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

Diff of /Sprout/SproutLoad.pm

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

revision 1.62, Sun Jul 30 05:44:57 2006 UTC revision 1.92, Sun Mar 23 16:33:15 2008 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 50  Line 52 
52    
53  =head3 new  =head3 new
54    
55  C<< my $spl = SproutLoad->new($sprout, $fig, $genomeFile, $subsysFile, $options); >>      my $spl = SproutLoad->new($sprout, $fig, $genomeFile, $subsysFile, $options);
56    
57  Construct a new Sprout Loader object, specifying the two participating databases and  Construct a new Sprout Loader object, specifying the two participating databases and
58  the name of the files containing the list of genomes and subsystems to use.  the name of the files containing the list of genomes and subsystems to use.
# Line 80  Line 82 
82  Either the name of the file containing the list of trusted subsystems or a reference  Either the name of the file containing the list of trusted subsystems or a reference
83  to a list of subsystem names. If nothing is specified, all NMPDR subsystems will be  to a list of subsystem names. If nothing is specified, all NMPDR subsystems will be
84  considered trusted. (A subsystem is considered NMPDR if it has a file named C<NMPDR>  considered trusted. (A subsystem is considered NMPDR if it has a file named C<NMPDR>
85  in its data directory.) Only subsystem data related to the trusted subsystems is loaded.  in its data directory.) Only subsystem data related to the NMPDR subsystems is loaded.
86    
87  =item options  =item options
88    
# 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 120  Line 123 
123                      # an omitted access code can be defaulted to 1.                      # an omitted access code can be defaulted to 1.
124                      for my $genomeLine (@genomeList) {                      for my $genomeLine (@genomeList) {
125                          my ($genomeID, $accessCode) = split("\t", $genomeLine);                          my ($genomeID, $accessCode) = split("\t", $genomeLine);
126                          if (undef $accessCode) {                          if (! defined($accessCode)) {
127                              $accessCode = 1;                              $accessCode = 1;
128                          }                          }
129                          $genomes{$genomeID} = $accessCode;                          $genomes{$genomeID} = $accessCode;
# Line 138  Line 141 
141          if (! defined $subsysFile || $subsysFile eq '') {          if (! defined $subsysFile || $subsysFile eq '') {
142              # Here we want all the usable subsystems. First we get the whole list.              # Here we want all the usable subsystems. First we get the whole list.
143              my @subs = $fig->all_subsystems();              my @subs = $fig->all_subsystems();
144              # Loop through, checking for usability.              # Loop through, checking for the NMPDR file.
145              for my $sub (@subs) {              for my $sub (@subs) {
146                  if ($fig->usable_subsystem($sub)) {                  if ($fig->nmpdr_subsystem($sub)) {
147                      $subsystems{$sub} = 1;                      $subsystems{$sub} = 1;
148                  }                  }
149              }              }
# Line 163  Line 166 
166                  Confess("Invalid subsystem parameter in SproutLoad constructor.");                  Confess("Invalid subsystem parameter in SproutLoad constructor.");
167              }              }
168          }          }
169            # Go through the subsys hash again, creating the keyword list for each subsystem.
170            for my $subsystem (keys %subsystems) {
171                my $name = $subsystem;
172                $name =~ s/_/ /g;
173                $subsystems{$subsystem} = $name;
174      }      }
175        }
176        # Get the list of NMPDR-oriented attribute keys.
177        my @propKeys = $fig->get_group_keys("NMPDR");
178      # Get the data directory from the Sprout object.      # Get the data directory from the Sprout object.
179      my ($directory) = $sprout->LoadInfo();      my ($directory) = $sprout->LoadInfo();
180      # Create the Sprout load object.      # Create the Sprout load object.
# Line 175  Line 186 
186                    loadDirectory => $directory,                    loadDirectory => $directory,
187                    erdb => $sprout,                    erdb => $sprout,
188                    loaders => [],                    loaders => [],
189                    options => $options                    options => $options,
190                      propKeys => \@propKeys,
191                   };                   };
192      # Bless and return it.      # Bless and return it.
193      bless $retVal, $class;      bless $retVal, $class;
# Line 184  Line 196 
196    
197  =head3 LoadOnly  =head3 LoadOnly
198    
199  C<< my $flag = $spl->LoadOnly; >>      my $flag = $spl->LoadOnly;
200    
201  Return TRUE if we are in load-only mode, else FALSE.  Return TRUE if we are in load-only mode, else FALSE.
202    
# Line 195  Line 207 
207      return $self->{options}->{loadOnly};      return $self->{options}->{loadOnly};
208  }  }
209    
 =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};  
 }  
210    
211  =head3 LoadGenomeData  =head3 LoadGenomeData
212    
213  C<< my $stats = $spl->LoadGenomeData(); >>      my $stats = $spl->LoadGenomeData();
214    
215  Load the Genome, Contig, and Sequence data from FIG into Sprout.  Load the Genome, Contig, and Sequence data from FIG into Sprout.
216    
# Line 247  Line 247 
247      my $genomeCount = (keys %{$genomeHash});      my $genomeCount = (keys %{$genomeHash});
248      # Create load objects for each of the tables we're loading.      # Create load objects for each of the tables we're loading.
249      my $loadGenome = $self->_TableLoader('Genome');      my $loadGenome = $self->_TableLoader('Genome');
250      my $loadHasContig = $self->_TableLoader('HasContig', $self->PrimaryOnly);      my $loadHasContig = $self->_TableLoader('HasContig');
251      my $loadContig = $self->_TableLoader('Contig', $self->PrimaryOnly);      my $loadContig = $self->_TableLoader('Contig');
252      my $loadIsMadeUpOf = $self->_TableLoader('IsMadeUpOf', $self->PrimaryOnly);      my $loadIsMadeUpOf = $self->_TableLoader('IsMadeUpOf');
253      my $loadSequence = $self->_TableLoader('Sequence', $self->PrimaryOnly);      my $loadSequence = $self->_TableLoader('Sequence');
254      if ($self->{options}->{loadOnly}) {      if ($self->{options}->{loadOnly}) {
255          Trace("Loading from existing files.") if T(2);          Trace("Loading from existing files.") if T(2);
256      } else {      } else {
257          Trace("Generating genome data.") if T(2);          Trace("Generating genome data.") if T(2);
258            # Get the full info for the FIG genomes.
259            my %genomeInfo = map { $_->[0] => { gname => $_->[1], szdna => $_->[2], maindomain => $_->[3],
260                                                pegs => $_->[4], rnas => $_->[5], complete => $_->[6] } } @{$fig->genome_info()};
261          # Now we loop through the genomes, generating the data for each one.          # Now we loop through the genomes, generating the data for each one.
262          for my $genomeID (sort keys %{$genomeHash}) {          for my $genomeID (sort keys %{$genomeHash}) {
263              Trace("Generating data for genome $genomeID.") if T(3);              Trace("Generating data for genome $genomeID.") if T(3);
# Line 266  Line 269 
269              my $extra = join " ", @extraData;              my $extra = join " ", @extraData;
270              # Get the full taxonomy.              # Get the full taxonomy.
271              my $taxonomy = $fig->taxonomy_of($genomeID);              my $taxonomy = $fig->taxonomy_of($genomeID);
272                # Get the version. If no version is specified, we default to the genome ID by itself.
273                my $version = $fig->genome_version($genomeID);
274                if (! defined($version)) {
275                    $version = $genomeID;
276                }
277                # Get the DNA size.
278                my $dnaSize = $fig->genome_szdna($genomeID);
279                # Open the NMPDR group file for this genome.
280                my $group;
281                if (open(TMP, "<$FIG_Config::organisms/$genomeID/NMPDR") &&
282                    defined($group = <TMP>)) {
283                    # Clean the line ending.
284                    chomp $group;
285                } else {
286                    # No group, so use the default.
287                    $group = $FIG_Config::otherGroup;
288                }
289                close TMP;
290                # Get the contigs.
291                my @contigs = $fig->all_contigs($genomeID);
292                # Get this genome's info array.
293                my $info = $genomeInfo{$genomeID};
294              # Output the genome record.              # Output the genome record.
295              $loadGenome->Put($genomeID, $accessCode, $fig->is_complete($genomeID), $genus,              $loadGenome->Put($genomeID, $accessCode, $info->{complete}, scalar(@contigs),
296                               $species, $extra, $taxonomy);                               $dnaSize, $genus, $info->{pegs}, $group, $info->{rnas}, $species, $extra, $version, $taxonomy);
297              # Now we loop through each of the genome's contigs.              # Now we loop through each of the genome's contigs.
             my @contigs = $fig->all_contigs($genomeID);  
298              for my $contigID (@contigs) {              for my $contigID (@contigs) {
299                  Trace("Processing contig $contigID for $genomeID.") if T(4);                  Trace("Processing contig $contigID for $genomeID.") if T(4);
300                  $loadContig->Add("contigIn");                  $loadContig->Add("contigIn");
# Line 306  Line 330 
330      return $retVal;      return $retVal;
331  }  }
332    
 =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;  
 }  
   
333  =head3 LoadFeatureData  =head3 LoadFeatureData
334    
335  C<< my $stats = $spl->LoadFeatureData(); >>      my $stats = $spl->LoadFeatureData();
336    
337  Load the feature data from FIG into Sprout.  Load the feature data from FIG into Sprout.
338    
# Line 444  Line 342 
342    
343      Feature      Feature
344      FeatureAlias      FeatureAlias
345        IsAliasOf
346      FeatureLink      FeatureLink
347      FeatureTranslation      FeatureTranslation
348      FeatureUpstream      FeatureUpstream
349      IsLocatedIn      IsLocatedIn
350      HasFeature      HasFeature
351        HasRoleInSubsystem
352        FeatureEssential
353        FeatureVirulent
354        FeatureIEDB
355        CDD
356        IsPresentOnProteinOf
357    
358  =over 4  =over 4
359    
# Line 463  Line 368 
368  sub LoadFeatureData {  sub LoadFeatureData {
369      # Get this object instance.      # Get this object instance.
370      my ($self) = @_;      my ($self) = @_;
371      # Get the FIG object.      # Get the FIG and Sprout objects.
372      my $fig = $self->{fig};      my $fig = $self->{fig};
373        my $sprout = $self->{sprout};
374      # Get the table of genome IDs.      # Get the table of genome IDs.
375      my $genomeHash = $self->{genomes};      my $genomeHash = $self->{genomes};
376      # Create load objects for each of the tables we're loading.      # Create load objects for each of the tables we're loading.
377      my $loadFeature = $self->_TableLoader('Feature');      my $loadFeature = $self->_TableLoader('Feature');
378      my $loadIsLocatedIn = $self->_TableLoader('IsLocatedIn', $self->PrimaryOnly);      my $loadIsLocatedIn = $self->_TableLoader('IsLocatedIn');
379      my $loadFeatureAlias = $self->_TableLoader('FeatureAlias');      my $loadFeatureAlias = $self->_TableLoader('FeatureAlias');
380        my $loadIsAliasOf = $self->_TableLoader('IsAliasOf');
381      my $loadFeatureLink = $self->_TableLoader('FeatureLink');      my $loadFeatureLink = $self->_TableLoader('FeatureLink');
382      my $loadFeatureTranslation = $self->_TableLoader('FeatureTranslation');      my $loadFeatureTranslation = $self->_TableLoader('FeatureTranslation');
383      my $loadFeatureUpstream = $self->_TableLoader('FeatureUpstream');      my $loadFeatureUpstream = $self->_TableLoader('FeatureUpstream');
384      my $loadHasFeature = $self->_TableLoader('HasFeature');      my $loadHasFeature = $self->_TableLoader('HasFeature');
385        my $loadHasRoleInSubsystem = $self->_TableLoader('HasRoleInSubsystem');
386        my $loadFeatureEssential = $self->_TableLoader('FeatureEssential');
387        my $loadFeatureVirulent = $self->_TableLoader('FeatureVirulent');
388        my $loadFeatureIEDB = $self->_TableLoader('FeatureIEDB');
389        my $loadCDD = $self->_TableLoader('CDD');
390        my $loadIsPresentOnProteinOf = $self->_TableLoader('IsPresentOnProteinOf');
391        # Get the subsystem hash.
392        my $subHash = $self->{subsystems};
393        # Get the property keys.
394        my $propKeys = $self->{propKeys};
395        # Create a hashes to hold CDD and alias values.
396        my %CDD = ();
397        my %alias = ();
398      # 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
399      # locations.      # locations.
400      my $chunkSize = $self->{sprout}->MaxSegment();      my $chunkSize = $self->{sprout}->MaxSegment();
# Line 483  Line 403 
403      } else {      } else {
404          Trace("Generating feature data.") if T(2);          Trace("Generating feature data.") if T(2);
405          # Now we loop through the genomes, generating the data for each one.          # Now we loop through the genomes, generating the data for each one.
406          for my $genomeID (sort keys %{$genomeHash}) {          my @allGenomes = sort keys %{$genomeHash};
407            Trace(scalar(@allGenomes) . " genomes found in list.") if T(3);
408            for my $genomeID (@allGenomes) {
409              Trace("Loading features for genome $genomeID.") if T(3);              Trace("Loading features for genome $genomeID.") if T(3);
410              $loadFeature->Add("genomeIn");              $loadFeature->Add("genomeIn");
411              # Get the feature list for this genome.              # Get the feature list for this genome.
412              my $features = $fig->all_features_detailed($genomeID);              my $features = $fig->all_features_detailed_fast($genomeID);
413              # Sort and count the list.              # Sort and count the list.
414              my @featureTuples = sort { $a->[0] cmp $b->[0] } @{$features};              my @featureTuples = sort { $a->[0] cmp $b->[0] } @{$features};
415              my $count = scalar @featureTuples;              my $count = scalar @featureTuples;
416                my @fids = map { $_->[0] } @featureTuples;
417              Trace("$count features found for genome $genomeID.") if T(3);              Trace("$count features found for genome $genomeID.") if T(3);
418                # Get the attributes for this genome and put them in a hash by feature ID.
419                my $attributes = GetGenomeAttributes($fig, $genomeID, \@fids, $propKeys);
420                Trace("Looping through features for $genomeID.") if T(3);
421              # Set up for our duplicate-feature check.              # Set up for our duplicate-feature check.
422              my $oldFeatureID = "";              my $oldFeatureID = "";
423              # Loop through the features.              # Loop through the features.
424              for my $featureTuple (@featureTuples) {              for my $featureTuple (@featureTuples) {
425                  # Split the tuple.                  # Split the tuple.
426                  my ($featureID, $locations, undef, $type) = @{$featureTuple};                  my ($featureID, $locations, undef, $type, $minloc, $maxloc, $assignment, $user, $quality) = @{$featureTuple};
427                  # Check for duplicates.                  # Check for duplicates.
428                  if ($featureID eq $oldFeatureID) {                  if ($featureID eq $oldFeatureID) {
429                      Trace("Duplicate feature $featureID found.") if T(1);                      Trace("Duplicate feature $featureID found.") if T(1);
# Line 505  Line 431 
431                      $oldFeatureID = $featureID;                      $oldFeatureID = $featureID;
432                      # Count this feature.                      # Count this feature.
433                      $loadFeature->Add("featureIn");                      $loadFeature->Add("featureIn");
434                      # Create the feature record.                      # Fix the quality. It is almost always a space, but some odd stuff might sneak through, and the
435                      $loadFeature->Put($featureID, 1, $type);                      # Sprout database requires a single character.
436                      # Link it to the parent genome.                      if (! defined($quality) || $quality eq "") {
437                      $loadHasFeature->Put($genomeID, $featureID, $type);                          $quality = " ";
438                        }
439                        # Begin building the keywords. We start with the genome ID, the
440                        # feature ID, the taxonomy, and the organism name.
441                        my @keywords = ($genomeID, $featureID, $fig->genus_species($genomeID),
442                                        $fig->taxonomy_of($genomeID));
443                      # Create the aliases.                      # Create the aliases.
444                      for my $alias ($fig->feature_aliases($featureID)) {                      for my $alias ($fig->feature_aliases($featureID)) {
445                          $loadFeatureAlias->Put($featureID, $alias);                          #Connect this alias to this feature.
446                      }                          $loadIsAliasOf->Put($alias, $featureID);
447                            push @keywords, $alias;
448                            # If this is a locus tag, also add its natural form as a keyword.
449                            my $naturalName = AliasAnalysis::Type(LocusTag => $alias);
450                            if ($naturalName) {
451                                push @keywords, $naturalName;
452                            }
453                            # If this is the first time for the specified alias, create its
454                            # alias record.
455                            if (! exists $alias{$alias}) {
456                                $loadFeatureAlias->Put($alias);
457                                $alias{$alias} = 1;
458                            }
459                        }
460                        # Add the corresponding IDs. Note we have to remove the FIG ID from the
461                        # return list. It's already among the keywords.
462                        my @corresponders = grep { $_ !~ /^fig/} $fig->get_corresponding_ids($featureID);
463                        push @keywords, @corresponders;
464                        Trace("Assignment for $featureID is: $assignment") if T(4);
465                        # Break the assignment into words and shove it onto the
466                        # keyword list.
467                        push @keywords, split(/\s+/, $assignment);
468                        # Link this feature to the parent genome.
469                        $loadHasFeature->Put($genomeID, $featureID, $type);
470                      # Get the links.                      # Get the links.
471                      my @links = $fig->fid_links($featureID);                      my @links = $fig->fid_links($featureID);
472                      for my $link (@links) {                      for my $link (@links) {
# Line 531  Line 485 
485                              $loadFeatureUpstream->Put($featureID, $upstream);                              $loadFeatureUpstream->Put($featureID, $upstream);
486                          }                          }
487                      }                      }
488                        # Now we need to find the subsystems this feature participates in.
489                        # We also add the subsystems to the keyword list. Before we do that,
490                        # we must convert underscores to spaces.
491                        my @subsystems = $fig->peg_to_subsystems($featureID);
492                        for my $subsystem (@subsystems) {
493                            # Only proceed if we like this subsystem.
494                            if (exists $subHash->{$subsystem}) {
495                                # Store the has-role link.
496                                $loadHasRoleInSubsystem->Put($featureID, $subsystem, $genomeID, $type);
497                                # Save the subsystem's keyword data.
498                                my $subKeywords = $subHash->{$subsystem};
499                                push @keywords, split /\s+/, $subKeywords;
500                                # Now we need to get this feature's role in the subsystem.
501                                my $subObject = $fig->get_subsystem($subsystem);
502                                my @roleColumns = $subObject->get_peg_roles($featureID);
503                                my @allRoles = $subObject->get_roles();
504                                for my $col (@roleColumns) {
505                                    my $role = $allRoles[$col];
506                                    push @keywords, split /\s+/, $role;
507                                    push @keywords, $subObject->get_role_abbr($col);
508                                }
509                            }
510                        }
511                        # There are three special attributes computed from property
512                        # data that we build next. If the special attribute is non-empty,
513                        # its name will be added to the keyword list. First, we get all
514                        # the attributes for this feature. They will come back as
515                        # 4-tuples: [peg, name, value, URL]. We use a 3-tuple instead:
516                        # [name, value, value with URL]. (We don't need the PEG, since
517                        # we already know it.)
518                        my @attributes = map { [$_->[1], $_->[2], Tracer::CombineURL($_->[2], $_->[3])] }
519                                             @{$attributes->{$featureID}};
520                        # Now we process each of the special attributes.
521                        if (SpecialAttribute($featureID, \@attributes,
522                                             1, [0,2], '^(essential|potential_essential)$',
523                                             $loadFeatureEssential)) {
524                            push @keywords, 'essential';
525                            $loadFeature->Add('essential');
526                        }
527                        if (SpecialAttribute($featureID, \@attributes,
528                                             0, [2], '^virulen',
529                                             $loadFeatureVirulent)) {
530                            push @keywords, 'virulent';
531                            $loadFeature->Add('virulent');
532                        }
533                        if (SpecialAttribute($featureID, \@attributes,
534                                             0, [0,2], '^iedb_',
535                                             $loadFeatureIEDB)) {
536                            push @keywords, 'iedb';
537                            $loadFeature->Add('iedb');
538                        }
539                        # Now we have some other attributes we need to process. Currently,
540                        # this is CDD and CELLO, but we expect the number to increase.
541                        my %attributeHash = ();
542                        for my $attrRow (@{$attributes->{$featureID}}) {
543                            my (undef, $key, @values) = @{$attrRow};
544                            $key =~ /^([^:]+)::(.+)/;
545                            if (exists $attributeHash{$1}) {
546                                $attributeHash{$1}->{$2} = \@values;
547                            } else {
548                                $attributeHash{$1} = {$2 => \@values};
549                            }
550                        }
551                        my $celloValue = "unknown";
552                        # Pull in the CELLO attribute. There will never be more than one.
553                        # If we have one, it's a feature attribute AND a keyword.
554                        my @celloData = keys %{$attributeHash{CELLO}};
555                        if (@celloData) {
556                            $celloValue = $celloData[0];
557                            push @keywords, $celloValue;
558                        }
559                        # Now we handle CDD. This is a bit more complicated, because
560                        # there are multiple CDDs per protein.
561                        if (exists $attributeHash{CDD}) {
562                            # Get the hash of CDD IDs to scores for this feature. We
563                            # already know it exists because of the above IF.
564                            my $cddHash = $attributeHash{CDD};
565                            my @cddData = sort keys %{$cddHash};
566                            for my $cdd (@cddData) {
567                                # Extract the score for this CDD and decode it.
568                                my ($codeScore) = split(/\s*,\s*/, $cddHash->{$cdd}->[1]);
569                                my $realScore = FIGRules::DecodeScore($codeScore);
570                                # We can't afford to crash because of a bad attribute
571                                # value, hence the IF below.
572                                if (! defined($realScore)) {
573                                    # Bad score, so count it.
574                                    $loadFeature->Add('badCDDscore');
575                                } else {
576                                    # Create the connection.
577                                    $loadIsPresentOnProteinOf->Put($cdd, $featureID, $realScore);
578                                    # If this CDD does not yet exist, create its record.
579                                    if (! exists $CDD{$cdd}) {
580                                        $CDD{$cdd} = 1;
581                                        $loadCDD->Put($cdd);
582                                    }
583                                }
584                            }
585                        }
586                        # Now we need to bust up hyphenated words in the keyword
587                        # list. We keep them separate and put them at the end so
588                        # the original word order is available.
589                        my $keywordString = "";
590                        my $bustedString = "";
591                        for my $keyword (@keywords) {
592                            if (length $keyword >= 3) {
593                                $keywordString .= " $keyword";
594                                if ($keyword =~ /-/) {
595                                    my @words = split /-/, $keyword;
596                                    $bustedString .= join(" ", "", @words);
597                                }
598                            }
599                        }
600                        $keywordString .= $bustedString;
601                        # Get rid of annoying punctuation.
602                        $keywordString =~ s/[();]//g;
603                        # Clean the keyword list.
604                        my $cleanWords = $sprout->CleanKeywords($keywordString);
605                        Trace("Keyword string for $featureID: $cleanWords") if T(4);
606                        # Now we need to process the feature's locations. First, we split them up.
607                        my @locationList = split /\s*,\s*/, $locations;
608                        # Next, we convert them to Sprout location objects.
609                        my @locObjectList = map { BasicLocation->new("$genomeID:$_") } @locationList;
610                        # Assemble them into a sprout location string for later.
611                        my $locationString = join(", ", map { $_->String } @locObjectList);
612                      # 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
613                      # 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
614                      # the maximum segment size. This simplifies the genes_in_region processing                      # the maximum segment size. This simplifies the genes_in_region processing
615                      # for Sprout.                      # for Sprout. To start, we create the location position indicator.
                     my @locationList = split /\s*,\s*/, $locations;  
                     # Create the location position indicator.  
616                      my $i = 1;                      my $i = 1;
617                      # Loop through the locations.                      # Loop through the locations.
618                      for my $location (@locationList) {                      for my $locObject (@locObjectList) {
619                          # Parse the location.                          # Split this location into a list of chunks.
                         my $locObject = BasicLocation->new("$genomeID:$location");  
                         # Split it into a list of chunks.  
620                          my @locOList = ();                          my @locOList = ();
621                          while (my $peeling = $locObject->Peel($chunkSize)) {                          while (my $peeling = $locObject->Peel($chunkSize)) {
622                              $loadIsLocatedIn->Add("peeling");                              $loadIsLocatedIn->Add("peeling");
# Line 557  Line 631 
631                              $i++;                              $i++;
632                          }                          }
633                      }                      }
634                        # Finally, reassemble the location objects into a list of Sprout location strings.
635                        # Create the feature record.
636                        $loadFeature->Put($featureID, 1, $user, $quality, $celloValue, $type, $assignment, $cleanWords, $locationString);
637                  }                  }
638              }              }
639          }              Trace("Genome $genomeID processed.") if T(3);
     }  
     # Finish the loads.  
     my $retVal = $self->_FinishAll();  
     return $retVal;  
 }  
   
 =head3 LoadBBHData  
   
 C<< my $stats = $spl->LoadBBHData(); >>  
   
 Load the bidirectional best hit data from FIG into Sprout.  
   
 Sprout does not store information on similarities. Instead, it has only the  
 bi-directional best hits. Even so, the BBH table is one of the largest in  
 the database.  
   
 The following relations are loaded by this method.  
   
     IsBidirectionalBestHitOf  
   
 =over 4  
   
 =item RETURNS  
   
 Returns a statistics object for the loads.  
   
 =back  
   
 =cut  
 #: Return Type $%;  
 sub LoadBBHData {  
     # Get this object instance.  
     my ($self) = @_;  
     # Get the FIG object.  
     my $fig = $self->{fig};  
     # Get the table of genome IDs.  
     my $genomeHash = $self->{genomes};  
     # Create load objects for each of the tables we're loading.  
     my $loadIsBidirectionalBestHitOf = $self->_TableLoader('IsBidirectionalBestHitOf');  
     if ($self->{options}->{loadOnly}) {  
         Trace("Loading from existing files.") if T(2);  
     } else {  
         Trace("Generating BBH data.") if T(2);  
         # Now we loop through the genomes, generating the data for each one.  
         for my $genomeID (sort keys %{$genomeHash}) {  
             $loadIsBidirectionalBestHitOf->Add("genomeIn");  
             Trace("Processing features for genome $genomeID.") if T(3);  
             # Get the feature list for this genome.  
             my $features = $fig->all_features_detailed($genomeID);  
             # Loop through the features.  
             for my $featureData (@{$features}) {  
                 # Split the tuple.  
                 my ($featureID, $locations, $aliases, $type) = @{$featureData};  
                 # Get the bi-directional best hits.  
                 my @bbhList = $fig->bbhs($featureID);  
                 for my $bbhEntry (@bbhList) {  
                     # Get the target feature ID and the score.  
                     my ($targetID, $score) = @{$bbhEntry};  
                     # Check the target feature's genome.  
                     my $targetGenomeID = $fig->genome_of($targetID);  
                     # Only proceed if it's one of our genomes.  
                     if ($genomeHash->{$targetGenomeID}) {  
                         $loadIsBidirectionalBestHitOf->Put($featureID, $targetID, $targetGenomeID,  
                                                            $score);  
                     }  
                 }  
             }  
640          }          }
641      }      }
642      # Finish the loads.      # Finish the loads.
# Line 636  Line 646 
646    
647  =head3 LoadSubsystemData  =head3 LoadSubsystemData
648    
649  C<< my $stats = $spl->LoadSubsystemData(); >>      my $stats = $spl->LoadSubsystemData();
650    
651  Load the subsystem data from FIG into Sprout.  Load the subsystem data from FIG into Sprout.
652    
# Line 652  Line 662 
662      SubsystemClass      SubsystemClass
663      Role      Role
664      RoleEC      RoleEC
665        IsIdentifiedByEC
666      SSCell      SSCell
667      ContainsFeature      ContainsFeature
668      IsGenomeOf      IsGenomeOf
# Line 693  Line 704 
704      # Get the map list.      # Get the map list.
705      my @maps = $fig->all_maps;      my @maps = $fig->all_maps;
706      # Create load objects for each of the tables we're loading.      # Create load objects for each of the tables we're loading.
707      my $loadDiagram = $self->_TableLoader('Diagram', $self->PrimaryOnly);      my $loadDiagram = $self->_TableLoader('Diagram');
708      my $loadRoleOccursIn = $self->_TableLoader('RoleOccursIn', $self->PrimaryOnly);      my $loadRoleOccursIn = $self->_TableLoader('RoleOccursIn');
709      my $loadSubsystem = $self->_TableLoader('Subsystem');      my $loadSubsystem = $self->_TableLoader('Subsystem');
710      my $loadRole = $self->_TableLoader('Role', $self->PrimaryOnly);      my $loadRole = $self->_TableLoader('Role');
711      my $loadRoleEC = $self->_TableLoader('RoleEC', $self->PrimaryOnly);      my $loadRoleEC = $self->_TableLoader('RoleEC');
712      my $loadCatalyzes = $self->_TableLoader('Catalyzes', $self->PrimaryOnly);      my $loadIsIdentifiedByEC = $self->_TableLoader('IsIdentifiedByEC');
713      my $loadSSCell = $self->_TableLoader('SSCell', $self->PrimaryOnly);      my $loadCatalyzes = $self->_TableLoader('Catalyzes');
714      my $loadContainsFeature = $self->_TableLoader('ContainsFeature', $self->PrimaryOnly);      my $loadSSCell = $self->_TableLoader('SSCell');
715      my $loadIsGenomeOf = $self->_TableLoader('IsGenomeOf', $self->PrimaryOnly);      my $loadContainsFeature = $self->_TableLoader('ContainsFeature');
716      my $loadIsRoleOf = $self->_TableLoader('IsRoleOf', $self->PrimaryOnly);      my $loadIsGenomeOf = $self->_TableLoader('IsGenomeOf');
717      my $loadOccursInSubsystem = $self->_TableLoader('OccursInSubsystem', $self->PrimaryOnly);      my $loadIsRoleOf = $self->_TableLoader('IsRoleOf');
718      my $loadParticipatesIn = $self->_TableLoader('ParticipatesIn', $self->PrimaryOnly);      my $loadOccursInSubsystem = $self->_TableLoader('OccursInSubsystem');
719      my $loadHasSSCell = $self->_TableLoader('HasSSCell', $self->PrimaryOnly);      my $loadParticipatesIn = $self->_TableLoader('ParticipatesIn');
720      my $loadRoleSubset = $self->_TableLoader('RoleSubset', $self->PrimaryOnly);      my $loadHasSSCell = $self->_TableLoader('HasSSCell');
721      my $loadGenomeSubset = $self->_TableLoader('GenomeSubset', $self->PrimaryOnly);      my $loadRoleSubset = $self->_TableLoader('RoleSubset');
722      my $loadConsistsOfRoles = $self->_TableLoader('ConsistsOfRoles', $self->PrimaryOnly);      my $loadGenomeSubset = $self->_TableLoader('GenomeSubset');
723      my $loadConsistsOfGenomes = $self->_TableLoader('ConsistsOfGenomes', $self->PrimaryOnly);      my $loadConsistsOfRoles = $self->_TableLoader('ConsistsOfRoles');
724      my $loadHasRoleSubset = $self->_TableLoader('HasRoleSubset', $self->PrimaryOnly);      my $loadConsistsOfGenomes = $self->_TableLoader('ConsistsOfGenomes');
725      my $loadHasGenomeSubset = $self->_TableLoader('HasGenomeSubset', $self->PrimaryOnly);      my $loadHasRoleSubset = $self->_TableLoader('HasRoleSubset');
726      my $loadSubsystemClass = $self->_TableLoader('SubsystemClass', $self->PrimaryOnly);      my $loadHasGenomeSubset = $self->_TableLoader('HasGenomeSubset');
727        my $loadSubsystemClass = $self->_TableLoader('SubsystemClass');
728      if ($self->{options}->{loadOnly}) {      if ($self->{options}->{loadOnly}) {
729          Trace("Loading from existing files.") if T(2);          Trace("Loading from existing files.") if T(2);
730      } else {      } else {
731          Trace("Generating subsystem data.") if T(2);          Trace("Generating subsystem data.") if T(2);
732          # 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
733          # information will be used to generate the Catalyzes table.          # information will be used to generate the Catalyzes table.
734          my %ecToRoles = ();          my %ecToRoles = ();
735          # 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 731  Line 743 
743              # Get the subsystem object.              # Get the subsystem object.
744              my $sub = $fig->get_subsystem($subsysID);              my $sub = $fig->get_subsystem($subsysID);
745              # Only proceed if the subsystem has a spreadsheet.              # Only proceed if the subsystem has a spreadsheet.
746              if (! $sub->{empty_ss}) {              if (defined($sub) && ! $sub->{empty_ss}) {
747                  Trace("Creating subsystem $subsysID.") if T(3);                  Trace("Creating subsystem $subsysID.") if T(3);
748                  $loadSubsystem->Add("subsystemIn");                  $loadSubsystem->Add("subsystemIn");
749                  # Create the subsystem record.                  # Create the subsystem record.
750                  my $curator = $sub->get_curator();                  my $curator = $sub->get_curator();
751                  my $notes = $sub->get_notes();                  my $notes = $sub->get_notes();
752                  $loadSubsystem->Put($subsysID, $curator, $notes);                  my $description = $sub->get_description();
753                  my $class = $fig->subsystem_classification($subsysID);                  $loadSubsystem->Put($subsysID, $curator, $description, $notes);
754                  if ($class) {                  # Now for the classification string. This comes back as a list
755                      $loadSubsystemClass->Put($subsysID, $class);                  # reference and we convert it to a space-delimited string.
756                  }                  my $classList = $fig->subsystem_classification($subsysID);
757                    my $classString = join($FIG_Config::splitter, grep { $_ } @$classList);
758                    $loadSubsystemClass->Put($subsysID, $classString);
759                  # 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.
760                  for (my $col = 0; defined($roleID = $sub->get_role($col)); $col++) {                  for (my $col = 0; defined($roleID = $sub->get_role($col)); $col++) {
761                        # Get the role's abbreviation.
762                        my $abbr = $sub->get_role_abbr($col);
763                      # Connect to this role.                      # Connect to this role.
764                      $loadOccursInSubsystem->Add("roleIn");                      $loadOccursInSubsystem->Add("roleIn");
765                      $loadOccursInSubsystem->Put($roleID, $subsysID, $col);                      $loadOccursInSubsystem->Put($roleID, $subsysID, $abbr, $col);
766                      # If it's a new role, add it to the role table.                      # If it's a new role, add it to the role table.
767                      if (! exists $roleData{$roleID}) {                      if (! exists $roleData{$roleID}) {
768                          # Get the role's abbreviation.                          # Get the role's abbreviation.
                         my $abbr = $sub->get_role_abbr($col);  
769                          # Add the role.                          # Add the role.
770                          $loadRole->Put($roleID, $abbr);                          $loadRole->Put($roleID);
771                          $roleData{$roleID} = 1;                          $roleData{$roleID} = 1;
772                          # Check for an EC number.                          # Check for an EC number.
773                          if ($roleID =~ /\(EC ([^.]+\.[^.]+\.[^.]+\.[^)]+)\)\s*$/) {                          if ($roleID =~ /\(EC (\d+\.\d+\.\d+\.\d+)\s*\)\s*$/) {
774                              my $ec = $1;                              my $ec = $1;
775                              $loadRoleEC->Put($roleID, $ec);                              $loadIsIdentifiedByEC->Put($roleID, $ec);
776                              $ecToRoles{$ec} = $roleID;                              # Check to see if this is our first encounter with this EC.
777                                if (exists $ecToRoles{$ec}) {
778                                    # No, so just add this role to the EC list.
779                                    push @{$ecToRoles{$ec}}, $roleID;
780                                } else {
781                                    # Output this EC.
782                                    $loadRoleEC->Put($ec);
783                                    # Create its role list.
784                                    $ecToRoles{$ec} = [$roleID];
785                                }
786                          }                          }
787                      }                      }
788                  }                  }
# Line 871  Line 895 
895              # Now we need to link all the map's roles to it.              # Now we need to link all the map's roles to it.
896              # A hash is used to prevent duplicates.              # A hash is used to prevent duplicates.
897              my %roleHash = ();              my %roleHash = ();
898              for my $role ($fig->map_to_ecs($map)) {              for my $ec ($fig->map_to_ecs($map)) {
899                  if (exists $ecToRoles{$role} && ! $roleHash{$role}) {                  if (exists $ecToRoles{$ec}) {
900                      $loadRoleOccursIn->Put($ecToRoles{$role}, $map);                      for my $role (@{$ecToRoles{$ec}}) {
901                            if (! $roleHash{$role}) {
902                                $loadRoleOccursIn->Put($role, $map);
903                      $roleHash{$role} = 1;                      $roleHash{$role} = 1;
904                  }                  }
905              }              }
906          }          }
907                }
908            }
909          # 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,
910          # then use the "ecToRoles" table to convert EC numbers to role IDs.          # then use the "ecToRoles" table to convert EC numbers to role IDs.
911          my @reactions = $fig->all_reactions();          my @reactions = $fig->all_reactions();
912          for my $reactionID (@reactions) {          for my $reactionID (@reactions) {
913              # 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.
914              my @roles = $fig->catalyzed_by($reactionID);              my @ecs = $fig->catalyzed_by($reactionID);
915              # Loop through the roles, creating catalyzation records.              # Loop through the roles, creating catalyzation records.
916              for my $thisRole (@roles) {              for my $thisEC (@ecs) {
917                  if (exists $ecToRoles{$thisRole}) {                  if (exists $ecToRoles{$thisEC}) {
918                      $loadCatalyzes->Put($ecToRoles{$thisRole}, $reactionID);                      for my $thisRole (@{$ecToRoles{$thisEC}}) {
919                            $loadCatalyzes->Put($thisRole, $reactionID);
920                        }
921                  }                  }
922              }              }
923          }          }
# Line 899  Line 929 
929    
930  =head3 LoadPropertyData  =head3 LoadPropertyData
931    
932  C<< my $stats = $spl->LoadPropertyData(); >>      my $stats = $spl->LoadPropertyData();
933    
934  Load the attribute data from FIG into Sprout.  Load the attribute data from FIG into Sprout.
935    
# Line 935  Line 965 
965      my $genomeHash = $self->{genomes};      my $genomeHash = $self->{genomes};
966      # Create load objects for each of the tables we're loading.      # Create load objects for each of the tables we're loading.
967      my $loadProperty = $self->_TableLoader('Property');      my $loadProperty = $self->_TableLoader('Property');
968      my $loadHasProperty = $self->_TableLoader('HasProperty', $self->PrimaryOnly);      my $loadHasProperty = $self->_TableLoader('HasProperty');
969      if ($self->{options}->{loadOnly}) {      if ($self->{options}->{loadOnly}) {
970          Trace("Loading from existing files.") if T(2);          Trace("Loading from existing files.") if T(2);
971      } else {      } else {
# Line 943  Line 973 
973          # Create a hash for storing property IDs.          # Create a hash for storing property IDs.
974          my %propertyKeys = ();          my %propertyKeys = ();
975          my $nextID = 1;          my $nextID = 1;
976            # Get the attributes we intend to store in the property table.
977            my $propKeys = $self->{propKeys};
978          # Loop through the genomes.          # Loop through the genomes.
979          for my $genomeID (keys %{$genomeHash}) {          for my $genomeID (sort keys %{$genomeHash}) {
980              $loadProperty->Add("genomeIn");              $loadProperty->Add("genomeIn");
981              Trace("Generating properties for $genomeID.") if T(3);              Trace("Generating properties for $genomeID.") if T(3);
982              # 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;  
983              my $propertyCount = 0;              my $propertyCount = 0;
984              # Loop through the features, creating HasProperty records.              # Get the properties for this genome's features.
985              for my $fid (@features) {              my @attributes = $fig->get_attributes("fig|$genomeID%", $propKeys);
986                  # Get all attributes for this feature. We do this one feature at a time              Trace("Property list built for $genomeID.") if T(3);
987                  # to insure we do not get any genome attributes.              # Loop through the results, creating HasProperty records.
988                  my @attributeList = $fig->get_attributes($fid, '', '', '');              for my $attributeData (@attributes) {
989                  if (scalar @attributeList) {                  # Pull apart the attribute tuple.
990                      $featureCount++;                  my ($fid, $key, $value, $url) = @{$attributeData};
                 }  
                 # 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};  
991                      # Concatenate the key and value and check the "propertyKeys" hash to                      # Concatenate the key and value and check the "propertyKeys" hash to
992                      # 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
993                      # character.                      # character.
# Line 984  Line 1005 
1005                      # Create the HasProperty entry for this feature/property association.                      # Create the HasProperty entry for this feature/property association.
1006                      $loadHasProperty->Put($fid, $propertyID, $url);                      $loadHasProperty->Put($fid, $propertyID, $url);
1007                  }                  }
             }  
1008              # Update the statistics.              # Update the statistics.
1009              Trace("$propertyCount attributes processed for $featureCount features.") if T(3);              Trace("$propertyCount attributes processed.") if T(3);
             $loadHasProperty->Add("featuresIn", $featureCount);  
1010              $loadHasProperty->Add("propertiesIn", $propertyCount);              $loadHasProperty->Add("propertiesIn", $propertyCount);
1011          }          }
1012      }      }
# Line 998  Line 1017 
1017    
1018  =head3 LoadAnnotationData  =head3 LoadAnnotationData
1019    
1020  C<< my $stats = $spl->LoadAnnotationData(); >>      my $stats = $spl->LoadAnnotationData();
1021    
1022  Load the annotation data from FIG into Sprout.  Load the annotation data from FIG into Sprout.
1023    
# Line 1032  Line 1051 
1051      my $genomeHash = $self->{genomes};      my $genomeHash = $self->{genomes};
1052      # Create load objects for each of the tables we're loading.      # Create load objects for each of the tables we're loading.
1053      my $loadAnnotation = $self->_TableLoader('Annotation');      my $loadAnnotation = $self->_TableLoader('Annotation');
1054      my $loadIsTargetOfAnnotation = $self->_TableLoader('IsTargetOfAnnotation', $self->PrimaryOnly);      my $loadIsTargetOfAnnotation = $self->_TableLoader('IsTargetOfAnnotation');
1055      my $loadSproutUser = $self->_TableLoader('SproutUser', $self->PrimaryOnly);      my $loadSproutUser = $self->_TableLoader('SproutUser');
1056      my $loadUserAccess = $self->_TableLoader('UserAccess', $self->PrimaryOnly);      my $loadUserAccess = $self->_TableLoader('UserAccess');
1057      my $loadMadeAnnotation = $self->_TableLoader('MadeAnnotation', $self->PrimaryOnly);      my $loadMadeAnnotation = $self->_TableLoader('MadeAnnotation');
1058      if ($self->{options}->{loadOnly}) {      if ($self->{options}->{loadOnly}) {
1059          Trace("Loading from existing files.") if T(2);          Trace("Loading from existing files.") if T(2);
1060      } else {      } else {
# Line 1105  Line 1124 
1124    
1125  =head3 LoadSourceData  =head3 LoadSourceData
1126    
1127  C<< my $stats = $spl->LoadSourceData(); >>      my $stats = $spl->LoadSourceData();
1128    
1129  Load the source data from FIG into Sprout.  Load the source data from FIG into Sprout.
1130    
# Line 1139  Line 1158 
1158      # Get the genome hash.      # Get the genome hash.
1159      my $genomeHash = $self->{genomes};      my $genomeHash = $self->{genomes};
1160      # Create load objects for each of the tables we're loading.      # Create load objects for each of the tables we're loading.
1161      my $loadComesFrom = $self->_TableLoader('ComesFrom', $self->PrimaryOnly);      my $loadComesFrom = $self->_TableLoader('ComesFrom');
1162      my $loadSource = $self->_TableLoader('Source');      my $loadSource = $self->_TableLoader('Source');
1163      my $loadSourceURL = $self->_TableLoader('SourceURL');      my $loadSourceURL = $self->_TableLoader('SourceURL');
1164      if ($self->{options}->{loadOnly}) {      if ($self->{options}->{loadOnly}) {
# Line 1183  Line 1202 
1202    
1203  =head3 LoadExternalData  =head3 LoadExternalData
1204    
1205  C<< my $stats = $spl->LoadExternalData(); >>      my $stats = $spl->LoadExternalData();
1206    
1207  Load the external data from FIG into Sprout.  Load the external data from FIG into Sprout.
1208    
# Line 1263  Line 1282 
1282    
1283  =head3 LoadReactionData  =head3 LoadReactionData
1284    
1285  C<< my $stats = $spl->LoadReactionData(); >>      my $stats = $spl->LoadReactionData();
1286    
1287  Load the reaction data from FIG into Sprout.  Load the reaction data from FIG into Sprout.
1288    
# Line 1276  Line 1295 
1295      Compound      Compound
1296      CompoundName      CompoundName
1297      CompoundCAS      CompoundCAS
1298        IsIdentifiedByCAS
1299        HasCompoundName
1300      IsAComponentOf      IsAComponentOf
1301    
1302  This method proceeds reaction by reaction rather than genome by genome.  This method proceeds reaction by reaction rather than genome by genome.
# Line 1297  Line 1318 
1318      my $fig = $self->{fig};      my $fig = $self->{fig};
1319      # Create load objects for each of the tables we're loading.      # Create load objects for each of the tables we're loading.
1320      my $loadReaction = $self->_TableLoader('Reaction');      my $loadReaction = $self->_TableLoader('Reaction');
1321      my $loadReactionURL = $self->_TableLoader('ReactionURL', $self->PrimaryOnly);      my $loadReactionURL = $self->_TableLoader('ReactionURL');
1322      my $loadCompound = $self->_TableLoader('Compound', $self->PrimaryOnly);      my $loadCompound = $self->_TableLoader('Compound');
1323      my $loadCompoundName = $self->_TableLoader('CompoundName', $self->PrimaryOnly);      my $loadCompoundName = $self->_TableLoader('CompoundName');
1324      my $loadCompoundCAS = $self->_TableLoader('CompoundCAS', $self->PrimaryOnly);      my $loadCompoundCAS = $self->_TableLoader('CompoundCAS');
1325      my $loadIsAComponentOf = $self->_TableLoader('IsAComponentOf', $self->PrimaryOnly);      my $loadIsAComponentOf = $self->_TableLoader('IsAComponentOf');
1326        my $loadIsIdentifiedByCAS = $self->_TableLoader('IsIdentifiedByCAS');
1327        my $loadHasCompoundName = $self->_TableLoader('HasCompoundName');
1328      if ($self->{options}->{loadOnly}) {      if ($self->{options}->{loadOnly}) {
1329          Trace("Loading from existing files.") if T(2);          Trace("Loading from existing files.") if T(2);
1330      } else {      } else {
1331          Trace("Generating annotation data.") if T(2);          Trace("Generating reaction data.") if T(2);
1332            # We need some hashes to prevent duplicates.
1333            my %compoundNames = ();
1334            my %compoundCASes = ();
1335          # First we create the compounds.          # First we create the compounds.
1336          my @compounds = $fig->all_compounds();          my @compounds = $fig->all_compounds();
1337          for my $cid (@compounds) {          for my $cid (@compounds) {
# Line 1314  Line 1340 
1340              # Each name will be given a priority number, starting with 1.              # Each name will be given a priority number, starting with 1.
1341              my $prio = 1;              my $prio = 1;
1342              for my $name (@names) {              for my $name (@names) {
1343                  $loadCompoundName->Put($cid, $name, $prio++);                  if (! exists $compoundNames{$name}) {
1344                        $loadCompoundName->Put($name);
1345                        $compoundNames{$name} = 1;
1346                    }
1347                    $loadHasCompoundName->Put($cid, $name, $prio++);
1348              }              }
1349              # Create the main compound record. Note that the first name              # Create the main compound record. Note that the first name
1350              # becomes the label.              # becomes the label.
# Line 1323  Line 1353 
1353              # Check for a CAS ID.              # Check for a CAS ID.
1354              my $cas = $fig->cas($cid);              my $cas = $fig->cas($cid);
1355              if ($cas) {              if ($cas) {
1356                  $loadCompoundCAS->Put($cid, $cas);                  $loadIsIdentifiedByCAS->Put($cid, $cas);
1357                    if (! exists $compoundCASes{$cas}) {
1358                        $loadCompoundCAS->Put($cas);
1359                        $compoundCASes{$cas} = 1;
1360                    }
1361              }              }
1362          }          }
1363          # 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 1360  Line 1394 
1394      return $retVal;      return $retVal;
1395  }  }
1396    
 =head3 LoadGroupData  
   
 C<< my $stats = $spl->LoadGroupData(); >>  
   
 Load the genome Groups into Sprout.  
   
 The following relations are loaded by this method.  
   
     GenomeGroups  
   
 There 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);  
         # Loop through the genomes.  
         my $line;  
         for my $genomeID (keys %{$genomeHash}) {  
             Trace("Processing $genomeID.") if T(3);  
             # Open the NMPDR group file for this genome.  
             if (open(TMP, "<$FIG_Config::organisms/$genomeID/NMPDR") &&  
                 defined($line = <TMP>)) {  
                 # Clean the line ending.  
                 chomp $line;  
                 # Add the group to the table. Note that there can only be one group  
                 # per genome.  
                 $loadGenomeGroups->Put($genomeID, $line);  
             }  
             close TMP;  
         }  
     }  
     # Finish the load.  
     my $retVal = $self->_FinishAll();  
     return $retVal;  
 }  
   
1397  =head3 LoadSynonymData  =head3 LoadSynonymData
1398    
1399  C<< my $stats = $spl->LoadSynonymData(); >>      my $stats = $spl->LoadSynonymData();
1400    
1401  Load the synonym groups into Sprout.  Load the synonym groups into Sprout.
1402    
# Line 1458  Line 1435 
1435          Trace("Generating synonym group data.") if T(2);          Trace("Generating synonym group data.") if T(2);
1436          # Get the database handle.          # Get the database handle.
1437          my $dbh = $fig->db_handle();          my $dbh = $fig->db_handle();
1438          # 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.
1439          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");
1440          my $result = $sth->execute();          my $result = $sth->execute();
1441          if (! defined($result)) {          if (! defined($result)) {
1442              Confess("Database error in Synonym load: " . $sth->errstr());              Confess("Database error in Synonym load: " . $sth->errstr());
1443          } else {          } else {
1444                Trace("Processing synonym results.") if T(2);
1445              # Remember the current synonym.              # Remember the current synonym.
1446              my $current_syn = "";              my $current_syn = "";
1447              # Count the features.              # Count the features.
1448              my $featureCount = 0;              my $featureCount = 0;
1449                my $entryCount = 0;
1450              # Loop through the synonym/peg pairs.              # Loop through the synonym/peg pairs.
1451              while (my @row = $sth->fetchrow()) {              while (my @row = $sth->fetchrow()) {
1452                  # Get the synonym ID and feature ID.                  # Get the synonym group ID and feature ID.
1453                  my ($syn_id, $peg) = @row;                  my ($syn_id, $peg) = @row;
1454                    # Count this row.
1455                    $entryCount++;
1456                    if ($entryCount % 1000 == 0) {
1457                        Trace("$entryCount rows processed.") if T(3);
1458                    }
1459                  # Insure it's for one of our genomes.                  # Insure it's for one of our genomes.
1460                  my $genomeID = FIG::genome_of($peg);                  my $genomeID = FIG::genome_of($peg);
1461                  if (exists $genomeHash->{$genomeID}) {                  if (exists $genomeHash->{$genomeID}) {
# Line 1490  Line 1474 
1474                      }                      }
1475                  }                  }
1476              }              }
1477                Trace("$entryCount rows produced $featureCount features.") if T(2);
1478          }          }
1479      }      }
1480      # Finish the load.      # Finish the load.
# Line 1499  Line 1484 
1484    
1485  =head3 LoadFamilyData  =head3 LoadFamilyData
1486    
1487  C<< my $stats = $spl->LoadFamilyData(); >>      my $stats = $spl->LoadFamilyData();
1488    
1489  Load the protein families into Sprout.  Load the protein families into Sprout.
1490    
1491  The following relations are loaded by this method.  The following relations are loaded by this method.
1492    
1493      Family      Family
1494      ContainsFeature      IsFamilyForFeature
1495    
1496  The source information for these relations is taken from the C<families_for_protein>,  The source information for these relations is taken from the C<families_for_protein>,
1497  C<family_function>, and C<sz_family> methods of the B<FIG> object.  C<family_function>, and C<sz_family> methods of the B<FIG> object.
# Line 1530  Line 1515 
1515      my $genomeHash = $self->{genomes};      my $genomeHash = $self->{genomes};
1516      # Create load objects for the tables we're loading.      # Create load objects for the tables we're loading.
1517      my $loadFamily = $self->_TableLoader('Family');      my $loadFamily = $self->_TableLoader('Family');
1518      my $loadContainsFeature = $self->_TableLoader('ContainsFeature');      my $loadIsFamilyForFeature = $self->_TableLoader('IsFamilyForFeature');
1519      if ($self->{options}->{loadOnly}) {      if ($self->{options}->{loadOnly}) {
1520          Trace("Loading from existing files.") if T(2);          Trace("Loading from existing files.") if T(2);
1521      } else {      } else {
# Line 1542  Line 1527 
1527              Trace("Processing features for $genomeID.") if T(2);              Trace("Processing features for $genomeID.") if T(2);
1528              # Loop through this genome's PEGs.              # Loop through this genome's PEGs.
1529              for my $fid ($fig->all_features($genomeID, "peg")) {              for my $fid ($fig->all_features($genomeID, "peg")) {
1530                  $loadContainsFeature->Add("features", 1);                  $loadIsFamilyForFeature->Add("features", 1);
1531                  # Get this feature's families.                  # Get this feature's families.
1532                  my @families = $fig->families_for_protein($fid);                  my @families = $fig->families_for_protein($fid);
1533                  # Loop through the families, connecting them to the feature.                  # Loop through the families, connecting them to the feature.
1534                  for my $family (@families) {                  for my $family (@families) {
1535                      $loadContainsFeature->Put($family, $fid);                      $loadIsFamilyForFeature->Put($family, $fid);
1536                      # If this is a new family, create a record for it.                      # If this is a new family, create a record for it.
1537                      if (! exists $familyHash{$family}) {                      if (! exists $familyHash{$family}) {
1538                          $familyHash{$family} = 1;                          $familyHash{$family} = 1;
# Line 1565  Line 1550 
1550      return $retVal;      return $retVal;
1551  }  }
1552    
1553    =head3 LoadDrugData
1554    
1555        my $stats = $spl->LoadDrugData();
1556    
1557    Load the drug target data into Sprout.
1558    
1559    The following relations are loaded by this method.
1560    
1561        PDB
1562        DocksWith
1563        IsProteinForFeature
1564        Ligand
1565    
1566    The source information for these relations is taken from attributes. The
1567    C<PDB> attribute links a PDB to a feature, and is used to build B<IsProteinForFeature>.
1568    The C<zinc_name> attribute describes the ligands. The C<docking_results>
1569    attribute contains the information for the B<DocksWith> relationship. It is
1570    expected that additional attributes and tables will be added in the future.
1571    
1572    =over 4
1573    
1574    =item RETURNS
1575    
1576    Returns a statistics object for the loads.
1577    
1578    =back
1579    
1580    =cut
1581    #: Return Type $%;
1582    sub LoadDrugData {
1583        # Get this object instance.
1584        my ($self) = @_;
1585        # Get the FIG object.
1586        my $fig = $self->{fig};
1587        # Get the genome hash.
1588        my $genomeHash = $self->{genomes};
1589        # Create load objects for the tables we're loading.
1590        my $loadPDB = $self->_TableLoader('PDB');
1591        my $loadLigand = $self->_TableLoader('Ligand');
1592        my $loadIsProteinForFeature = $self->_TableLoader('IsProteinForFeature');
1593        my $loadDocksWith = $self->_TableLoader('DocksWith');
1594        if ($self->{options}->{loadOnly}) {
1595            Trace("Loading from existing files.") if T(2);
1596        } else {
1597            Trace("Generating drug target data.") if T(2);
1598            # First comes the "DocksWith" relationship. This will give us a list of PDBs.
1599            # We can also encounter PDBs when we process "IsProteinForFeature". To manage
1600            # this process, PDB information is collected in a hash table and then
1601            # unspooled after both relationships are created.
1602            my %pdbHash = ();
1603            Trace("Generating docking data.") if T(2);
1604            # Get all the docking data. This may cause problems if there are too many PDBs,
1605            # at which point we'll need another algorithm. The indicator that this is
1606            # happening will be a timeout error in the next statement.
1607            my @dockData = $fig->query_attributes('$key = ? AND $value < ?',
1608                                                  ['docking_results', $FIG_Config::dockLimit]);
1609            Trace(scalar(@dockData) . " rows of docking data found.") if T(3);
1610            for my $dockData (@dockData) {
1611                # Get the docking data components.
1612                my ($pdbID, $docking_key, @valueData) = @{$dockData};
1613                # Fix the PDB ID. It's supposed to be lower-case, but this does not always happen.
1614                $pdbID = lc $pdbID;
1615                # Strip off the object type.
1616                $pdbID =~ s/pdb://;
1617                # Extract the ZINC ID from the docking key. Note that there are two possible
1618                # formats.
1619                my (undef, $zinc_id) = $docking_key =~ /^docking_results::(ZINC)?(\d+)$/;
1620                if (! $zinc_id) {
1621                    Trace("Invalid docking result key $docking_key for $pdbID.") if T(0);
1622                    $loadDocksWith->Add("errors");
1623                } else {
1624                    # Get the pieces of the value and parse the energy.
1625                    # Note that we don't care about the rank, since
1626                    # we can sort on the energy level itself in our database.
1627                    my ($energy, $tool, $type) = @valueData;
1628                    my ($rank, $total, $vanderwaals, $electrostatic) = split /\s*;\s*/, $energy;
1629                    # Ignore predicted results.
1630                    if ($type ne "Predicted") {
1631                        # Count this docking result.
1632                        if (! exists $pdbHash{$pdbID}) {
1633                            $pdbHash{$pdbID} = 1;
1634                        } else {
1635                            $pdbHash{$pdbID}++;
1636                        }
1637                        # Write the result to the output.
1638                        $loadDocksWith->Put($pdbID, $zinc_id, $electrostatic, $type, $tool,
1639                                            $total, $vanderwaals);
1640                    }
1641                }
1642            }
1643            Trace("Connecting features.") if T(2);
1644            # Loop through the genomes.
1645            for my $genome (sort keys %{$genomeHash}) {
1646                Trace("Generating PDBs for $genome.") if T(3);
1647                # Get all of the PDBs that BLAST against this genome's features.
1648                my @attributeData = $fig->get_attributes("fig|$genome%", 'PDB::%');
1649                for my $pdbData (@attributeData) {
1650                    # The PDB ID is coded as a subkey.
1651                    if ($pdbData->[1] !~ /PDB::(.+)/i) {
1652                        Trace("Invalid PDB ID \"$pdbData->[1]\" in attribute table.") if T(0);
1653                        $loadPDB->Add("errors");
1654                    } else {
1655                        my $pdbID = $1;
1656                        # Insure the PDB is in the hash.
1657                        if (! exists $pdbHash{$pdbID}) {
1658                            $pdbHash{$pdbID} = 0;
1659                        }
1660                        # The score and locations are coded in the attribute value.
1661                        if ($pdbData->[2] !~ /^([^;]+)(.*)$/) {
1662                            Trace("Invalid PDB data for $pdbID and feature $pdbData->[0].") if T(0);
1663                            $loadIsProteinForFeature->Add("errors");
1664                        } else {
1665                            my ($score, $locData) = ($1,$2);
1666                            # The location data may not be present, so we have to start with some
1667                            # defaults and then check.
1668                            my ($start, $end) = (1, 0);
1669                            if ($locData) {
1670                                $locData =~ /(\d+)-(\d+)/;
1671                                $start = $1;
1672                                $end = $2;
1673                            }
1674                            # If we still don't have the end location, compute it from
1675                            # the feature length.
1676                            if (! $end) {
1677                                # Most features have one location, but we do a list iteration
1678                                # just in case.
1679                                my @locations = $fig->feature_location($pdbData->[0]);
1680                                $end = 0;
1681                                for my $loc (@locations) {
1682                                    my $locObject = BasicLocation->new($loc);
1683                                    $end += $locObject->Length;
1684                                }
1685                            }
1686                            # Decode the score.
1687                            my $realScore = FIGRules::DecodeScore($score);
1688                            # Connect the PDB to the feature.
1689                            $loadIsProteinForFeature->Put($pdbID, $pdbData->[0], $start, $realScore, $end);
1690                        }
1691                    }
1692                }
1693            }
1694            # We've got all our PDBs now, so we unspool them from the hash.
1695            Trace("Generating PDBs. " . scalar(keys %pdbHash) . " found.") if T(2);
1696            my $count = 0;
1697            for my $pdbID (sort keys %pdbHash) {
1698                $loadPDB->Put($pdbID, $pdbHash{$pdbID});
1699                $count++;
1700                Trace("$count PDBs processed.") if T(3) && ($count % 500 == 0);
1701            }
1702            # Finally we create the ligand table. This information can be found in the
1703            # zinc_name attribute.
1704            Trace("Loading ligands.") if T(2);
1705            # The ligand list is huge, so we have to get it in pieces. We also have to check for duplicates.
1706            my $last_zinc_id = "";
1707            my $zinc_id = "";
1708            my $done = 0;
1709            while (! $done) {
1710                # Get the next 10000 ligands. We insist that the object ID is greater than
1711                # the last ID we processed.
1712                Trace("Loading batch starting with ZINC:$zinc_id.") if T(3);
1713                my @attributeData = $fig->query_attributes('$object > ? AND $key = ? ORDER BY $object LIMIT 10000',
1714                                                           ["ZINC:$zinc_id", "zinc_name"]);
1715                Trace(scalar(@attributeData) . " attribute rows returned.") if T(3);
1716                if (! @attributeData) {
1717                    # Here there are no attributes left, so we quit the loop.
1718                    $done = 1;
1719                } else {
1720                    # Process the attribute data we've received.
1721                    for my $zinc_data (@attributeData) {
1722                        # The ZINC ID is found in the first return column, prefixed with the word ZINC.
1723                        if ($zinc_data->[0] =~ /^ZINC:(\d+)$/) {
1724                            $zinc_id = $1;
1725                            # Check for a duplicate.
1726                            if ($zinc_id eq $last_zinc_id) {
1727                                $loadLigand->Add("duplicate");
1728                            } else {
1729                                # Here it's safe to output the ligand. The ligand name is the attribute value
1730                                # (third column in the row).
1731                                $loadLigand->Put($zinc_id, $zinc_data->[2]);
1732                                # Insure we don't try to add this ID again.
1733                                $last_zinc_id = $zinc_id;
1734                            }
1735                        } else {
1736                            Trace("Invalid zinc ID \"$zinc_data->[0]\" in attribute table.") if T(0);
1737                            $loadLigand->Add("errors");
1738                        }
1739                    }
1740                }
1741            }
1742            Trace("Ligands loaded.") if T(2);
1743        }
1744        # Finish the load.
1745        my $retVal = $self->_FinishAll();
1746        return $retVal;
1747    }
1748    
1749    
1750  =head2 Internal Utility Methods  =head2 Internal Utility Methods
1751    
1752    =head3 SpecialAttribute
1753    
1754        my $count = SproutLoad::SpecialAttribute($id, \@attributes, $idxMatch, \@idxValues, $pattern, $loader);
1755    
1756    Look for special attributes of a given type. A special attribute is found by comparing one of
1757    the columns of the incoming attribute list to a search pattern. If a match is found, then
1758    a set of columns is put into an output table connected to the specified ID.
1759    
1760    For example, when processing features, the attribute list we look at has three columns: attribute
1761    name, attribute value, and attribute value HTML. The IEDB attribute exists if the attribute name
1762    begins with C<iedb_>. The call signature is therefore
1763    
1764        my $found = SpecialAttribute($fid, \@attributeList, 0, [0,2], '^iedb_', $loadFeatureIEDB);
1765    
1766    The pattern is matched against column 0, and if we have a match, then column 2's value is put
1767    to the output along with the specified feature ID.
1768    
1769    =over 4
1770    
1771    =item id
1772    
1773    ID of the object whose special attributes are being loaded. This forms the first column of the
1774    output.
1775    
1776    =item attributes
1777    
1778    Reference to a list of tuples.
1779    
1780    =item idxMatch
1781    
1782    Index in each tuple of the column to be matched against the pattern. If the match is
1783    successful, an output record will be generated.
1784    
1785    =item idxValues
1786    
1787    Reference to a list containing the indexes in each tuple of the columns to be put as
1788    the second column of the output.
1789    
1790    =item pattern
1791    
1792    Pattern to be matched against the specified column. The match will be case-insensitive.
1793    
1794    =item loader
1795    
1796    An object to which each output record will be put. Usually this is an B<ERDBLoad> object,
1797    but technically it could be anything with a C<Put> method.
1798    
1799    =item RETURN
1800    
1801    Returns a count of the matches found.
1802    
1803    =item
1804    
1805    =back
1806    
1807    =cut
1808    
1809    sub SpecialAttribute {
1810        # Get the parameters.
1811        my ($id, $attributes, $idxMatch, $idxValues, $pattern, $loader) = @_;
1812        # Declare the return variable.
1813        my $retVal = 0;
1814        # Loop through the attribute rows.
1815        for my $row (@{$attributes}) {
1816            # Check for a match.
1817            if ($row->[$idxMatch] =~ m/$pattern/i) {
1818                # We have a match, so output a row. This is a bit tricky, since we may
1819                # be putting out multiple columns of data from the input.
1820                my $value = join(" ", map { $row->[$_] } @{$idxValues});
1821                $loader->Put($id, $value);
1822                $retVal++;
1823            }
1824        }
1825        Trace("$retVal special attributes found for $id and loader " . $loader->RelName() . ".") if T(4) && $retVal;
1826        # Return the number of matches.
1827        return $retVal;
1828    }
1829    
1830  =head3 TableLoader  =head3 TableLoader
1831    
1832  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 1581  Line 1841 
1841    
1842  Name of the table (relation) being loaded.  Name of the table (relation) being loaded.
1843    
 =item ignore  
   
 TRUE if the table should be ignored entirely, else FALSE.  
   
1844  =item RETURN  =item RETURN
1845    
1846  Returns an ERDBLoad object for loading the specified table.  Returns an ERDBLoad object for loading the specified table.
# Line 1595  Line 1851 
1851    
1852  sub _TableLoader {  sub _TableLoader {
1853      # Get the parameters.      # Get the parameters.
1854      my ($self, $tableName, $ignore) = @_;      my ($self, $tableName) = @_;
1855      # Create the load object.      # Create the load object.
1856      my $retVal = ERDBLoad->new($self->{erdb}, $tableName, $self->{loadDirectory}, $self->LoadOnly,      my $retVal = ERDBLoad->new($self->{erdb}, $tableName, $self->{loadDirectory}, $self->LoadOnly);
                                $ignore);  
1857      # Cache it in the loader list.      # Cache it in the loader list.
1858      push @{$self->{loaders}}, $retVal;      push @{$self->{loaders}}, $retVal;
1859      # Return it to the caller.      # Return it to the caller.
# Line 1670  Line 1925 
1925      return $retVal;      return $retVal;
1926  }  }
1927    
1928    =head3 GetGenomeAttributes
1929    
1930        my $aHashRef = GetGenomeAttributes($fig, $genomeID, \@fids, \@propKeys);
1931    
1932    Return a hash of attributes keyed on feature ID. This method gets all the NMPDR-related
1933    attributes for all the features of a genome in a single call, then organizes them into
1934    a hash.
1935    
1936    =over 4
1937    
1938    =item fig
1939    
1940    FIG-like object for accessing attributes.
1941    
1942    =item genomeID
1943    
1944    ID of the genome who's attributes are desired.
1945    
1946    =item fids
1947    
1948    Reference to a list of the feature IDs whose attributes are to be kept.
1949    
1950    =item propKeys
1951    
1952    A list of the keys to retrieve.
1953    
1954    =item RETURN
1955    
1956    Returns a reference to a hash. The key of the hash is the feature ID. The value is the
1957    reference to a list of the feature's attribute tuples. Each tuple contains the feature ID,
1958    the attribute key, and one or more attribute values.
1959    
1960    =back
1961    
1962    =cut
1963    
1964    sub GetGenomeAttributes {
1965        # Get the parameters.
1966        my ($fig, $genomeID, $fids, $propKeys) = @_;
1967        # Declare the return variable.
1968        my $retVal = {};
1969        # Initialize the hash. This not only enables us to easily determine which FIDs to
1970        # keep, it insures that the caller sees a list reference for every known fid,
1971        # simplifying the logic.
1972        for my $fid (@{$fids}) {
1973            $retVal->{$fid} = [];
1974        }
1975        # Get the attributes. If ev_code_cron is running, we may get a timeout error, so
1976        # an eval is used.
1977        my @aList = ();
1978        eval {
1979            @aList = $fig->get_attributes("fig|$genomeID%", $propKeys);
1980            Trace(scalar(@aList) . " attributes returned for genome $genomeID.") if T(3);
1981        };
1982        # Check for a problem.
1983        if ($@) {
1984            Trace("Retrying attributes for $genomeID due to error: $@") if T(1);
1985            # Our fallback plan is to process the attributes in blocks of 100. This is much slower,
1986            # but allows us to continue processing.
1987            my $nFids = scalar @{$fids};
1988            for (my $i = 0; $i < $nFids; $i += 100) {
1989                # Determine the index of the last feature ID we'll be specifying on this pass.
1990                # Normally it's $i + 99, but if we're close to the end it may be less.
1991                my $end = ($i + 100 > $nFids ? $nFids - 1 : $i + 99);
1992                # Get a slice of the fid list.
1993                my @slice = @{$fids}[$i .. $end];
1994                # Get the relevant attributes.
1995                Trace("Retrieving attributes for fids $i to $end.") if T(3);
1996                my @aShort = $fig->get_attributes(\@slice, $propKeys);
1997                Trace(scalar(@aShort) . " attributes returned for fids $i to $end.") if T(3);
1998                push @aList, @aShort;
1999            }
2000        }
2001        # Now we should have all the interesting attributes in @aList. Populate the hash with
2002        # them.
2003        for my $aListEntry (@aList) {
2004            my $fid = $aListEntry->[0];
2005            if (exists $retVal->{$fid}) {
2006                push @{$retVal->{$fid}}, $aListEntry;
2007            }
2008        }
2009        # Return the result.
2010        return $retVal;
2011    }
2012    
2013    
2014  1;  1;

Legend:
Removed from v.1.62  
changed lines
  Added in v.1.92

MCS Webmaster
ViewVC Help
Powered by ViewVC 1.0.3