loadFeatureLink = $self->_TableLoader('FeatureLink');
378      my $loadFeatureTranslation = $self->_TableLoader('FeatureTranslation');      my $loadFeatureTranslation = $self->_TableLoader('FeatureTranslation');
379      my $loadFeatureUpstream = $self->_TableLoader('FeatureUpstream');      my $loadFeatureUpstream = $self->_TableLoader('FeatureUpstream');
380      my $loadHasFeature = $self->_TableLoader('HasFeature', $self->PrimaryOnly);      my $loadHasFeature = $self->_TableLoader('HasFeature');
381      my $loadHasRoleInSubsystem = $self->_TableLoader('HasRoleInSubsystem', $self->PrimaryOnly);      my $loadHasRoleInSubsystem = $self->_TableLoader('HasRoleInSubsystem');
382      my $loadFeatureEssential = $self->_TableLoader('FeatureEssential');      my $loadFeatureEssential = $self->_TableLoader('FeatureEssential');
383      my $loadFeatureVirulent = $self->_TableLoader('FeatureVirulent');      my $loadFeatureVirulent = $self->_TableLoader('FeatureVirulent');
384      my $loadFeatureIEDB = $self->_TableLoader('FeatureIEDB');      my $loadFeatureIEDB = $self->_TableLoader('FeatureIEDB');
385        my $loadCDD = $self->_TableLoader('CDD');
386        my $loadIsPresentOnProteinOf = $self->_TableLoader('IsPresentOnProteinOf');
387      # Get the subsystem hash.      # Get the subsystem hash.
388      my $subHash = $self->{subsystems};      my $subHash = $self->{subsystems};
389        # Get the property keys.
390        my $propKeys = $self->{propKeys};
391        # Create a hashes to hold CDD and alias values.
392        my %CDD = ();
393        my %alias = ();
394      # Get the maximum sequence size. We need this later for splitting up the      # Get the maximum sequence size. We need this later for splitting up the
395      # locations.      # locations.
396      my $chunkSize = $self->{sprout}->MaxSegment();      my $chunkSize = $self->{sprout}->MaxSegment();
# Line 513  Line 399 
399      } else {      } else {
400          Trace("Generating feature data.") if T(2);          Trace("Generating feature data.") if T(2);
401          # Now we loop through the genomes, generating the data for each one.          # Now we loop through the genomes, generating the data for each one.
402          for my $genomeID (sort keys %{$genomeHash}) {          my @allGenomes = sort keys %{$genomeHash};
403            Trace(scalar(@allGenomes) . " genomes found in list.") if T(3);
404            for my $genomeID (@allGenomes) {
405              Trace("Loading features for genome $genomeID.") if T(3);              Trace("Loading features for genome $genomeID.") if T(3);
406              $loadFeature->Add("genomeIn");              $loadFeature->Add("genomeIn");
407              # Get the feature list for this genome.              # Get the feature list for this genome.
408              my $features = $fig->all_features_detailed($genomeID);              my $features = $fig->all_features_detailed_fast($genomeID);
409              # Sort and count the list.              # Sort and count the list.
410              my @featureTuples = sort { $a->[0] cmp $b->[0] } @{$features};              my @featureTuples = sort { $a->[0] cmp $b->[0] } @{$features};
411              my $count = scalar @featureTuples;              my $count = scalar @featureTuples;
412                my @fids = map { $_->[0] } @featureTuples;
413              Trace("$count features found for genome $genomeID.") if T(3);              Trace("$count features found for genome $genomeID.") if T(3);
414                # Get the attributes for this genome and put them in a hash by feature ID.
415                my $attributes = GetGenomeAttributes($fig, $genomeID, \@fids, $propKeys);
416                Trace("Looping through features for $genomeID.") if T(3);
417              # Set up for our duplicate-feature check.              # Set up for our duplicate-feature check.
418              my $oldFeatureID = "";              my $oldFeatureID = "";
419              # Loop through the features.              # Loop through the features.
420              for my $featureTuple (@featureTuples) {              for my $featureTuple (@featureTuples) {
421                  # Split the tuple.                  # Split the tuple.
422                  my ($featureID, $locations, undef, $type) = @{$featureTuple};                  my ($featureID, $locations, undef, $type, $minloc, $maxloc, $assignment, $user, $quality) = @{$featureTuple};
423                  # Check for duplicates.                  # Check for duplicates.
424                  if ($featureID eq $oldFeatureID) {                  if ($featureID eq $oldFeatureID) {
425                      Trace("Duplicate feature $featureID found.") if T(1);                      Trace("Duplicate feature $featureID found.") if T(1);
# Line 535  Line 427 
427                      $oldFeatureID = $featureID;                      $oldFeatureID = $featureID;
428                      # Count this feature.                      # Count this feature.
429                      $loadFeature->Add("featureIn");                      $loadFeature->Add("featureIn");
430                        # Fix the quality. It is almost always a space, but some odd stuff might sneak through, and the
431                        # Sprout database requires a single character.
432                        if (! defined($quality) || $quality eq "") {
433                            $quality = " ";
434                        }
435                      # Begin building the keywords. We start with the genome ID, the                      # Begin building the keywords. We start with the genome ID, the
436                      # feature ID, and the organism name.                      # feature ID, the taxonomy, and the organism name.
437                      my @keywords = ($genomeID, $featureID, $fig->genus_species($genomeID));                      my @keywords = ($genomeID, $featureID, $fig->genus_species($genomeID),
438                      # Get the functional assignment and aliases. This                                      $fig->taxonomy_of($genomeID));
                     # depends on the feature type.  
                     my $assignment;  
                     if ($type eq "peg") {  
                         $assignment = $fig->function_of($featureID);  
439                          # Create the aliases.                          # Create the aliases.
440                          for my $alias ($fig->feature_aliases($featureID)) {                          for my $alias ($fig->feature_aliases($featureID)) {
441                              $loadFeatureAlias->Put($featureID, $alias);                          #Connect this alias to this feature.
442                            $loadIsAliasOf->Put($alias, $featureID);
443                              push @keywords, $alias;                              push @keywords, $alias;
444                            # If this is a locus tag, also add its natural form as a keyword.
445                            my $naturalName = AliasAnalysis::Type(LocusTag => $alias);
446                            if ($naturalName) {
447                                push @keywords, $naturalName;
448                            }
449                            # If this is the first time for the specified alias, create its
450                            # alias record.
451                            if (! exists $alias{$alias}) {
452                                $loadFeatureAlias->Put($alias);
453                                $alias{$alias} = 1;
454                          }                          }
                     } else {  
                         # For other types, the assignment is the first (and ONLY) alias.  
                         ($assignment) = $fig->feature_aliases($featureID);  
455                      }                      }
456                      Trace("Assignment for $featureID is: $assignment") if T(4);                      Trace("Assignment for $featureID is: $assignment") if T(4);
457                      # Break the assignment into words and shove it onto the                      # Break the assignment into words and shove it onto the
# Line 578  Line 479 
479                      }                      }
480                      # Now we need to find the subsystems this feature participates in.                      # Now we need to find the subsystems this feature participates in.
481                      # We also add the subsystems to the keyword list. Before we do that,                      # We also add the subsystems to the keyword list. Before we do that,
482                      # we must convert underscores to spaces and tack on the classifications.                      # we must convert underscores to spaces.
483                      my @subsystems = $fig->peg_to_subsystems($featureID);                      my @subsystems = $fig->peg_to_subsystems($featureID);
484                      for my $subsystem (@subsystems) {                      for my $subsystem (@subsystems) {
485                          # Only proceed if we like this subsystem.                          # Only proceed if we like this subsystem.
# Line 607  Line 508 
508                      # [name, value, value with URL]. (We don't need the PEG, since                      # [name, value, value with URL]. (We don't need the PEG, since
509                      # we already know it.)                      # we already know it.)
510                      my @attributes = map { [$_->[1], $_->[2], Tracer::CombineURL($_->[2], $_->[3])] }                      my @attributes = map { [$_->[1], $_->[2], Tracer::CombineURL($_->[2], $_->[3])] }
511                                           $fig->get_attributes($featureID);                                           @{$attributes->{$featureID}};
512                      # Now we process each of the special attributes.                      # Now we process each of the special attributes.
513                      if (SpecialAttribute($featureID, \@attributes,                      if (SpecialAttribute($featureID, \@attributes,
514                                           1, 2, '^(essential|potential_essential)$',                                           1, [0,2], '^(essential|potential_essential)$',
515                                           $loadFeatureEssential)) {                                           $loadFeatureEssential)) {
516                          push @keywords, 'essential';                          push @keywords, 'essential';
517                          $loadFeature->Add('essential');                          $loadFeature->Add('essential');
518                      }                      }
519                      if (SpecialAttribute($featureID, \@attributes,                      if (SpecialAttribute($featureID, \@attributes,
520                                           0, 2, '^virulen',                                           0, [2], '^virulen',
521                                           $loadFeatureVirulent)) {                                           $loadFeatureVirulent)) {
522                          push @keywords, 'virulent';                          push @keywords, 'virulent';
523                          $loadFeature->Add('virulent');                          $loadFeature->Add('virulent');
524                      }                      }
525                      if (SpecialAttribute($featureID, \@attributes,                      if (SpecialAttribute($featureID, \@attributes,
526                                           0, 2, '^iedb_',                                           0, [0,2], '^iedb_',
527                                           $loadFeatureIEDB)) {                                           $loadFeatureIEDB)) {
528                          push @keywords, 'iedb';                          push @keywords, 'iedb';
529                          $loadFeature->Add('iedb');                          $loadFeature->Add('iedb');
530                      }                      }
531                        # Now we have some other attributes we need to process. Currently,
532                        # this is CDD and CELLO, but we expect the number to increase.
533                        my %attributeHash = ();
534                        for my $attrRow (@{$attributes->{$featureID}}) {
535                            my (undef, $key, @values) = @{$attrRow};
536                            $key =~ /^([^:]+)::(.+)/;
537                            if (exists $attributeHash{$1}) {
538                                $attributeHash{$1}->{$2} = \@values;
539                            } else {
540                                $attributeHash{$1} = {$2 => \@values};
541                            }
542                        }
543                        my $celloValue = "unknown";
544                        # Pull in the CELLO attribute. There will never be more than one.
545                        # If we have one, it's a feature attribute AND a keyword.
546                        my @celloData = keys %{$attributeHash{CELLO}};
547                        if (@celloData) {
548                            $celloValue = $celloData[0];
549                            push @keywords, $celloValue;
550                        }
551                        # Now we handle CDD. This is a bit more complicated, because
552                        # there are multiple CDDs per protein.
553                        if (exists $attributeHash{CDD}) {
554                            # Get the hash of CDD IDs to scores for this feature. We
555                            # already know it exists because of the above IF.
556                            my $cddHash = $attributeHash{CDD};
557                            my @cddData = sort keys %{$cddHash};
558                            for my $cdd (@cddData) {
559                                # Extract the score for this CDD and decode it.
560                                my ($codeScore) = split(/\s*,\s*/, $cddHash->{$cdd}->[0]);
561                                my $realScore = FIGRules::DecodeScore($codeScore);
562                                # Create the connection.
563                                $loadIsPresentOnProteinOf->Put($cdd, $featureID, $realScore);
564                                # If this CDD does not yet exist, create its record.
565                                if (! exists $CDD{$cdd}) {
566                                    $CDD{$cdd} = 1;
567                                    $loadCDD->Put($cdd);
568                                }
569                            }
570                        }
571                      # Now we need to bust up hyphenated words in the keyword                      # Now we need to bust up hyphenated words in the keyword
572                      # list.                      # list. We keep them separate and put them at the end so
573                        # the original word order is available.
574                      my $keywordString = "";                      my $keywordString = "";
575                        my $bustedString = "";
576                      for my $keyword (@keywords) {                      for my $keyword (@keywords) {
577                          if (length $keyword >= 4) {                          if (length $keyword >= 3) {
578                              $keywordString .= " $keyword";                              $keywordString .= " $keyword";
579                              if ($keyword =~ /-/) {                              if ($keyword =~ /-/) {
580                                  my @words = grep { length($_) >= 4 } split /-/, $keyword;                                  my @words = split /-/, $keyword;
581                                  $keywordString .= join(" ", "", @words);                                  $bustedString .= join(" ", "", @words);
582                              }                              }
583                          }                          }
584                      }                      }
585                        $keywordString .= $bustedString;
586                        # Get rid of annoying punctuation.
587                        $keywordString =~ s/[();]//g;
588                      # Clean the keyword list.                      # Clean the keyword list.
589                      my $cleanWords = $sprout->CleanKeywords($keywordString);                      my $cleanWords = $sprout->CleanKeywords($keywordString);
590                      Trace("Keyword string for $featureID: $cleanWords") if T(4);                      Trace("Keyword string for $featureID: $cleanWords") if T(4);
591                      # Create the feature record.                      # Now we need to process the feature's locations. First, we split them up.
592                      $loadFeature->Put($featureID, 1, $type, $assignment, $cleanWords);                      my @locationList = split /\s*,\s*/, $locations;
593                        # Next, we convert them to Sprout location objects.
594                        my @locObjectList = map { BasicLocation->new("$genomeID:$_") } @locationList;
595                      # This part is the roughest. We need to relate the features to contig                      # This part is the roughest. We need to relate the features to contig
596                      # locations, and the locations must be split so that none of them exceed                      # locations, and the locations must be split so that none of them exceed
597                      # the maximum segment size. This simplifies the genes_in_region processing                      # the maximum segment size. This simplifies the genes_in_region processing
598                      # for Sprout.                      # for Sprout. To start, we create the location position indicator.
                     my @locationList = split /\s*,\s*/, $locations;  
                     # Create the location position indicator.  
599                      my $i = 1;                      my $i = 1;
600                      # Loop through the locations.                      # Loop through the locations.
601                      for my $location (@locationList) {                      for my $locObject (@locObjectList) {
602                          # Parse the location.                          # Split this location into a list of chunks.
                         my $locObject = BasicLocation->new("$genomeID:$location");  
                         # Split it into a list of chunks.  
603                          my @locOList = ();                          my @locOList = ();
604                          while (my $peeling = $locObject->Peel($chunkSize)) {                          while (my $peeling = $locObject->Peel($chunkSize)) {
605                              $loadIsLocatedIn->Add("peeling");                              $loadIsLocatedIn->Add("peeling");
# Line 670  Line 614 
614                              $i++;                              $i++;
615                          }                          }
616                      }                      }
617                        # Finally, reassemble the location objects into a list of Sprout location strings.
618                        $locations = join(", ", map { $_->String } @locObjectList);
619                        # Create the feature record.
620                        $loadFeature->Put($featureID, 1, $user, $quality, $celloValue, $type, $assignment, $cleanWords, $locations);
621                  }                  }
622              }              }
623                Trace("Genome $genomeID processed.") if T(3);
624          }          }
625      }      }
626      # Finish the loads.      # Finish the loads.
# Line 697  Line 646 
646      SubsystemClass      SubsystemClass
647      Role      Role
648      RoleEC      RoleEC
649        IsIdentifiedByEC
650      SSCell      SSCell
651      ContainsFeature      ContainsFeature
652      IsGenomeOf      IsGenomeOf
# Line 738  Line 688 
688      # Get the map list.      # Get the map list.
689      my @maps = $fig->all_maps;      my @maps = $fig->all_maps;
690      # Create load objects for each of the tables we're loading.      # Create load objects for each of the tables we're loading.
691      my $loadDiagram = $self->_TableLoader('Diagram', $self->PrimaryOnly);      my $loadDiagram = $self->_TableLoader('Diagram');
692      my $loadRoleOccursIn = $self->_TableLoader('RoleOccursIn', $self->PrimaryOnly);      my $loadRoleOccursIn = $self->_TableLoader('RoleOccursIn');
693      my $loadSubsystem = $self->_TableLoader('Subsystem');      my $loadSubsystem = $self->_TableLoader('Subsystem');
694      my $loadRole = $self->_TableLoader('Role', $self->PrimaryOnly);      my $loadRole = $self->_TableLoader('Role');
695      my $loadRoleEC = $self->_TableLoader('RoleEC', $self->PrimaryOnly);      my $loadRoleEC = $self->_TableLoader('RoleEC');
696      my $loadCatalyzes = $self->_TableLoader('Catalyzes', $self->PrimaryOnly);      my $loadIsIdentifiedByEC = $self->_TableLoader('IsIdentifiedByEC');
697      my $loadSSCell = $self->_TableLoader('SSCell', $self->PrimaryOnly);      my $loadCatalyzes = $self->_TableLoader('Catalyzes');
698      my $loadContainsFeature = $self->_TableLoader('ContainsFeature', $self->PrimaryOnly);      my $loadSSCell = $self->_TableLoader('SSCell');
699      my $loadIsGenomeOf = $self->_TableLoader('IsGenomeOf', $self->PrimaryOnly);      my $loadContainsFeature = $self->_TableLoader('ContainsFeature');
700      my $loadIsRoleOf = $self->_TableLoader('IsRoleOf', $self->PrimaryOnly);      my $loadIsGenomeOf = $self->_TableLoader('IsGenomeOf');
701      my $loadOccursInSubsystem = $self->_TableLoader('OccursInSubsystem', $self->PrimaryOnly);      my $loadIsRoleOf = $self->_TableLoader('IsRoleOf');
702      my $loadParticipatesIn = $self->_TableLoader('ParticipatesIn', $self->PrimaryOnly);      my $loadOccursInSubsystem = $self->_TableLoader('OccursInSubsystem');
703      my $loadHasSSCell = $self->_TableLoader('HasSSCell', $self->PrimaryOnly);      my $loadParticipatesIn = $self->_TableLoader('ParticipatesIn');
704      my $loadRoleSubset = $self->_TableLoader('RoleSubset', $self->PrimaryOnly);      my $loadHasSSCell = $self->_TableLoader('HasSSCell');
705      my $loadGenomeSubset = $self->_TableLoader('GenomeSubset', $self->PrimaryOnly);      my $loadRoleSubset = $self->_TableLoader('RoleSubset');
706      my $loadConsistsOfRoles = $self->_TableLoader('ConsistsOfRoles', $self->PrimaryOnly);      my $loadGenomeSubset = $self->_TableLoader('GenomeSubset');
707      my $loadConsistsOfGenomes = $self->_TableLoader('ConsistsOfGenomes', $self->PrimaryOnly);      my $loadConsistsOfRoles = $self->_TableLoader('ConsistsOfRoles');
708      my $loadHasRoleSubset = $self->_TableLoader('HasRoleSubset', $self->PrimaryOnly);      my $loadConsistsOfGenomes = $self->_TableLoader('ConsistsOfGenomes');
709      my $loadHasGenomeSubset = $self->_TableLoader('HasGenomeSubset', $self->PrimaryOnly);      my $loadHasRoleSubset = $self->_TableLoader('HasRoleSubset');
710      my $loadSubsystemClass = $self->_TableLoader('SubsystemClass', $self->PrimaryOnly);      my $loadHasGenomeSubset = $self->_TableLoader('HasGenomeSubset');
711        my $loadSubsystemClass = $self->_TableLoader('SubsystemClass');
712      if ($self->{options}->{loadOnly}) {      if ($self->{options}->{loadOnly}) {
713          Trace("Loading from existing files.") if T(2);          Trace("Loading from existing files.") if T(2);
714      } else {      } else {
715          Trace("Generating subsystem data.") if T(2);          Trace("Generating subsystem data.") if T(2);
716          # This hash will contain the role for each EC. When we're done, this          # This hash will contain the roles for each EC. When we're done, this
717          # information will be used to generate the Catalyzes table.          # information will be used to generate the Catalyzes table.
718          my %ecToRoles = ();          my %ecToRoles = ();
719          # Loop through the subsystems. Our first task will be to create the          # Loop through the subsystems. Our first task will be to create the
# Line 776  Line 727 
727              # Get the subsystem object.              # Get the subsystem object.
728              my $sub = $fig->get_subsystem($subsysID);              my $sub = $fig->get_subsystem($subsysID);
729              # Only proceed if the subsystem has a spreadsheet.              # Only proceed if the subsystem has a spreadsheet.
730              if (! $sub->{empty_ss}) {              if (defined($sub) && ! $sub->{empty_ss}) {
731                  Trace("Creating subsystem $subsysID.") if T(3);                  Trace("Creating subsystem $subsysID.") if T(3);
732                  $loadSubsystem->Add("subsystemIn");                  $loadSubsystem->Add("subsystemIn");
733                  # Create the subsystem record.                  # Create the subsystem record.
# Line 786  Line 737 
737                  # Now for the classification string. This comes back as a list                  # Now for the classification string. This comes back as a list
738                  # reference and we convert it to a space-delimited string.                  # reference and we convert it to a space-delimited string.
739                  my $classList = $fig->subsystem_classification($subsysID);                  my $classList = $fig->subsystem_classification($subsysID);
740                  my $classString = join(" : ", grep { $_ } @$classList);                  my $classString = join($FIG_Config::splitter, grep { $_ } @$classList);
741                  $loadSubsystemClass->Put($subsysID, $classString);                  $loadSubsystemClass->Put($subsysID, $classString);
742                  # Connect it to its roles. Each role is a column in the subsystem spreadsheet.                  # Connect it to its roles. Each role is a column in the subsystem spreadsheet.
743                  for (my $col = 0; defined($roleID = $sub->get_role($col)); $col++) {                  for (my $col = 0; defined($roleID = $sub->get_role($col)); $col++) {
744                        # Get the role's abbreviation.
745                        my $abbr = $sub->get_role_abbr($col);
746                      # Connect to this role.                      # Connect to this role.
747                      $loadOccursInSubsystem->Add("roleIn");                      $loadOccursInSubsystem->Add("roleIn");
748                      $loadOccursInSubsystem->Put($roleID, $subsysID, $col);                      $loadOccursInSubsystem->Put($roleID, $subsysID, $abbr, $col);
749                      # If it's a new role, add it to the role table.                      # If it's a new role, add it to the role table.
750                      if (! exists $roleData{$roleID}) {                      if (! exists $roleData{$roleID}) {
751                          # Get the role's abbreviation.                          # Get the role's abbreviation.
                         my $abbr = $sub->get_role_abbr($col);  
752                          # Add the role.                          # Add the role.
753                          $loadRole->Put($roleID, $abbr);                          $loadRole->Put($roleID);
754                          $roleData{$roleID} = 1;                          $roleData{$roleID} = 1;
755                          # Check for an EC number.                          # Check for an EC number.
756                          if ($roleID =~ /\(EC ([^.]+\.[^.]+\.[^.]+\.[^)]+)\)\s*$/) {                          if ($roleID =~ /\(EC (\d+\.\d+\.\d+\.\d+)\s*\)\s*$/) {
757                              my $ec = $1;                              my $ec = $1;
758                              $loadRoleEC->Put($roleID, $ec);                              $loadIsIdentifiedByEC->Put($roleID, $ec);
759                              $ecToRoles{$ec} = $roleID;                              # Check to see if this is our first encounter with this EC.
760                                if (exists $ecToRoles{$ec}) {
761                                    # No, so just add this role to the EC list.
762                                    push @{$ecToRoles{$ec}}, $roleID;
763                                } else {
764                                    # Output this EC.
765                                    $loadRoleEC->Put($ec);
766                                    # Create its role list.
767                                    $ecToRoles{$ec} = [$roleID];
768                                }
769                          }                          }
770                      }                      }
771                  }                  }
# Line 917  Line 878 
878              # Now we need to link all the map's roles to it.              # Now we need to link all the map's roles to it.
879              # A hash is used to prevent duplicates.              # A hash is used to prevent duplicates.
880              my %roleHash = ();              my %roleHash = ();
881              for my $role ($fig->map_to_ecs($map)) {              for my $ec ($fig->map_to_ecs($map)) {
882                  if (exists $ecToRoles{$role} && ! $roleHash{$role}) {                  if (exists $ecToRoles{$ec}) {
883                      $loadRoleOccursIn->Put($ecToRoles{$role}, $map);                      for my $role (@{$ecToRoles{$ec}}) {
884                            if (! $roleHash{$role}) {
885                                $loadRoleOccursIn->Put($role, $map);
886                      $roleHash{$role} = 1;                      $roleHash{$role} = 1;
887                  }                  }
888              }              }
889          }          }
890                }
891            }
892          # Before we leave, we must create the Catalyzes table. We start with the reactions,          # Before we leave, we must create the Catalyzes table. We start with the reactions,
893          # then use the "ecToRoles" table to convert EC numbers to role IDs.          # then use the "ecToRoles" table to convert EC numbers to role IDs.
894          my @reactions = $fig->all_reactions();          my @reactions = $fig->all_reactions();
895          for my $reactionID (@reactions) {          for my $reactionID (@reactions) {
896              # Get this reaction's list of roles. The results will be EC numbers.              # Get this reaction's list of roles. The results will be EC numbers.
897              my @roles = $fig->catalyzed_by($reactionID);              my @ecs = $fig->catalyzed_by($reactionID);
898              # Loop through the roles, creating catalyzation records.              # Loop through the roles, creating catalyzation records.
899              for my $thisRole (@roles) {              for my $thisEC (@ecs) {
900                  if (exists $ecToRoles{$thisRole}) {                  if (exists $ecToRoles{$thisEC}) {
901                      $loadCatalyzes->Put($ecToRoles{$thisRole}, $reactionID);                      for my $thisRole (@{$ecToRoles{$thisEC}}) {
902                            $loadCatalyzes->Put($thisRole, $reactionID);
903                        }
904                  }                  }
905              }              }
906          }          }
# Line 981  Line 948 
948      my $genomeHash = $self->{genomes};      my $genomeHash = $self->{genomes};
949      # Create load objects for each of the tables we're loading.      # Create load objects for each of the tables we're loading.
950      my $loadProperty = $self->_TableLoader('Property');      my $loadProperty = $self->_TableLoader('Property');
951      my $loadHasProperty = $self->_TableLoader('HasProperty', $self->PrimaryOnly);      my $loadHasProperty = $self->_TableLoader('HasProperty');
952      if ($self->{options}->{loadOnly}) {      if ($self->{options}->{loadOnly}) {
953          Trace("Loading from existing files.") if T(2);          Trace("Loading from existing files.") if T(2);
954      } else {      } else {
# Line 989  Line 956 
956          # Create a hash for storing property IDs.          # Create a hash for storing property IDs.
957          my %propertyKeys = ();          my %propertyKeys = ();
958          my $nextID = 1;          my $nextID = 1;
959            # Get the attributes we intend to store in the property table.
960            my $propKeys = $self->{propKeys};
961          # Loop through the genomes.          # Loop through the genomes.
962          for my $genomeID (sort keys %{$genomeHash}) {          for my $genomeID (sort keys %{$genomeHash}) {
963              $loadProperty->Add("genomeIn");              $loadProperty->Add("genomeIn");
964              Trace("Generating properties for $genomeID.") if T(3);              Trace("Generating properties for $genomeID.") if T(3);
965              # Get the genome's features. The feature ID is the first field in the              # Initialize a counter.
             # tuples returned by "all_features_detailed". We use "all_features_detailed"  
             # rather than "all_features" because we want all features regardless of type.  
             my @features = map { $_->[0] } @{$fig->all_features_detailed($genomeID)};  
             my $featureCount = 0;  
966              my $propertyCount = 0;              my $propertyCount = 0;
967              # Loop through the features, creating HasProperty records.              # Get the properties for this genome's features.
968              for my $fid (@features) {              my @attributes = $fig->get_attributes("fig|$genomeID%", $propKeys);
969                  # Get all attributes for this feature. We do this one feature at a time              Trace("Property list built for $genomeID.") if T(3);
970                  # to insure we do not get any genome attributes.              # Loop through the results, creating HasProperty records.
971                  my @attributeList = $fig->get_attributes($fid, '', '', '');              for my $attributeData (@attributes) {
972                  # Add essentiality and virulence attributes.                  # Pull apart the attribute tuple.
973                  if ($fig->essential($fid)) {                  my ($fid, $key, $value, $url) = @{$attributeData};
                     push @attributeList, [$fid, 'essential', 1, ''];  
                 }  
                 if ($fig->virulent($fid)) {  
                     push @attributeList, [$fid, 'virulent', 1, ''];  
                 }  
                 if (scalar @attributeList) {  
                     $featureCount++;  
                 }  
                 # Loop through the attributes.  
                 for my $tuple (@attributeList) {  
                     $propertyCount++;  
                     # Get this attribute value's data. Note that we throw away the FID,  
                     # since it will always be the same as the value if "$fid".  
                     my (undef, $key, $value, $url) = @{$tuple};  
974                      # Concatenate the key and value and check the "propertyKeys" hash to                      # Concatenate the key and value and check the "propertyKeys" hash to
975                      # see if we already have an ID for it. We use a tab for the separator                      # see if we already have an ID for it. We use a tab for the separator
976                      # character.                      # character.
# Line 1037  Line 988 
988                      # Create the HasProperty entry for this feature/property association.                      # Create the HasProperty entry for this feature/property association.
989                      $loadHasProperty->Put($fid, $propertyID, $url);                      $loadHasProperty->Put($fid, $propertyID, $url);
990                  }                  }
             }  
991              # Update the statistics.              # Update the statistics.
992              Trace("$propertyCount attributes processed for $featureCount features.") if T(3);              Trace("$propertyCount attributes processed.") if T(3);
             $loadHasProperty->Add("featuresIn", $featureCount);  
993              $loadHasProperty->Add("propertiesIn", $propertyCount);              $loadHasProperty->Add("propertiesIn", $propertyCount);
994          }          }
995      }      }
# Line 1085  Line 1034 
1034      my $genomeHash = $self->{genomes};      my $genomeHash = $self->{genomes};
1035      # Create load objects for each of the tables we're loading.      # Create load objects for each of the tables we're loading.
1036      my $loadAnnotation = $self->_TableLoader('Annotation');      my $loadAnnotation = $self->_TableLoader('Annotation');
1037      my $loadIsTargetOfAnnotation = $self->_TableLoader('IsTargetOfAnnotation', $self->PrimaryOnly);      my $loadIsTargetOfAnnotation = $self->_TableLoader('IsTargetOfAnnotation');
1038      my $loadSproutUser = $self->_TableLoader('SproutUser', $self->PrimaryOnly);      my $loadSproutUser = $self->_TableLoader('SproutUser');
1039      my $loadUserAccess = $self->_TableLoader('UserAccess', $self->PrimaryOnly);      my $loadUserAccess = $self->_TableLoader('UserAccess');
1040      my $loadMadeAnnotation = $self->_TableLoader('MadeAnnotation', $self->PrimaryOnly);      my $loadMadeAnnotation = $self->_TableLoader('MadeAnnotation');
1041      if ($self->{options}->{loadOnly}) {      if ($self->{options}->{loadOnly}) {
1042          Trace("Loading from existing files.") if T(2);          Trace("Loading from existing files.") if T(2);
1043      } else {      } else {
# Line 1192  Line 1141 
1141      # Get the genome hash.      # Get the genome hash.
1142      my $genomeHash = $self->{genomes};      my $genomeHash = $self->{genomes};
1143      # Create load objects for each of the tables we're loading.      # Create load objects for each of the tables we're loading.
1144      my $loadComesFrom = $self->_TableLoader('ComesFrom', $self->PrimaryOnly);      my $loadComesFrom = $self->_TableLoader('ComesFrom');
1145      my $loadSource = $self->_TableLoader('Source');      my $loadSource = $self->_TableLoader('Source');
1146      my $loadSourceURL = $self->_TableLoader('SourceURL');      my $loadSourceURL = $self->_TableLoader('SourceURL');
1147      if ($self->{options}->{loadOnly}) {      if ($self->{options}->{loadOnly}) {
# Line 1329  Line 1278 
1278      Compound      Compound
1279      CompoundName      CompoundName
1280      CompoundCAS      CompoundCAS
1281        IsIdentifiedByCAS
1282        HasCompoundName
1283      IsAComponentOf      IsAComponentOf
1284    
1285  This method proceeds reaction by reaction rather than genome by genome.  This method proceeds reaction by reaction rather than genome by genome.
# Line 1350  Line 1301 
1301      my $fig = $self->{fig};      my $fig = $self->{fig};
1302      # Create load objects for each of the tables we're loading.      # Create load objects for each of the tables we're loading.
1303      my $loadReaction = $self->_TableLoader('Reaction');      my $loadReaction = $self->_TableLoader('Reaction');
1304      my $loadReactionURL = $self->_TableLoader('ReactionURL', $self->PrimaryOnly);      my $loadReactionURL = $self->_TableLoader('ReactionURL');
1305      my $loadCompound = $self->_TableLoader('Compound', $self->PrimaryOnly);      my $loadCompound = $self->_TableLoader('Compound');
1306      my $loadCompoundName = $self->_TableLoader('CompoundName', $self->PrimaryOnly);      my $loadCompoundName = $self->_TableLoader('CompoundName');
1307      my $loadCompoundCAS = $self->_TableLoader('CompoundCAS', $self->PrimaryOnly);      my $loadCompoundCAS = $self->_TableLoader('CompoundCAS');
1308      my $loadIsAComponentOf = $self->_TableLoader('IsAComponentOf', $self->PrimaryOnly);      my $loadIsAComponentOf = $self->_TableLoader('IsAComponentOf');
1309        my $loadIsIdentifiedByCAS = $self->_TableLoader('IsIdentifiedByCAS');
1310        my $loadHasCompoundName = $self->_TableLoader('HasCompoundName');
1311      if ($self->{options}->{loadOnly}) {      if ($self->{options}->{loadOnly}) {
1312          Trace("Loading from existing files.") if T(2);          Trace("Loading from existing files.") if T(2);
1313      } else {      } else {
1314          Trace("Generating annotation data.") if T(2);          Trace("Generating reaction data.") if T(2);
1315            # We need some hashes to prevent duplicates.
1316            my %compoundNames = ();
1317            my %compoundCASes = ();
1318          # First we create the compounds.          # First we create the compounds.
1319          my @compounds = $fig->all_compounds();          my @compounds = $fig->all_compounds();
1320          for my $cid (@compounds) {          for my $cid (@compounds) {
# Line 1367  Line 1323 
1323              # Each name will be given a priority number, starting with 1.              # Each name will be given a priority number, starting with 1.
1324              my $prio = 1;              my $prio = 1;
1325              for my $name (@names) {              for my $name (@names) {
1326                  $loadCompoundName->Put($cid, $name, $prio++);                  if (! exists $compoundNames{$name}) {
1327                        $loadCompoundName->Put($name);
1328                        $compoundNames{$name} = 1;
1329                    }
1330                    $loadHasCompoundName->Put($cid, $name, $prio++);
1331              }              }
1332              # Create the main compound record. Note that the first name              # Create the main compound record. Note that the first name
1333              # becomes the label.              # becomes the label.
# Line 1376  Line 1336 
1336              # Check for a CAS ID.              # Check for a CAS ID.
1337              my $cas = $fig->cas($cid);              my $cas = $fig->cas($cid);
1338              if ($cas) {              if ($cas) {
1339                  $loadCompoundCAS->Put($cid, $cas);                  $loadIsIdentifiedByCAS->Put($cid, $cas);
1340                    if (! exists $compoundCASes{$cas}) {
1341                        $loadCompoundCAS->Put($cas);
1342                        $compoundCASes{$cas} = 1;
1343                    }
1344              }              }
1345          }          }
1346          # All the compounds are set up, so we need to loop through the reactions next. First,          # All the compounds are set up, so we need to loop through the reactions next. First,
# Line 1413  Line 1377 
1377      return $retVal;      return $retVal;
1378  }  }
1379    
 =head3 LoadGroupData  
   
 C<< my $stats = $spl->LoadGroupData(); >>  
   
 Load the genome Groups into Sprout.  
   
 The following relations are loaded by this method.  
   
     GenomeGroups  
   
 Currently, we do not use groups. We used to use them for NMPDR groups,  
 butThere is no direct support for genome groups in FIG, so we access the SEED  
 files directly.  
   
 =over 4  
   
 =item RETURNS  
   
 Returns a statistics object for the loads.  
   
 =back  
   
 =cut  
 #: Return Type $%;  
 sub LoadGroupData {  
     # Get this object instance.  
     my ($self) = @_;  
     # Get the FIG object.  
     my $fig = $self->{fig};  
     # Get the genome hash.  
     my $genomeHash = $self->{genomes};  
     # Create a load object for the table we're loading.  
     my $loadGenomeGroups = $self->_TableLoader('GenomeGroups');  
     if ($self->{options}->{loadOnly}) {  
         Trace("Loading from existing files.") if T(2);  
     } else {  
         Trace("Generating group data.") if T(2);  
         # Currently there are no groups.  
     }  
     # Finish the load.  
     my $retVal = $self->_FinishAll();  
     return $retVal;  
 }  
   
1380  =head3 LoadSynonymData  =head3 LoadSynonymData
1381    
1382  C<< my $stats = $spl->LoadSynonymData(); >>  C<< my $stats = $spl->LoadSynonymData(); >>
# Line 1498  Line 1418 
1418          Trace("Generating synonym group data.") if T(2);          Trace("Generating synonym group data.") if T(2);
1419          # Get the database handle.          # Get the database handle.
1420          my $dbh = $fig->db_handle();          my $dbh = $fig->db_handle();
1421          # Ask for the synonyms.          # Ask for the synonyms. Note that "maps_to" is a group name, and "syn_id" is a PEG ID or alias.
1422          my $sth = $dbh->prepare_command("SELECT maps_to, syn_id FROM peg_synonyms ORDER BY maps_to");          my $sth = $dbh->prepare_command("SELECT maps_to, syn_id FROM peg_synonyms ORDER BY maps_to");
1423          my $result = $sth->execute();          my $result = $sth->execute();
1424          if (! defined($result)) {          if (! defined($result)) {
# Line 1510  Line 1430 
1430              my $featureCount = 0;              my $featureCount = 0;
1431              # Loop through the synonym/peg pairs.              # Loop through the synonym/peg pairs.
1432              while (my @row = $sth->fetchrow()) {              while (my @row = $sth->fetchrow()) {
1433                  # Get the synonym ID and feature ID.                  # Get the synonym group ID and feature ID.
1434                  my ($syn_id, $peg) = @row;                  my ($syn_id, $peg) = @row;
1435                  # Insure it's for one of our genomes.                  # Insure it's for one of our genomes.
1436                  my $genomeID = FIG::genome_of($peg);                  my $genomeID = FIG::genome_of($peg);
# Line 1613  Line 1533 
1533    
1534  The following relations are loaded by this method.  The following relations are loaded by this method.
1535    
     DrugProject  
     ContainsTopic  
     DrugTopic  
     ContainsAnalysisOf  
1536      PDB      PDB
1537      IncludesBound      DocksWith
1538      IsBoundIn      IsProteinForFeature
     BindsWith  
1539      Ligand      Ligand
     DescribesProteinForFeature  
     FeatureConservation  
1540    
1541  The source information for these relations is taken from flat files in the  The source information for these relations is taken from attributes. The
1542  C<$FIG_Config::drug_directory>. The file C<master_tables.list> contains  C<PDB> attribute links a PDB to a feature, and is used to build B<IsProteinForFeature>.
1543  a list of drug project names paired with file names. The named file (in the  The C<zinc_name> attribute describes the ligands. The C<docking_results>
1544  same directory) contains all the data for the project.  attribute contains the information for the B<DocksWith> relationship. It is
1545    expected that additional attributes and tables will be added in the future.
1546    
1547  =over 4  =over 4
1548    
# Line 1648  Line 1562 
1562      # Get the genome hash.      # Get the genome hash.
1563      my $genomeHash = $self->{genomes};      my $genomeHash = $self->{genomes};
1564      # Create load objects for the tables we're loading.      # Create load objects for the tables we're loading.
     my $loadDrugProject = $self->_TableLoader('DrugProject');  
     my $loadContainsTopic = $self->_TableLoader('ContainsTopic');  
     my $loadDrugTopic = $self->_TableLoader('DrugTopic');  
     my $loadContainsAnalysisOf = $self->_TableLoader('ContainsAnalysisOf');  
1565      my $loadPDB = $self->_TableLoader('PDB');      my $loadPDB = $self->_TableLoader('PDB');
     my $loadIncludesBound = $self->_TableLoader('IncludesBound');  
     my $loadIsBoundIn = $self->_TableLoader('IsBoundIn');  
     my $loadBindsWith = $self->_TableLoader('BindsWith');  
1566      my $loadLigand = $self->_TableLoader('Ligand');      my $loadLigand = $self->_TableLoader('Ligand');
1567      my $loadDescribesProteinForFeature = $self->_TableLoader('DescribesProteinForFeature');      my $loadIsProteinForFeature = $self->_TableLoader('IsProteinForFeature');
1568      my $loadFeatureConservation = $self->_TableLoader('FeatureConservation');      my $loadDocksWith = $self->_TableLoader('DocksWith');
1569      if ($self->{options}->{loadOnly}) {      if ($self->{options}->{loadOnly}) {
1570          Trace("Loading from existing files.") if T(2);          Trace("Loading from existing files.") if T(2);
1571      } else {      } else {
1572          Trace("Generating drug target data.") if T(2);          Trace("Generating drug target data.") if T(2);
1573          # Load the project list. The file comes in as a list of chomped lines,          # First comes the "DocksWith" relationship. This will give us a list of PDBs.
1574          # and we split them on the TAB character to make the project name the          # We can also encounter PDBs when we process "IsProteinForFeature". To manage
1575          # key and the file name the value of the resulting hash.          # this process, PDB information is collected in a hash table and then
1576          my %projects = map { split /\t/, $_ } Tracer::GetFile("$FIG_Config::drug_directory/master_tables.list");          # unspooled after both relationships are created.
1577          # Create hashes for the derived objects: PDBs, Features, and Ligands. These objects          my %pdbHash = ();
1578          # may occur multiple times in a single project file or even in multiple project          Trace("Generating docking data.") if T(2);
1579          # files.          # Get all the docking data. This may cause problems if there are too many PDBs,
1580          my %ligands = ();          # at which point we'll need another algorithm. The indicator that this is
1581          my %pdbs = ();          # happening will be a timeout error in the next statement.
1582          my %features = ();          my @dockData = $fig->query_attributes('$key = ? AND $value < ?',
1583          my %bindings = ();                                                ['docking_results', $FIG_Config::dockLimit]);
1584          # Set up a counter for drug topics. This will be used as the key.          Trace(scalar(@dockData) . " rows of docking data found.") if T(3);
1585          my $topicCounter = 0;          for my $dockData (@dockData) {
1586          # Loop through the projects. We sort the keys not because we need them sorted, but              # Get the docking data components.
1587          # because it makes it easier to infer our progress from trace messages.              my ($pdbID, $docking_key, @valueData) = @{$dockData};
1588          for my $project (sort keys %projects) {              # Fix the PDB ID. It's supposed to be lower-case, but this does not always happen.
1589              Trace("Processing project $project.") if T(3);              $pdbID = lc $pdbID;
1590              # Only proceed if the download file exists.              # Strip off the object type.
1591              my $projectFile = "$FIG_Config::drug_directory/$projects{$project}";              $pdbID =~ s/pdb://;
1592              if (! -f $projectFile) {              # Extract the ZINC ID from the docking key. Note that there are two possible
1593                  Trace("Project file $projectFile not found.") if T(0);              # formats.
1594              } else {              my (undef, $zinc_id) = $docking_key =~ /^docking_results::(ZINC)?(\d+)$/;
1595                  # Create the project record.              if (! $zinc_id) {
1596                  $loadDrugProject->Put($project);                  Trace("Invalid docking result key $docking_key for $pdbID.") if T(0);
1597                  # Create a hash for the topics. Each project has one or more topics. The                  $loadDocksWith->Add("errors");
1598                  # topic is identified by a URL, a category, and an identifier.              } else {
1599                  my %topics = ();                  # Get the pieces of the value and parse the energy.
1600                  # Now we can open the project file.                  # Note that we don't care about the rank, since
1601                  Trace("Reading project file $projectFile.") if T(3);                  # we can sort on the energy level itself in our database.
1602                  Open(\*PROJECT, "<$projectFile");                  my ($energy, $tool, $type) = @valueData;
1603                  # Get the first record, which is a list of column headers. We don't use this                  my ($rank, $total, $vanderwaals, $electrostatic) = split /\s*;\s*/, $energy;
1604                  # for anything, but it may be useful for debugging.                  # Ignore predicted results.
1605                  my $headerLine = <PROJECT>;                  if ($type ne "Predicted") {
1606                  # Loop through the rest of the records.                      # Count this docking result.
1607                  while (! eof PROJECT) {                      if (! exists $pdbHash{$pdbID}) {
1608                      # Get the current line of data. Note that not all lines will have all                          $pdbHash{$pdbID} = 1;
1609                      # the fields. In particular, the CLIBE data is fairly rare.                      } else {
1610                      my ($authorOrganism, $category, $tag, $refURL, $peg, $conservation,                          $pdbHash{$pdbID}++;
1611                          $pdbBound, $pdbBoundEval, $pdbFree, $pdbFreeEval, $pdbFreeTitle,                      }
1612                          $protDistInfo, $passAspInfo, $passAspFile, $passWeightInfo,                      # Write the result to the output.
1613                          $passWeightFile, $clibeInfo, $clibeURL, $clibeTotalEnergy,                      $loadDocksWith->Put($pdbID, $zinc_id, $electrostatic, $type, $tool,
1614                          $clibeVanderwaals, $clibeHBonds, $clibeEI, $clibeSolvationE)                                          $total, $vanderwaals);
                        = Tracer::GetLine(\*PROJECT);  
                     # The tag contains an identifier for the current line of data followed  
                     # by a text statement that generally matches a property name in the  
                     # main database. We split it up, since the identifier goes with  
                     # the PDB data and the text statement is part of the topic.  
                     my ($lineID, $topicTag) = split /\s*,\s*/, $tag;  
                     $loadDrugProject->Add("data line");  
                     # Check for a new topic.  
                     my $topicData = "$category\t$topicTag\t$refURL";  
                     if (! exists $topics{$topicData}) {  
                         # Here we have a new topic. Compute its ID.  
                         $topicCounter++;  
                         $topics{$topicData} = $topicCounter;  
                         # Create its database record.  
                         $loadDrugTopic->Put($topicCounter, $refURL, $category, $authorOrganism,  
                                             $topicTag);  
                         # Connect it to the project.  
                         $loadContainsTopic->Put($project, $topicCounter);  
                         $loadDrugTopic->Add("topic");  
                     }  
                     # Now we know the topic ID exists in the hash and the topic will  
                     # appear in the database, so we get this topic's ID.  
                     my $topicID = $topics{$topicData};  
                     # If the feature in this line is new, we need to save its conservation  
                     # number.  
                     if (! exists $features{$peg}) {  
                         $loadFeatureConservation->Put($peg, $conservation);  
                         $features{$peg} = 1;  
                     }  
                     # Now we have two PDBs to deal with-- a bound PDB and a free PDB.  
                     # The free PDB will have data about docking points; the bound PDB  
                     # will have data about docking. We store both types as PDBs, and  
                     # the special data comes from relationships. First we process the  
                     # bound PDB.  
                     if ($pdbBound) {  
                         $loadPDB->Add("bound line");  
                         # Insure this PDB is in the database.  
                         $self->CreatePDB($pdbBound, lc "$pdbFreeTitle (bound)", "bound", \%pdbs, $loadPDB);  
                         # Connect it to this topic.  
                         $loadIncludesBound->Put($topicID, $pdbBound);  
                         # Check for CLIBE data.  
                         if ($clibeInfo) {  
                             $loadLigand->Add("clibes");  
                             # We have CLIBE data, so we create a ligand and relate it to the PDB.  
                             if (! exists $ligands{$clibeInfo}) {  
                                 # This is a new ligand, so create its record.  
                                 $loadLigand->Put($clibeInfo);  
                                 $loadLigand->Add("ligand");  
                                 # Make sure we know this ligand already exists.  
                                 $ligands{$clibeInfo} = 1;  
                             }  
                             # Now connect the PDB to the ligand using the CLIBE data.  
                             $loadBindsWith->Put($pdbBound, $clibeInfo, $clibeURL, $clibeHBonds, $clibeEI,  
                                                 $clibeSolvationE, $clibeVanderwaals);  
                         }  
                         # Connect this PDB to the feature.  
                         $loadDescribesProteinForFeature->Put($pdbBound, $peg, $protDistInfo, $pdbBoundEval);  
                     }  
                     # Next is the free PDB.  
                     if ($pdbFree) {  
                         $loadPDB->Add("free line");  
                         # Insure this PDB is in the database.  
                         $self->CreatePDB($pdbFree, lc $pdbFreeTitle, "free", \%pdbs, $loadPDB);  
                         # Connect it to this topic.  
                         $loadContainsAnalysisOf->Put($topicID, $pdbFree, $passAspInfo,  
                                                      $passWeightFile, $passWeightInfo, $passAspFile);  
                         # Connect this PDB to the feature.  
                         $loadDescribesProteinForFeature->Put($pdbFree, $peg, $protDistInfo, $pdbFreeEval);  
                     }  
                     # If we have both PDBs, we may need to link them.  
                     if ($pdbFree && $pdbBound) {  
                         $loadIsBoundIn->Add("connection");  
                         # Insure we only link them once.  
                         my $bindingKey =  "$pdbFree\t$pdbBound";  
                         if (! exists $bindings{$bindingKey}) {  
                             $loadIsBoundIn->Add("newConnection");  
                             $loadIsBoundIn->Put($pdbFree, $pdbBound);  
                             $bindings{$bindingKey} = 1;  
1615                          }                          }
1616                      }                      }
1617                  }                  }
1618                  # Close off this project.          Trace("Connecting features.") if T(2);
1619                  close PROJECT;          # Loop through the genomes.
1620            for my $genome (sort keys %{$genomeHash}) {
1621                Trace("Generating PDBs for $genome.") if T(3);
1622                # Get all of the PDBs that BLAST against this genome's features.
1623                my @attributeData = $fig->get_attributes("fig|$genome%", 'PDB::%');
1624                for my $pdbData (@attributeData) {
1625                    # The PDB ID is coded as a subkey.
1626                    if ($pdbData->[1] !~ /PDB::(.+)/i) {
1627                        Trace("Invalid PDB ID \"$pdbData->[1]\" in attribute table.") if T(0);
1628                        $loadPDB->Add("errors");
1629                    } else {
1630                        my $pdbID = $1;
1631                        # Insure the PDB is in the hash.
1632                        if (! exists $pdbHash{$pdbID}) {
1633                            $pdbHash{$pdbID} = 0;
1634                        }
1635                        # The score and locations are coded in the attribute value.
1636                        if ($pdbData->[2] !~ /^([^;]+)(.*)$/) {
1637                            Trace("Invalid PDB data for $pdbID and feature $pdbData->[0].") if T(0);
1638                            $loadIsProteinForFeature->Add("errors");
1639                        } else {
1640                            my ($score, $locData) = ($1,$2);
1641                            # The location data may not be present, so we have to start with some
1642                            # defaults and then check.
1643                            my ($start, $end) = (1, 0);
1644                            if ($locData) {
1645                                $locData =~ /(\d+)-(\d+)/;
1646                                $start = $1;
1647                                $end = $2;
1648                            }
1649                            # If we still don't have the end location, compute it from
1650                            # the feature length.
1651                            if (! $end) {
1652                                # Most features have one location, but we do a list iteration
1653                                # just in case.
1654                                my @locations = $fig->feature_location($pdbData->[0]);
1655                                $end = 0;
1656                                for my $loc (@locations) {
1657                                    my $locObject = BasicLocation->new($loc);
1658                                    $end += $locObject->Length;
1659                                }
1660                            }
1661                            # Decode the score.
1662                            my $realScore = FIGRules::DecodeScore($score);
1663                            # Connect the PDB to the feature.
1664                            $loadIsProteinForFeature->Put($pdbData->[0], $pdbID, $start, $realScore, $end);
1665                        }
1666                    }
1667                }
1668            }
1669            # We've got all our PDBs now, so we unspool them from the hash.
1670            Trace("Generating PDBs. " . scalar(keys %pdbHash) . " found.") if T(2);
1671            my $count = 0;
1672            for my $pdbID (sort keys %pdbHash) {
1673                $loadPDB->Put($pdbID, $pdbHash{$pdbID});
1674                $count++;
1675                Trace("$count PDBs processed.") if T(3) && ($count % 500 == 0);
1676            }
1677            # Finally we create the ligand table. This information can be found in the
1678            # zinc_name attribute.
1679            Trace("Loading ligands.") if T(2);
1680            # The ligand list is huge, so we have to get it in pieces. We also have to check for duplicates.
1681            my $last_zinc_id = "";
1682            my $zinc_id = "";
1683            my $done = 0;
1684            while (! $done) {
1685                # Get the next 10000 ligands. We insist that the object ID is greater than
1686                # the last ID we processed.
1687                Trace("Loading batch starting with ZINC:$zinc_id.") if T(3);
1688                my @attributeData = $fig->query_attributes('$object > ? AND $key = ? ORDER BY $object LIMIT 10000',
1689                                                           ["ZINC:$zinc_id", "zinc_name"]);
1690                Trace(scalar(@attributeData) . " attribute rows returned.") if T(3);
1691                if (! @attributeData) {
1692                    # Here there are no attributes left, so we quit the loop.
1693                    $done = 1;
1694                } else {
1695                    # Process the attribute data we've received.
1696                    for my $zinc_data (@attributeData) {
1697                        # The ZINC ID is found in the first return column, prefixed with the word ZINC.
1698                        if ($zinc_data->[0] =~ /^ZINC:(\d+)$/) {
1699                            $zinc_id = $1;
1700                            # Check for a duplicate.
1701                            if ($zinc_id eq $last_zinc_id) {
1702                                $loadLigand->Add("duplicate");
1703                            } else {
1704                                # Here it's safe to output the ligand. The ligand name is the attribute value
1705                                # (third column in the row).
1706                                $loadLigand->Put($zinc_id, $zinc_data->[2]);
1707                                # Insure we don't try to add this ID again.
1708                                $last_zinc_id = $zinc_id;
1709                            }
1710                        } else {
1711                            Trace("Invalid zinc ID \"$zinc_data->[0]\" in attribute table.") if T(0);
1712                            $loadLigand->Add("errors");
1713                        }
1714                    }
1715              }              }
1716          }          }
1717            Trace("Ligands loaded.") if T(2);
1718      }      }
1719      # Finish the load.      # Finish the load.
1720      my $retVal = $self->_FinishAll();      my $retVal = $self->_FinishAll();
# Line 1801  Line 1726 
1726    
1727  =head3 SpecialAttribute  =head3 SpecialAttribute
1728    
1729  C<< my $count = SproutLoad::SpecialAttribute($id, \@attributes, $idxMatch, $idxValue, $pattern, $loader); >>  C<< my $count = SproutLoad::SpecialAttribute($id, \@attributes, $idxMatch, \@idxValues, $pattern, $loader); >>
1730    
1731  Look for special attributes of a given type. A special attribute is found by comparing one of  Look for special attributes of a given type. A special attribute is found by comparing one of
1732  the columns of the incoming attribute list to a search pattern. If a match is found, then  the columns of the incoming attribute list to a search pattern. If a match is found, then
1733  another column is put into an output table connected to the specified ID.  a set of columns is put into an output table connected to the specified ID.
1734    
1735  For example, when processing features, the attribute list we look at has three columns: attribute  For example, when processing features, the attribute list we look at has three columns: attribute
1736  name, attribute value, and attribute value HTML. The IEDB attribute exists if the attribute name  name, attribute value, and attribute value HTML. The IEDB attribute exists if the attribute name
1737  begins with C<iedb_>. The call signature is therefore  begins with C<iedb_>. The call signature is therefore
1738    
1739      my $found = SpecialAttribute($fid, \@attributeList, 0, 2, '^iedb_', $loadFeatureIEDB);      my $found = SpecialAttribute($fid, \@attributeList, 0, [0,2], '^iedb_', $loadFeatureIEDB);
1740    
1741  The pattern is matched against column 0, and if we have a match, then column 2's value is put  The pattern is matched against column 0, and if we have a match, then column 2's value is put
1742  to the output along with the specified feature ID.  to the output along with the specified feature ID.
# Line 1832  Line 1757 
1757  Index in each tuple of the column to be matched against the pattern. If the match is  Index in each tuple of the column to be matched against the pattern. If the match is
1758  successful, an output record will be generated.  successful, an output record will be generated.
1759    
1760  =item idxValue  =item idxValues
1761    
1762  Index in each tuple of the column to be put as the second column of the output.  Reference to a list containing the indexes in each tuple of the columns to be put as
1763    the second column of the output.
1764    
1765  =item pattern  =item pattern
1766    
# Line 1857  Line 1783 
1783    
1784  sub SpecialAttribute {  sub SpecialAttribute {
1785      # Get the parameters.      # Get the parameters.
1786      my ($id, $attributes, $idxMatch, $idxValue, $pattern, $loader) = @_;      my ($id, $attributes, $idxMatch, $idxValues, $pattern, $loader) = @_;
1787      # Declare the return variable.      # Declare the return variable.
1788      my $retVal = 0;      my $retVal = 0;
1789      # Loop through the attribute rows.      # Loop through the attribute rows.
1790      for my $row (@{$attributes}) {      for my $row (@{$attributes}) {
1791          # Check for a match.          # Check for a match.
1792          if ($row->[$idxMatch] =~ m/$pattern/i) {          if ($row->[$idxMatch] =~ m/$pattern/i) {
1793              # We have a match, so output a row.              # We have a match, so output a row. This is a bit tricky, since we may
1794              $loader->Put($id, $row->[$idxValue]);              # be putting out multiple columns of data from the input.
1795                my $value = join(" ", map { $row->[$_] } @{$idxValues});
1796                $loader->Put($id, $value);
1797              $retVal++;              $retVal++;
1798          }          }
1799      }      }
# Line 1874  Line 1802 
1802      return $retVal;      return $retVal;
1803  }  }
1804    
 =head3 CreatePDB  
   
 C<< $loader->CreatePDB($pdbID, $title, $type, \%pdbHash); >>  
   
 Insure that a PDB record exists for the identified PDB. If one does not exist, it will be  
 created.  
   
 =over 4  
   
 =item pdbID  
   
 ID string (usually an unqualified file name) for the desired PDB.  
   
 =item title  
   
 Title to use if the PDB must be created.  
   
 =item type  
   
 Type of PDB: C<free> or C<bound>  
   
 =item pdbHash  
   
 Hash containing the IDs of PDBs that have already been created.  
   
 =item pdbLoader  
   
 Load object for the PDB table.  
   
 =back  
   
 =cut  
   
 sub CreatePDB {  
     # Get the parameters.  
     my ($self, $pdbID, $title, $type, $pdbHash, $pdbLoader) = @_;  
     $pdbLoader->Add("PDB check");  
     # Check to see if this is a new PDB.  
     if (! exists $pdbHash->{$pdbID}) {  
         # It is, so we create it.  
         $pdbLoader->Put($pdbID, $title, $type);  
         $pdbHash->{$pdbID} = 1;  
         # Count it.  
         $pdbLoader->Add("PDB-$type");  
     }  
 }  
   
1805  =head3 TableLoader  =head3 TableLoader
1806    
1807  Create an ERDBLoad object for the specified table. The object is also added to  Create an ERDBLoad object for the specified table. The object is also added to
# Line 1935  Line 1816 
1816    
1817  Name of the table (relation) being loaded.  Name of the table (relation) being loaded.
1818    
 =item ignore  
   
 TRUE if the table should be ignored entirely, else FALSE.  
   
1819  =item RETURN  =item RETURN
1820    
1821  Returns an ERDBLoad object for loading the specified table.  Returns an ERDBLoad object for loading the specified table.
# Line 1949  Line 1826 
1826    
1827  sub _TableLoader {  sub _TableLoader {
1828      # Get the parameters.      # Get the parameters.
1829      my ($self, $tableName, $ignore) = @_;      my ($self, $tableName) = @_;
1830      # Create the load object.      # Create the load object.
1831      my $retVal = ERDBLoad->new($self->{erdb}, $tableName, $self->{loadDirectory}, $self->LoadOnly,      my $retVal = ERDBLoad->new($self->{erdb}, $tableName, $self->{loadDirectory}, $self->LoadOnly);
                                $ignore);  
1832      # Cache it in the loader list.      # Cache it in the loader list.
1833      push @{$self->{loaders}}, $retVal;      push @{$self->{loaders}}, $retVal;
1834      # Return it to the caller.      # Return it to the caller.
# Line 2024  Line 1900 
1900      return $retVal;      return $retVal;
1901  }  }
1902    
1903    =head3 GetGenomeAttributes
1904    
1905    C<< my $aHashRef = GetGenomeAttributes($fig, $genomeID, \@fids, \@propKeys); >>
1906    
1907    Return a hash of attributes keyed on feature ID. This method gets all the NMPDR-related
1908    attributes for all the features of a genome in a single call, then organizes them into
1909    a hash.
1910    
1911    =over 4
1912    
1913    =item fig
1914    
1915    FIG-like object for accessing attributes.
1916    
1917    =item genomeID
1918    
1919    ID of the genome who's attributes are desired.
1920    
1921    =item fids
1922    
1923    Reference to a list of the feature IDs whose attributes are to be kept.
1924    
1925    =item propKeys
1926    
1927    A list of the keys to retrieve.
1928    
1929    =item RETURN
1930    
1931    Returns a reference to a hash. The key of the hash is the feature ID. The value is the
1932    reference to a list of the feature's attribute tuples. Each tuple contains the feature ID,
1933    the attribute key, and one or more attribute values.
1934    
1935    =back
1936    
1937    =cut
1938    
1939    sub GetGenomeAttributes {
1940        # Get the parameters.
1941        my ($fig, $genomeID, $fids, $propKeys) = @_;
1942        # Declare the return variable.
1943        my $retVal = {};
1944        # Initialize the hash. This not only enables us to easily determine which FIDs to
1945        # keep, it insures that the caller sees a list reference for every known fid,
1946        # simplifying the logic.
1947        for my $fid (@{$fids}) {
1948            $retVal->{$fid} = [];
1949        }
1950        # Get the attributes. If ev_code_cron is running, we may get a timeout error, so
1951        # an eval is used.
1952        my @aList = ();
1953        eval {
1954            @aList = $fig->get_attributes("fig|$genomeID%", $propKeys);
1955            Trace(scalar(@aList) . " attributes returned for genome $genomeID.") if T(3);
1956        };
1957        # Check for a problem.
1958        if ($@) {
1959            Trace("Retrying attributes for $genomeID due to error: $@") if T(1);
1960            # Our fallback plan is to process the attributes in blocks of 100. This is much slower,
1961            # but allows us to continue processing.
1962            my $nFids = scalar @{$fids};
1963            for (my $i = 0; $i < $nFids; $i += 100) {
1964                # Determine the index of the last feature ID we'll be specifying on this pass.
1965                # Normally it's $i + 99, but if we're close to the end it may be less.
1966                my $end = ($i + 100 > $nFids ? $nFids - 1 : $i + 99);
1967                # Get a slice of the fid list.
1968                my @slice = @{$fids}[$i .. $end];
1969                # Get the relevant attributes.
1970                Trace("Retrieving attributes for fids $i to $end.") if T(3);
1971                my @aShort = $fig->get_attributes(\@slice, $propKeys);
1972                Trace(scalar(@aShort) . " attributes returned for fids $i to $end.") if T(3);
1973                push @aList, @aShort;
1974            }
1975        }
1976        # Now we should have all the interesting attributes in @aList. Populate the hash with
1977        # them.
1978        for my $aListEntry (@aList) {
1979            my $fid = $aListEntry->[0];
1980            if (exists $retVal->{$fid}) {
1981                push @{$retVal->{$fid}}, $aListEntry;
1982            }
1983        }
1984        # Return the result.
1985        return $retVal;
1986    }
1987    
1988    
1989  1;  1;

Legend:
Removed from v.1.76  
changed lines
  Added in v.1.88

MCS Webmaster
ViewVC Help
Powered by ViewVC 1.0.3