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

Diff of /Sprout/SproutLoad.pm

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

revision 1.5, Fri Sep 9 14:55:01 2005 UTC revision 1.48, Fri Jul 7 00:24:16 2006 UTC
# Line 10  Line 10 
10      use Sprout;      use Sprout;
11      use Stats;      use Stats;
12      use BasicLocation;      use BasicLocation;
13        use HTML;
14    
15  =head1 Sprout Load Methods  =head1 Sprout Load Methods
16    
# Line 29  Line 30 
30      $stats->Accumulate($spl->LoadFeatureData());      $stats->Accumulate($spl->LoadFeatureData());
31      print $stats->Show();      print $stats->Show();
32    
 This module makes use of the internal Sprout property C<_erdb>.  
   
33  It is worth noting that the FIG object does not need to be a real one. Any object  It is worth noting that the FIG object does not need to be a real one. Any object
34  that implements the FIG methods for data retrieval could be used. So, for example,  that implements the FIG methods for data retrieval could be used. So, for example,
35  this object could be used to copy data from one Sprout database to another, or  this object could be used to copy data from one Sprout database to another, or
# Line 51  Line 50 
50    
51  =head3 new  =head3 new
52    
53  C<< my $spl = SproutLoad->new($sprout, $fig, $genomeFile, $subsysFile); >>  C<< my $spl = SproutLoad->new($sprout, $fig, $genomeFile, $subsysFile, $options); >>
54    
55  Construct a new Sprout Loader object, specifying the two participating databases and  Construct a new Sprout Loader object, specifying the two participating databases and
56  the name of the files containing the list of genomes and subsystems to use.  the name of the files containing the list of genomes and subsystems to use.
# Line 79  Line 78 
78  =item subsysFile  =item subsysFile
79    
80  Either the name of the file containing the list of trusted subsystems or a reference  Either the name of the file containing the list of trusted subsystems or a reference
81  to a list of subsystem names. If nothing is specified, all known subsystems will be  to a list of subsystem names. If nothing is specified, all NMPDR subsystems will be
82  considered trusted. Only subsystem data related to the trusted subsystems is loaded.  considered trusted. (A subsystem is considered NMPDR if it has a file named C<NMPDR>
83    in its data directory.) Only subsystem data related to the trusted subsystems is loaded.
84    
85    =item options
86    
87    Reference to a hash of command-line options.
88    
89  =back  =back
90    
# Line 88  Line 92 
92    
93  sub new {  sub new {
94      # Get the parameters.      # Get the parameters.
95      my ($class, $sprout, $fig, $genomeFile, $subsysFile) = @_;      my ($class, $sprout, $fig, $genomeFile, $subsysFile, $options) = @_;
96      # Load the list of genomes into a hash.      # Create the genome hash.
97      my %genomes;      my %genomes = ();
98        # We only need it if load-only is NOT specified.
99        if (! $options->{loadOnly}) {
100      if (! defined($genomeFile) || $genomeFile eq '') {      if (! defined($genomeFile) || $genomeFile eq '') {
101          # 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.
102          my @genomeList = $fig->genomes(1);          my @genomeList = $fig->genomes(1);
# Line 124  Line 130 
130              Confess("Invalid genome parameter ($type) in SproutLoad constructor.");              Confess("Invalid genome parameter ($type) in SproutLoad constructor.");
131          }          }
132      }      }
133        }
134      # Load the list of trusted subsystems.      # Load the list of trusted subsystems.
135      my %subsystems = ();      my %subsystems = ();
136        # We only need it if load-only is NOT specified.
137        if (! $options->{loadOnly}) {
138      if (! defined $subsysFile || $subsysFile eq '') {      if (! defined $subsysFile || $subsysFile eq '') {
139          # Here we want all the subsystems.              # Here we want all the NMPDR subsystems. First we get the whole list.
140          %subsystems = map { $_ => 1 } $fig->all_subsystems();              my @subs = $fig->all_subsystems();
141                # Loop through, checking for the NMPDR file.
142                for my $sub (@subs) {
143                    if (-e "$FIG_Config::data/Subsystems/$sub/NMPDR") {
144                        $subsystems{$sub} = 1;
145                    }
146                }
147      } else {      } else {
148          my $type = ref $subsysFile;          my $type = ref $subsysFile;
149          if ($type eq 'ARRAY') {          if ($type eq 'ARRAY') {
# Line 148  Line 163 
163              Confess("Invalid subsystem parameter in SproutLoad constructor.");              Confess("Invalid subsystem parameter in SproutLoad constructor.");
164          }          }
165      }      }
166        }
167      # Get the data directory from the Sprout object.      # Get the data directory from the Sprout object.
168      my ($directory) = $sprout->LoadInfo();      my ($directory) = $sprout->LoadInfo();
169      # Create the Sprout load object.      # Create the Sprout load object.
# Line 157  Line 173 
173                    subsystems => \%subsystems,                    subsystems => \%subsystems,
174                    sprout => $sprout,                    sprout => $sprout,
175                    loadDirectory => $directory,                    loadDirectory => $directory,
176                    erdb => $sprout->{_erdb},                    erdb => $sprout,
177                    loaders => []                    loaders => [],
178                      options => $options
179                   };                   };
180      # Bless and return it.      # Bless and return it.
181      bless $retVal, $class;      bless $retVal, $class;
182      return $retVal;      return $retVal;
183  }  }
184    
185    =head3 LoadOnly
186    
187    C<< my $flag = $spl->LoadOnly; >>
188    
189    Return TRUE if we are in load-only mode, else FALSE.
190    
191    =cut
192    
193    sub LoadOnly {
194        my ($self) = @_;
195        return $self->{options}->{loadOnly};
196    }
197    
198    =head3 PrimaryOnly
199    
200    C<< my $flag = $spl->PrimaryOnly; >>
201    
202    Return TRUE if only the main entity is to be loaded, else FALSE.
203    
204    =cut
205    
206    sub PrimaryOnly {
207        my ($self) = @_;
208        return $self->{options}->{primaryOnly};
209    }
210    
211  =head3 LoadGenomeData  =head3 LoadGenomeData
212    
213  C<< my $stats = $spl->LoadGenomeData(); >>  C<< my $stats = $spl->LoadGenomeData(); >>
# Line 192  Line 235 
235    
236  =back  =back
237    
 B<TO DO>  
   
 Real quality vectors instead of C<unknown> for everything.  
   
 GenomeGroup relation. (The original script took group information from the C<NMPDR> file  
 in each genome's main directory, but no such file exists anywhere in my version of the  
 data store.)  
   
238  =cut  =cut
239  #: Return Type $%;  #: Return Type $%;
240  sub LoadGenomeData {  sub LoadGenomeData {
# Line 210  Line 245 
245      # Get the genome count.      # Get the genome count.
246      my $genomeHash = $self->{genomes};      my $genomeHash = $self->{genomes};
247      my $genomeCount = (keys %{$genomeHash});      my $genomeCount = (keys %{$genomeHash});
     Trace("Beginning genome data load.") if T(2);  
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', $genomeCount);      my $loadGenome = $self->_TableLoader('Genome');
250      my $loadHasContig = $self->_TableLoader('HasContig', $genomeCount * 300);      my $loadHasContig = $self->_TableLoader('HasContig', $self->PrimaryOnly);
251      my $loadContig = $self->_TableLoader('Contig', $genomeCount * 300);      my $loadContig = $self->_TableLoader('Contig', $self->PrimaryOnly);
252      my $loadIsMadeUpOf = $self->_TableLoader('IsMadeUpOf', $genomeCount * 60000);      my $loadIsMadeUpOf = $self->_TableLoader('IsMadeUpOf', $self->PrimaryOnly);
253      my $loadSequence = $self->_TableLoader('Sequence', $genomeCount * 60000);      my $loadSequence = $self->_TableLoader('Sequence', $self->PrimaryOnly);
254        if ($self->{options}->{loadOnly}) {
255            Trace("Loading from existing files.") if T(2);
256        } else {
257            Trace("Generating genome data.") if T(2);
258      # Now we loop through the genomes, generating the data for each one.      # Now we loop through the genomes, generating the data for each one.
259      for my $genomeID (sort keys %{$genomeHash}) {      for my $genomeID (sort keys %{$genomeHash}) {
260          Trace("Loading data for genome $genomeID.") if T(3);              Trace("Generating data for genome $genomeID.") if T(3);
261                $loadGenome->Add("genomeIn");
262          # The access code comes in via the genome hash.          # The access code comes in via the genome hash.
263          my $accessCode = $genomeHash->{$genomeID};          my $accessCode = $genomeHash->{$genomeID};
264          # Get the genus, species, and strain from the scientific name. Note that we append              # Get the genus, species, and strain from the scientific name.
         # the genome ID to the strain. In some cases this is the totality of the strain name.  
265          my ($genus, $species, @extraData) = split / /, $self->{fig}->genus_species($genomeID);          my ($genus, $species, @extraData) = split / /, $self->{fig}->genus_species($genomeID);
266          my $extra = join " ", @extraData, "[$genomeID]";              my $extra = join " ", @extraData;
267          # Get the full taxonomy.          # Get the full taxonomy.
268          my $taxonomy = $fig->taxonomy_of($genomeID);          my $taxonomy = $fig->taxonomy_of($genomeID);
269          # Output the genome record.          # Output the genome record.
# Line 235  Line 273 
273          my @contigs = $fig->all_contigs($genomeID);          my @contigs = $fig->all_contigs($genomeID);
274          for my $contigID (@contigs) {          for my $contigID (@contigs) {
275              Trace("Processing contig $contigID for $genomeID.") if T(4);              Trace("Processing contig $contigID for $genomeID.") if T(4);
276                    $loadContig->Add("contigIn");
277                    $loadSequence->Add("contigIn");
278              # Create the contig ID.              # Create the contig ID.
279              my $sproutContigID = "$genomeID:$contigID";              my $sproutContigID = "$genomeID:$contigID";
280              # Create the contig record and relate it to the genome.              # Create the contig record and relate it to the genome.
# Line 246  Line 286 
286              # Now we get the sequence a chunk at a time.              # Now we get the sequence a chunk at a time.
287              my $contigLen = $fig->contig_ln($genomeID, $contigID);              my $contigLen = $fig->contig_ln($genomeID, $contigID);
288              for (my $i = 1; $i <= $contigLen; $i += $chunkSize) {              for (my $i = 1; $i <= $contigLen; $i += $chunkSize) {
289                        $loadSequence->Add("chunkIn");
290                  # Compute the endpoint of this chunk.                  # Compute the endpoint of this chunk.
291                  my $end = FIG::min($i + $chunkSize - 1, $contigLen);                  my $end = FIG::min($i + $chunkSize - 1, $contigLen);
292                  # Get the actual DNA.                  # Get the actual DNA.
# Line 258  Line 299 
299              }              }
300          }          }
301      }      }
302        }
303      # Finish the loads.      # Finish the loads.
304      my $retVal = $self->_FinishAll();      my $retVal = $self->_FinishAll();
305      # Return the result.      # Return the result.
# Line 301  Line 343 
343      my $genomeCount = (keys %{$genomeFilter});      my $genomeCount = (keys %{$genomeFilter});
344      my $featureCount = $genomeCount * 4000;      my $featureCount = $genomeCount * 4000;
345      # Start the loads.      # Start the loads.
346      my $loadCoupling = $self->_TableLoader('Coupling', $featureCount * $genomeCount);      my $loadCoupling = $self->_TableLoader('Coupling');
347      my $loadIsEvidencedBy = $self->_TableLoader('IsEvidencedBy', $featureCount * 8000);      my $loadIsEvidencedBy = $self->_TableLoader('IsEvidencedBy', $self->PrimaryOnly);
348      my $loadPCH = $self->_TableLoader('PCH', $featureCount * 2000);      my $loadPCH = $self->_TableLoader('PCH', $self->PrimaryOnly);
349      my $loadParticipatesInCoupling = $self->_TableLoader('ParticipatesInCoupling', $featureCount * 2000);      my $loadParticipatesInCoupling = $self->_TableLoader('ParticipatesInCoupling', $self->PrimaryOnly);
350      my $loadUsesAsEvidence = $self->_TableLoader('UsesAsEvidence', $featureCount * 8000);      my $loadUsesAsEvidence = $self->_TableLoader('UsesAsEvidence', $self->PrimaryOnly);
351      Trace("Beginning coupling data load.") if T(2);      if ($self->{options}->{loadOnly}) {
352            Trace("Loading from existing files.") if T(2);
353        } else {
354            Trace("Generating coupling data.") if T(2);
355      # Loop through the genomes found.      # Loop through the genomes found.
356      for my $genome (sort keys %{$genomeFilter}) {      for my $genome (sort keys %{$genomeFilter}) {
357          Trace("Generating coupling data for $genome.") if T(3);          Trace("Generating coupling data for $genome.") if T(3);
358                $loadCoupling->Add("genomeIn");
359          # Create a hash table for holding coupled pairs. We use this to prevent          # Create a hash table for holding coupled pairs. We use this to prevent
360          # duplicates. For example, if A is coupled to B, we don't want to also          # duplicates. For example, if A is coupled to B, we don't want to also
361          # assert that B is coupled to A, because we already know it. Fortunately,          # assert that B is coupled to A, because we already know it. Fortunately,
# Line 320  Line 366 
366          my @pegs = $fig->pegs_of($genome);          my @pegs = $fig->pegs_of($genome);
367          # Loop through the PEGs.          # Loop through the PEGs.
368          for my $peg1 (@pegs) {          for my $peg1 (@pegs) {
369                    $loadCoupling->Add("pegIn");
370              Trace("Processing PEG $peg1 for $genome.") if T(4);              Trace("Processing PEG $peg1 for $genome.") if T(4);
371              # Get a list of the coupled PEGs.              # Get a list of the coupled PEGs.
372              my @couplings = $fig->coupled_to($peg1);              my @couplings = $fig->coupled_to($peg1);
# Line 328  Line 375 
375              for my $coupleData (@couplings) {              for my $coupleData (@couplings) {
376                  my ($peg2, $score) = @{$coupleData};                  my ($peg2, $score) = @{$coupleData};
377                  # Compute the coupling ID.                  # Compute the coupling ID.
378                  my $coupleID = Sprout::CouplingID($peg1, $peg2);                      my $coupleID = $self->{erdb}->CouplingID($peg1, $peg2);
379                  if (! exists $dupHash{$coupleID}) {                  if (! exists $dupHash{$coupleID}) {
380                            $loadCoupling->Add("couplingIn");
381                      # Here we have a new coupling to store in the load files.                      # Here we have a new coupling to store in the load files.
382                      Trace("Storing coupling ($coupleID) with score $score.") if T(4);                      Trace("Storing coupling ($coupleID) with score $score.") if T(4);
383                      # Ensure we don't do this again.                      # Ensure we don't do this again.
# Line 345  Line 393 
393                      my %evidenceMap = ();                      my %evidenceMap = ();
394                      # Process each evidence item.                      # Process each evidence item.
395                      for my $evidenceData (@evidence) {                      for my $evidenceData (@evidence) {
396                                $loadPCH->Add("evidenceIn");
397                          my ($peg3, $peg4, $usage) = @{$evidenceData};                          my ($peg3, $peg4, $usage) = @{$evidenceData};
398                          # Only proceed if the evidence is from a Sprout                          # Only proceed if the evidence is from a Sprout
399                          # genome.                          # genome.
400                          if ($genomeFilter->{$fig->genome_of($peg3)}) {                          if ($genomeFilter->{$fig->genome_of($peg3)}) {
401                                    $loadUsesAsEvidence->Add("evidenceChosen");
402                              my $evidenceKey = "$coupleID $peg3 $peg4";                              my $evidenceKey = "$coupleID $peg3 $peg4";
403                              # We store this evidence in the hash if the usage                              # We store this evidence in the hash if the usage
404                              # is nonzero or no prior evidence has been found. This                              # is nonzero or no prior evidence has been found. This
405                              # insures that if there is duplicate evidence, we                              # insures that if there is duplicate evidence, we
406                              # at least keep the meaningful ones. Only evidence is                                  # at least keep the meaningful ones. Only evidence in
407                              # the hash makes it to the output.                              # the hash makes it to the output.
408                              if ($usage || ! exists $evidenceMap{$evidenceKey}) {                              if ($usage || ! exists $evidenceMap{$evidenceKey}) {
409                                  $evidenceMap{$evidenceKey} = $evidenceData;                                  $evidenceMap{$evidenceKey} = $evidenceData;
# Line 368  Line 418 
418                          $loadIsEvidencedBy->Put($coupleID, $evidenceID);                          $loadIsEvidencedBy->Put($coupleID, $evidenceID);
419                          # Connect it to the features.                          # Connect it to the features.
420                          $loadUsesAsEvidence->Put($evidenceID, $peg3, 1);                          $loadUsesAsEvidence->Put($evidenceID, $peg3, 1);
421                          $loadUsesAsEvidence->Put($evidenceID, $peg4, 1);                              $loadUsesAsEvidence->Put($evidenceID, $peg4, 2);
422                            }
423                      }                      }
424                  }                  }
425              }              }
# Line 395  Line 446 
446      FeatureTranslation      FeatureTranslation
447      FeatureUpstream      FeatureUpstream
448      IsLocatedIn      IsLocatedIn
449        HasFeature
450    
451  =over 4  =over 4
452    
# Line 413  Line 465 
465      my $fig = $self->{fig};      my $fig = $self->{fig};
466      # Get the table of genome IDs.      # Get the table of genome IDs.
467      my $genomeHash = $self->{genomes};      my $genomeHash = $self->{genomes};
     my $genomeCount = (keys %{$genomeHash});  
     my $featureCount = $genomeCount * 4000;  
468      # Create load objects for each of the tables we're loading.      # Create load objects for each of the tables we're loading.
469      my $loadFeature = $self->_TableLoader('Feature', $featureCount);      my $loadFeature = $self->_TableLoader('Feature');
470      my $loadFeatureAlias = $self->_TableLoader('FeatureAlias', $featureCount * 6);      my $loadIsLocatedIn = $self->_TableLoader('IsLocatedIn', $self->PrimaryOnly);
471      my $loadFeatureLink = $self->_TableLoader('FeatureLink', $featureCount * 10);      my $loadFeatureAlias = $self->_TableLoader('FeatureAlias');
472      my $loadFeatureTranslation = $self->_TableLoader('FeatureTranslation', $featureCount);      my $loadFeatureLink = $self->_TableLoader('FeatureLink');
473      my $loadFeatureUpstream = $self->_TableLoader('FeatureUpstream', $featureCount);      my $loadFeatureTranslation = $self->_TableLoader('FeatureTranslation');
474      my $loadIsLocatedIn = $self->_TableLoader('IsLocatedIn', $featureCount);      my $loadFeatureUpstream = $self->_TableLoader('FeatureUpstream');
475        my $loadHasFeature = $self->_TableLoader('HasFeature');
476      # 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
477      # locations.      # locations.
478      my $chunkSize = $self->{sprout}->MaxSegment();      my $chunkSize = $self->{sprout}->MaxSegment();
479      Trace("Beginning feature data load.") if T(2);      if ($self->{options}->{loadOnly}) {
480            Trace("Loading from existing files.") if T(2);
481        } else {
482            Trace("Generating feature data.") if T(2);
483      # Now we loop through the genomes, generating the data for each one.      # Now we loop through the genomes, generating the data for each one.
484      for my $genomeID (sort keys %{$genomeHash}) {      for my $genomeID (sort keys %{$genomeHash}) {
485          Trace("Loading features for genome $genomeID.") if T(3);          Trace("Loading features for genome $genomeID.") if T(3);
486                $loadFeature->Add("genomeIn");
487          # Get the feature list for this genome.          # Get the feature list for this genome.
488          my $features = $fig->all_features_detailed($genomeID);          my $features = $fig->all_features_detailed($genomeID);
489          # Loop through the features.          # Loop through the features.
490          for my $featureData (@{$features}) {          for my $featureData (@{$features}) {
491                    $loadFeature->Add("featureIn");
492              # Split the tuple.              # Split the tuple.
493              my ($featureID, $locations, $aliases, $type) = @{$featureData};                  my ($featureID, $locations, undef, $type) = @{$featureData};
494              # Create the feature record.              # Create the feature record.
495              $loadFeature->Put("$genomeID:$featureID", 1, $type);                  $loadFeature->Put($featureID, 1, $type);
496                    # Link it to the parent genome.
497                    $loadHasFeature->Put($genomeID, $featureID, $type);
498              # Create the aliases.              # Create the aliases.
499              for my $alias (split /\s*,\s*/, $aliases) {                  for my $alias ($fig->feature_aliases($featureID)) {
500                  $loadFeatureAlias->Put($featureID, $alias);                  $loadFeatureAlias->Put($featureID, $alias);
501              }              }
502              # Get the links.              # Get the links.
# Line 448  Line 506 
506              }              }
507              # If this is a peg, generate the translation and the upstream.              # If this is a peg, generate the translation and the upstream.
508              if ($type eq 'peg') {              if ($type eq 'peg') {
509                        $loadFeatureTranslation->Add("pegIn");
510                  my $translation = $fig->get_translation($featureID);                  my $translation = $fig->get_translation($featureID);
511                  if ($translation) {                  if ($translation) {
512                      $loadFeatureTranslation->Put($featureID, $translation);                      $loadFeatureTranslation->Put($featureID, $translation);
# Line 463  Line 522 
522              # the maximum segment size. This simplifies the genes_in_region processing              # the maximum segment size. This simplifies the genes_in_region processing
523              # for Sprout.              # for Sprout.
524              my @locationList = split /\s*,\s*/, $locations;              my @locationList = split /\s*,\s*/, $locations;
525                    # Create the location position indicator.
526                    my $i = 1;
527              # Loop through the locations.              # Loop through the locations.
528              for my $location (@locationList) {              for my $location (@locationList) {
529                  # Parse the location.                  # Parse the location.
530                  my $locObject = BasicLocation->new($location);                      my $locObject = BasicLocation->new("$genomeID:$location");
531                  # Split it into a list of chunks.                  # Split it into a list of chunks.
532                  my @locOList = ();                  my @locOList = ();
533                  while (my $peeling = $locObject->Peel($chunkSize)) {                  while (my $peeling = $locObject->Peel($chunkSize)) {
534                            $loadIsLocatedIn->Add("peeling");
535                      push @locOList, $peeling;                      push @locOList, $peeling;
536                  }                  }
537                  push @locOList, $locObject;                  push @locOList, $locObject;
538                  # Loop through the chunks, creating IsLocatedIn records. The variable                  # Loop through the chunks, creating IsLocatedIn records. The variable
539                  # "$i" will be used to keep the location index.                  # "$i" will be used to keep the location index.
                 my $i = 1;  
540                  for my $locChunk (@locOList) {                  for my $locChunk (@locOList) {
541                      $loadIsLocatedIn->Put($featureID, $locChunk->Contig, $locChunk->Left,                      $loadIsLocatedIn->Put($featureID, $locChunk->Contig, $locChunk->Left,
542                                            $locChunk->Dir, $locChunk->Length, $i);                                            $locChunk->Dir, $locChunk->Length, $i);
# Line 484  Line 545 
545              }              }
546          }          }
547      }      }
548        }
549      # Finish the loads.      # Finish the loads.
550      my $retVal = $self->_FinishAll();      my $retVal = $self->_FinishAll();
551      return $retVal;      return $retVal;
# Line 520  Line 582 
582      my $fig = $self->{fig};      my $fig = $self->{fig};
583      # Get the table of genome IDs.      # Get the table of genome IDs.
584      my $genomeHash = $self->{genomes};      my $genomeHash = $self->{genomes};
     my $genomeCount = (keys %{$genomeHash});  
     my $featureCount = $genomeCount * 4000;  
585      # Create load objects for each of the tables we're loading.      # Create load objects for each of the tables we're loading.
586      my $loadIsBidirectionalBestHitOf = $self->_TableLoader('IsBidirectionalBestHitOf',      my $loadIsBidirectionalBestHitOf = $self->_TableLoader('IsBidirectionalBestHitOf');
587                                                             $featureCount * $genomeCount);      if ($self->{options}->{loadOnly}) {
588      Trace("Beginning BBH load.") if T(2);          Trace("Loading from existing files.") if T(2);
589        } else {
590            Trace("Generating BBH data.") if T(2);
591      # Now we loop through the genomes, generating the data for each one.      # Now we loop through the genomes, generating the data for each one.
592      for my $genomeID (sort keys %{$genomeHash}) {      for my $genomeID (sort keys %{$genomeHash}) {
593                $loadIsBidirectionalBestHitOf->Add("genomeIn");
594          Trace("Processing features for genome $genomeID.") if T(3);          Trace("Processing features for genome $genomeID.") if T(3);
595          # Get the feature list for this genome.          # Get the feature list for this genome.
596          my $features = $fig->all_features_detailed($genomeID);          my $features = $fig->all_features_detailed($genomeID);
# Line 550  Line 613 
613              }              }
614          }          }
615      }      }
616        }
617      # Finish the loads.      # Finish the loads.
618      my $retVal = $self->_FinishAll();      my $retVal = $self->_FinishAll();
619      return $retVal;      return $retVal;
# Line 570  Line 634 
634  The following relations are loaded by this method.  The following relations are loaded by this method.
635    
636      Subsystem      Subsystem
637        SubsystemClass
638      Role      Role
639        RoleEC
640      SSCell      SSCell
641      ContainsFeature      ContainsFeature
642      IsGenomeOf      IsGenomeOf
# Line 578  Line 644 
644      OccursInSubsystem      OccursInSubsystem
645      ParticipatesIn      ParticipatesIn
646      HasSSCell      HasSSCell
647        ConsistsOfRoles
648        RoleSubset
649        HasRoleSubset
650        ConsistsOfGenomes
651        GenomeSubset
652        HasGenomeSubset
653        Catalyzes
654        Diagram
655        RoleOccursIn
656    
657  =over 4  =over 4
658    
# Line 587  Line 662 
662    
663  =back  =back
664    
 B<TO DO>  
   
 Generate RoleName table?  
   
665  =cut  =cut
666  #: Return Type $%;  #: Return Type $%;
667  sub LoadSubsystemData {  sub LoadSubsystemData {
# Line 604  Line 675 
675      # Get the subsystem hash. This lists the subsystems we'll process.      # Get the subsystem hash. This lists the subsystems we'll process.
676      my $subsysHash = $self->{subsystems};      my $subsysHash = $self->{subsystems};
677      my @subsysIDs = sort keys %{$subsysHash};      my @subsysIDs = sort keys %{$subsysHash};
678      my $subsysCount = @subsysIDs;      # Get the map list.
679      my $genomeCount = (keys %{$genomeHash});      my @maps = $fig->all_maps;
     my $featureCount = $genomeCount * 4000;  
680      # Create load objects for each of the tables we're loading.      # Create load objects for each of the tables we're loading.
681      my $loadSubsystem = $self->_TableLoader('Subsystem', $subsysCount);      my $loadDiagram = $self->_TableLoader('Diagram', $self->PrimaryOnly);
682      my $loadRole = $self->_TableLoader('Role', $featureCount * 6);      my $loadRoleOccursIn = $self->_TableLoader('RoleOccursIn', $self->PrimaryOnly);
683      my $loadSSCell = $self->_TableLoader('SSCell', $featureCount * $genomeCount);      my $loadSubsystem = $self->_TableLoader('Subsystem');
684      my $loadContainsFeature = $self->_TableLoader('ContainsFeature', $featureCount * $subsysCount);      my $loadRole = $self->_TableLoader('Role', $self->PrimaryOnly);
685      my $loadIsGenomeOf = $self->_TableLoader('IsGenomeOf', $featureCount * $genomeCount);      my $loadRoleEC = $self->_TableLoader('RoleEC', $self->PrimaryOnly);
686      my $loadIsRoleOf = $self->_TableLoader('IsRoleOf', $featureCount * $genomeCount);      my $loadCatalyzes = $self->_TableLoader('Catalyzes', $self->PrimaryOnly);
687      my $loadOccursInSubsystem = $self->_TableLoader('OccursInSubsystem', $featureCount * 6);      my $loadSSCell = $self->_TableLoader('SSCell', $self->PrimaryOnly);
688      my $loadParticipatesIn = $self->_TableLoader('ParticipatesIn', $subsysCount * $genomeCount);      my $loadContainsFeature = $self->_TableLoader('ContainsFeature', $self->PrimaryOnly);
689      my $loadHasSSCell = $self->_TableLoader('HasSSCell', $featureCount * $genomeCount);      my $loadIsGenomeOf = $self->_TableLoader('IsGenomeOf', $self->PrimaryOnly);
690      Trace("Beginning subsystem data load.") if T(2);      my $loadIsRoleOf = $self->_TableLoader('IsRoleOf', $self->PrimaryOnly);
691        my $loadOccursInSubsystem = $self->_TableLoader('OccursInSubsystem', $self->PrimaryOnly);
692        my $loadParticipatesIn = $self->_TableLoader('ParticipatesIn', $self->PrimaryOnly);
693        my $loadHasSSCell = $self->_TableLoader('HasSSCell', $self->PrimaryOnly);
694        my $loadRoleSubset = $self->_TableLoader('RoleSubset', $self->PrimaryOnly);
695        my $loadGenomeSubset = $self->_TableLoader('GenomeSubset', $self->PrimaryOnly);
696        my $loadConsistsOfRoles = $self->_TableLoader('ConsistsOfRoles', $self->PrimaryOnly);
697        my $loadConsistsOfGenomes = $self->_TableLoader('ConsistsOfGenomes', $self->PrimaryOnly);
698        my $loadHasRoleSubset = $self->_TableLoader('HasRoleSubset', $self->PrimaryOnly);
699        my $loadHasGenomeSubset = $self->_TableLoader('HasGenomeSubset', $self->PrimaryOnly);
700        my $loadSubsystemClass = $self->_TableLoader('SubsystemClass', $self->PrimaryOnly);
701        if ($self->{options}->{loadOnly}) {
702            Trace("Loading from existing files.") if T(2);
703        } else {
704            Trace("Generating subsystem data.") if T(2);
705            # This hash will contain the role for each EC. When we're done, this
706            # information will be used to generate the Catalyzes table.
707            my %ecToRoles = ();
708      # Loop through the subsystems. Our first task will be to create the      # Loop through the subsystems. Our first task will be to create the
709      # roles. We do this by looping through the subsystems and creating a      # roles. We do this by looping through the subsystems and creating a
710      # role hash. The hash tracks each role ID so that we don't create      # role hash. The hash tracks each role ID so that we don't create
711      # duplicates. As we move along, we'll connect the roles and subsystems.          # duplicates. As we move along, we'll connect the roles and subsystems
712            # and memorize up the reactions.
713            my ($genomeID, $roleID);
714      my %roleData = ();      my %roleData = ();
715      for my $subsysID (@subsysIDs) {      for my $subsysID (@subsysIDs) {
716                # Get the subsystem object.
717                my $sub = $fig->get_subsystem($subsysID);
718                # Only proceed if the subsystem has a spreadsheet.
719                if (! $sub->{empty_ss}) {
720          Trace("Creating subsystem $subsysID.") if T(3);          Trace("Creating subsystem $subsysID.") if T(3);
721                    $loadSubsystem->Add("subsystemIn");
722          # Create the subsystem record.          # Create the subsystem record.
723          $loadSubsystem->Put($subsysID);                  my $curator = $sub->get_curator();
724          # Get the subsystem's roles.                  my $notes = $sub->get_notes();
725          my @roles = $fig->subsys_to_roles($subsysID);                  $loadSubsystem->Put($subsysID, $curator, $notes);
726          # Connect the roles to the subsystem. If a role is new, we create                  my $class = $fig->subsystem_classification($subsysID);
727          # a role record for it.                  if ($class) {
728          for my $roleID (@roles) {                      $loadSubsystemClass->Put($subsysID, $class);
729              $loadOccursInSubsystem->Put($roleID, $subsysID);                  }
730                    # Connect it to its roles. Each role is a column in the subsystem spreadsheet.
731                    for (my $col = 0; defined($roleID = $sub->get_role($col)); $col++) {
732                        # Connect to this role.
733                        $loadOccursInSubsystem->Add("roleIn");
734                        $loadOccursInSubsystem->Put($roleID, $subsysID, $col);
735                        # If it's a new role, add it to the role table.
736              if (! exists $roleData{$roleID}) {              if (! exists $roleData{$roleID}) {
737                  $loadRole->Put($roleID);                          # Get the role's abbreviation.
738                            my $abbr = $sub->get_role_abbr($col);
739                            # Add the role.
740                            $loadRole->Put($roleID, $abbr);
741                  $roleData{$roleID} = 1;                  $roleData{$roleID} = 1;
742                            # Check for an EC number.
743                            if ($roleID =~ /\(EC ([^.]+\.[^.]+\.[^.]+\.[^)]+)\)\s*$/) {
744                                my $ec = $1;
745                                $loadRoleEC->Put($roleID, $ec);
746                                $ecToRoles{$ec} = $roleID;
747                            }
748              }              }
749          }          }
750          # Now all roles for this subsystem have been filled in. We create the                  # Now we create the spreadsheet for the subsystem by matching roles to
751          # spreadsheet by matches roles to genomes. To do this, we need to                  # genomes. Each genome is a row and each role is a column. We may need
752          # get the genomes on the sheet.                  # to actually create the roles as we find them.
753          Trace("Creating subsystem $subsysID spreadsheet.") if T(3);          Trace("Creating subsystem $subsysID spreadsheet.") if T(3);
754          my @genomes = map { $_->[0] } @{$fig->subsystem_genomes($subsysID)};                  for (my $row = 0; defined($genomeID = $sub->get_genome($row)); $row++) {
755          for my $genomeID (@genomes) {                      # Only proceed if this is one of our genomes.
             # Only process this genome if it's one of ours.  
756              if (exists $genomeHash->{$genomeID}) {              if (exists $genomeHash->{$genomeID}) {
757                  # Connect the genome to the subsystem.                          # Count the PEGs and cells found for verification purposes.
758                  $loadParticipatesIn->Put($genomeID, $subsysID);                          my $pegCount = 0;
759                            my $cellCount = 0;
760                            # Create a list for the PEGs we find. This list will be used
761                            # to generate cluster numbers.
762                            my @pegsFound = ();
763                            # Create a hash that maps spreadsheet IDs to PEGs. We will
764                            # use this to generate the ContainsFeature data after we have
765                            # the cluster numbers.
766                            my %cellPegs = ();
767                            # Get the genome's variant code for this subsystem.
768                            my $variantCode = $sub->get_variant_code($row);
769                  # Loop through the subsystem's roles. We use an index because it is                  # Loop through the subsystem's roles. We use an index because it is
770                  # part of the spreadsheet cell ID.                  # part of the spreadsheet cell ID.
771                  for (my $i = 0; $i <= $#roles; $i++) {                          for (my $col = 0; defined($roleID = $sub->get_role($col)); $col++) {
                     my $role = $roles[$i];  
772                      # Get the features in the spreadsheet cell for this genome and role.                      # Get the features in the spreadsheet cell for this genome and role.
773                      my @pegs = $fig->pegs_in_subsystem_coll($subsysID, $genomeID, $i);                              my @pegs = grep { !$fig->is_deleted_fid($_) } $sub->get_pegs_from_cell($row, $col);
774                      # Only proceed if features exist.                      # Only proceed if features exist.
775                      if (@pegs > 0) {                      if (@pegs > 0) {
776                          # Create the spreadsheet cell.                          # Create the spreadsheet cell.
777                          my $cellID = "$subsysID:$genomeID:$i";                                  $cellCount++;
778                                    my $cellID = "$subsysID:$genomeID:$col";
779                          $loadSSCell->Put($cellID);                          $loadSSCell->Put($cellID);
780                          $loadIsGenomeOf->Put($genomeID, $cellID);                          $loadIsGenomeOf->Put($genomeID, $cellID);
781                          $loadIsRoleOf->Put($role, $cellID);                                  $loadIsRoleOf->Put($roleID, $cellID);
782                          $loadHasSSCell->Put($subsysID, $cellID);                          $loadHasSSCell->Put($subsysID, $cellID);
783                          # Attach the features to it.                                  # Remember its features.
784                          for my $pegID (@pegs) {                                  push @pegsFound, @pegs;
785                              $loadContainsFeature->Put($cellID, $pegID);                                  $cellPegs{$cellID} = \@pegs;
786                                    $pegCount += @pegs;
787                                }
788                            }
789                            # If we found some cells for this genome, we need to compute clusters and
790                            # denote it participates in the subsystem.
791                            if ($pegCount > 0) {
792                                Trace("$pegCount PEGs in $cellCount cells for $genomeID.") if T(3);
793                                $loadParticipatesIn->Put($genomeID, $subsysID, $variantCode);
794                                # Create a hash mapping PEG IDs to cluster numbers.
795                                # We default to -1 for all of them.
796                                my %clusterOf = map { $_ => -1 } @pegsFound;
797                                # Partition the PEGs found into clusters.
798                                my @clusters = $fig->compute_clusters([keys %clusterOf], $sub);
799                                for (my $i = 0; $i <= $#clusters; $i++) {
800                                    my $subList = $clusters[$i];
801                                    for my $peg (@{$subList}) {
802                                        $clusterOf{$peg} = $i;
803                                    }
804                                }
805                                # Create the ContainsFeature data.
806                                for my $cellID (keys %cellPegs) {
807                                    my $cellList = $cellPegs{$cellID};
808                                    for my $cellPeg (@$cellList) {
809                                        $loadContainsFeature->Put($cellID, $cellPeg, $clusterOf{$cellPeg});
810                          }                          }
811                      }                      }
812                  }                  }
813              }              }
814          }          }
815                    # Now we need to generate the subsets. The subset names must be concatenated to
816                    # the subsystem name to make them unique keys. There are two types of subsets:
817                    # genome subsets and role subsets. We do the role subsets first.
818                    my @subsetNames = $sub->get_subset_names();
819                    for my $subsetID (@subsetNames) {
820                        # Create the subset record.
821                        my $actualID = "$subsysID:$subsetID";
822                        $loadRoleSubset->Put($actualID);
823                        # Connect the subset to the subsystem.
824                        $loadHasRoleSubset->Put($subsysID, $actualID);
825                        # Connect the subset to its roles.
826                        my @roles = $sub->get_subsetC_roles($subsetID);
827                        for my $roleID (@roles) {
828                            $loadConsistsOfRoles->Put($actualID, $roleID);
829      }      }
     # Finish the load.  
     my $retVal = $self->_FinishAll();  
     return $retVal;  
830  }  }
831                    # Next the genome subsets.
832  =head3 LoadDiagramData                  @subsetNames = $sub->get_subset_namesR();
833                    for my $subsetID (@subsetNames) {
834  C<< my $stats = $spl->LoadDiagramData(); >>                      # Create the subset record.
835                        my $actualID = "$subsysID:$subsetID";
836  Load the diagram data from FIG into Sprout.                      $loadGenomeSubset->Put($actualID);
837                        # Connect the subset to the subsystem.
838  Diagrams are used to organize functional roles. The diagram shows the                      $loadHasGenomeSubset->Put($subsysID, $actualID);
839  connections between chemicals that interact with a subsystem.                      # Connect the subset to its genomes.
840                        my @genomes = $sub->get_subsetR($subsetID);
841  The following relations are loaded by this method.                      for my $genomeID (@genomes) {
842                            $loadConsistsOfGenomes->Put($actualID, $genomeID);
843      Diagram                      }
844      RoleOccursIn                  }
845                }
846  =over 4              # Now we loop through the diagrams. We need to create the diagram records
847                # and link each diagram to its roles. Note that only roles which occur
848  =item RETURNS              # in subsystems (and therefore appear in the %ecToRoles hash) are
849                # included.
850  Returns a statistics object for the loads.              for my $map (@maps) {
   
 =back  
   
 =cut  
 #: Return Type $%;  
 sub LoadDiagramData {  
     # Get this object instance.  
     my ($self) = @_;  
     # Get the FIG object.  
     my $fig = $self->{fig};  
     # Get the map list.  
     my @maps = $fig->all_maps;  
     my $mapCount = @maps;  
     my $genomeCount = (keys %{$self->{genomes}});  
     my $featureCount = $genomeCount * 4000;  
     # Create load objects for each of the tables we're loading.  
     my $loadDiagram = $self->_TableLoader('Diagram', $mapCount);  
     my $loadRoleOccursIn = $self->_TableLoader('RoleOccursIn', $featureCount * 6);  
     Trace("Beginning diagram data load.") if T(2);  
     # Loop through the diagrams.  
     for my $map ($fig->all_maps) {  
851          Trace("Loading diagram $map.") if T(3);          Trace("Loading diagram $map.") if T(3);
852          # Get the diagram's descriptive name.          # Get the diagram's descriptive name.
853          my $name = $fig->map_name($map);          my $name = $fig->map_name($map);
# Line 724  Line 856 
856          # A hash is used to prevent duplicates.          # A hash is used to prevent duplicates.
857          my %roleHash = ();          my %roleHash = ();
858          for my $role ($fig->map_to_ecs($map)) {          for my $role ($fig->map_to_ecs($map)) {
859              if (! $roleHash{$role}) {                      if (exists $ecToRoles{$role} && ! $roleHash{$role}) {
860                  $loadRoleOccursIn->Put($role, $map);                          $loadRoleOccursIn->Put($ecToRoles{$role}, $map);
861                  $roleHash{$role} = 1;                  $roleHash{$role} = 1;
862              }              }
863          }          }
864      }      }
865                # Before we leave, we must create the Catalyzes table. We start with the reactions,
866                # then use the "ecToRoles" table to convert EC numbers to role IDs.
867                my @reactions = $fig->all_reactions();
868                for my $reactionID (@reactions) {
869                    # Get this reaction's list of roles. The results will be EC numbers.
870                    my @roles = $fig->catalyzed_by($reactionID);
871                    # Loop through the roles, creating catalyzation records.
872                    for my $thisRole (@roles) {
873                        if (exists $ecToRoles{$thisRole}) {
874                            $loadCatalyzes->Put($ecToRoles{$thisRole}, $reactionID);
875                        }
876                    }
877                }
878            }
879        }
880      # Finish the load.      # Finish the load.
881      my $retVal = $self->_FinishAll();      my $retVal = $self->_FinishAll();
882      return $retVal;      return $retVal;
# Line 771  Line 918 
918      my $fig = $self->{fig};      my $fig = $self->{fig};
919      # Get the genome hash.      # Get the genome hash.
920      my $genomeHash = $self->{genomes};      my $genomeHash = $self->{genomes};
     my $genomeCount = (keys %{$genomeHash});  
921      # Create load objects for each of the tables we're loading.      # Create load objects for each of the tables we're loading.
922      my $loadProperty = $self->_TableLoader('Property', $genomeCount * 1500);      my $loadProperty = $self->_TableLoader('Property');
923      my $loadHasProperty = $self->_TableLoader('HasProperty', $genomeCount * 1500);      my $loadHasProperty = $self->_TableLoader('HasProperty', $self->PrimaryOnly);
924      Trace("Beginning property data load.") if T(2);      if ($self->{options}->{loadOnly}) {
925            Trace("Loading from existing files.") if T(2);
926        } else {
927            Trace("Generating property data.") if T(2);
928      # Create a hash for storing property IDs.      # Create a hash for storing property IDs.
929      my %propertyKeys = ();      my %propertyKeys = ();
930      my $nextID = 1;      my $nextID = 1;
931      # Loop through the genomes.      # Loop through the genomes.
932      for my $genomeID (keys %{$genomeHash}) {      for my $genomeID (keys %{$genomeHash}) {
933                $loadProperty->Add("genomeIn");
934                Trace("Generating properties for $genomeID.") if T(3);
935          # Get the genome's features. The feature ID is the first field in the          # Get the genome's features. The feature ID is the first field in the
936          # tuples returned by "all_features_detailed". We use "all_features_detailed"          # tuples returned by "all_features_detailed". We use "all_features_detailed"
937          # rather than "all_features" because we want all features regardless of type.          # rather than "all_features" because we want all features regardless of type.
938          my @features = map { $_->[0] } @{$fig->all_features_detailed($genomeID)};          my @features = map { $_->[0] } @{$fig->all_features_detailed($genomeID)};
939                my $featureCount = 0;
940                my $propertyCount = 0;
941          # Loop through the features, creating HasProperty records.          # Loop through the features, creating HasProperty records.
942          for my $fid (@features) {          for my $fid (@features) {
943              # Get all attributes for this feature. We do this one feature at a time              # Get all attributes for this feature. We do this one feature at a time
944              # to insure we do not get any genome attributes.              # to insure we do not get any genome attributes.
945              my @attributeList = $fig->get_attributes($fid, '', '', '');              my @attributeList = $fig->get_attributes($fid, '', '', '');
946                    if (scalar @attributeList) {
947                        $featureCount++;
948                    }
949              # Loop through the attributes.              # Loop through the attributes.
950              for my $tuple (@attributeList) {              for my $tuple (@attributeList) {
951                        $propertyCount++;
952                  # Get this attribute value's data. Note that we throw away the FID,                  # Get this attribute value's data. Note that we throw away the FID,
953                  # since it will always be the same as the value if "$fid".                  # since it will always be the same as the value if "$fid".
954                  my (undef, $key, $value, $url) = @{$tuple};                  my (undef, $key, $value, $url) = @{$tuple};
# Line 813  Line 970 
970                  $loadHasProperty->Put($fid, $propertyID, $url);                  $loadHasProperty->Put($fid, $propertyID, $url);
971              }              }
972          }          }
973                # Update the statistics.
974                Trace("$propertyCount attributes processed for $featureCount features.") if T(3);
975                $loadHasProperty->Add("featuresIn", $featureCount);
976                $loadHasProperty->Add("propertiesIn", $propertyCount);
977            }
978      }      }
979      # Finish the load.      # Finish the load.
980      my $retVal = $self->_FinishAll();      my $retVal = $self->_FinishAll();
# Line 853  Line 1015 
1015      my $fig = $self->{fig};      my $fig = $self->{fig};
1016      # Get the genome hash.      # Get the genome hash.
1017      my $genomeHash = $self->{genomes};      my $genomeHash = $self->{genomes};
     my $genomeCount = (keys %{$genomeHash});  
1018      # Create load objects for each of the tables we're loading.      # Create load objects for each of the tables we're loading.
1019      my $loadAnnotation = $self->_TableLoader('Annotation', $genomeCount * 4000);      my $loadAnnotation = $self->_TableLoader('Annotation');
1020      my $loadIsTargetOfAnnotation = $self->_TableLoader('IsTargetOfAnnotation', $genomeCount * 4000);      my $loadIsTargetOfAnnotation = $self->_TableLoader('IsTargetOfAnnotation', $self->PrimaryOnly);
1021      my $loadSproutUser = $self->_TableLoader('SproutUser', 100);      my $loadSproutUser = $self->_TableLoader('SproutUser', $self->PrimaryOnly);
1022      my $loadUserAccess = $self->_TableLoader('UserAccess', 1000);      my $loadUserAccess = $self->_TableLoader('UserAccess', $self->PrimaryOnly);
1023      my $loadMadeAnnotation = $self->_TableLoader('MadeAnnotation', $genomeCount * 4000);      my $loadMadeAnnotation = $self->_TableLoader('MadeAnnotation', $self->PrimaryOnly);
1024      Trace("Beginning annotation data load.") if T(2);      if ($self->{options}->{loadOnly}) {
1025            Trace("Loading from existing files.") if T(2);
1026        } else {
1027            Trace("Generating annotation data.") if T(2);
1028      # Create a hash of user names. We'll use this to prevent us from generating duplicate      # Create a hash of user names. We'll use this to prevent us from generating duplicate
1029      # user records.      # user records.
1030      my %users = ( FIG => 1, master => 1 );      my %users = ( FIG => 1, master => 1 );
# Line 872  Line 1036 
1036      # Get the current time.      # Get the current time.
1037      my $time = time();      my $time = time();
1038      # Loop through the genomes.      # Loop through the genomes.
1039      for my $genomeID (%{$genomeHash}) {          for my $genomeID (sort keys %{$genomeHash}) {
1040          Trace("Processing $genomeID.") if T(3);          Trace("Processing $genomeID.") if T(3);
         # Get the genome's PEGs.  
         my @pegs = $fig->pegs_of($genomeID);  
         for my $peg (@pegs) {  
             Trace("Processing $peg.") if T(4);  
1041              # Create a hash of timestamps. We use this to prevent duplicate time stamps              # Create a hash of timestamps. We use this to prevent duplicate time stamps
1042              # from showing up for a single PEG's annotations.              # from showing up for a single PEG's annotations.
1043              my %seenTimestamps = ();              my %seenTimestamps = ();
1044              # Check for a functional assignment.              # Get the genome's annotations.
1045              my $func = $fig->function_of($peg);              my @annotations = $fig->read_all_annotations($genomeID);
1046              if ($func) {              Trace("Processing annotations.") if T(2);
1047                  # If this is NOT a hypothetical assignment, we create an              for my $tuple (@annotations) {
1048                  # assignment annotation for it.                  # Get the annotation tuple.
1049                  if (! FIG::hypo($peg)) {                  my ($peg, $timestamp, $user, $text) = @{$tuple};
                     # Note that we double the slashes so that what goes into the database is  
                     # a new-line escape sequence rather than an actual new-line.  
                     $loadAnnotation->Put("$peg:$time", $time, "FIG\\nSet function to\\n$func");  
                     $loadIsTargetOfAnnotation->Put($peg, "$peg:$time");  
                     $loadMadeAnnotation->Put("FIG", "$peg:$time");  
                     # Denote we've seen this timestamp.  
                     $seenTimestamps{$time} = 1;  
                 }  
                 # Now loop through the real annotations.  
                 for my $tuple ($fig->feature_annotations($peg, "raw")) {  
                     my ($fid, $timestamp, $user, $text) = $tuple;  
1050                      # Here we fix up the annotation text. "\r" is removed,                      # Here we fix up the annotation text. "\r" is removed,
1051                      # and "\t" and "\n" are escaped. Note we use the "s"                  # and "\t" and "\n" are escaped. Note we use the "gs"
1052                      # modifier so that new-lines inside the text do not                      # modifier so that new-lines inside the text do not
1053                      # stop the substitution search.                      # stop the substitution search.
1054                      $text =~ s/\r//gs;                      $text =~ s/\r//gs;
# Line 909  Line 1058 
1058                      $text =~ s/Set master function/Set FIG function/s;                      $text =~ s/Set master function/Set FIG function/s;
1059                      # Insure the time stamp is valid.                      # Insure the time stamp is valid.
1060                      if ($timestamp =~ /^\d+$/) {                      if ($timestamp =~ /^\d+$/) {
1061                          # Here it's a number. We need to insure it's unique.                      # Here it's a number. We need to insure the one we use to form
1062                          while ($seenTimestamps{$timestamp}) {                      # the key is unique.
1063                              $timestamp++;                      my $keyStamp = $timestamp;
1064                        while ($seenTimestamps{"$peg:$keyStamp"}) {
1065                            $keyStamp++;
1066                          }                          }
1067                          $seenTimestamps{$timestamp} = 1;                      my $annotationID = "$peg:$keyStamp";
1068                          my $annotationID = "$peg:$timestamp";                      $seenTimestamps{$annotationID} = 1;
1069                          # Insure the user exists.                          # Insure the user exists.
1070                          if (! $users{$user}) {                          if (! $users{$user}) {
1071                              $loadSproutUser->Put($user, "SEED user");                              $loadSproutUser->Put($user, "SEED user");
# Line 922  Line 1073 
1073                              $users{$user} = 1;                              $users{$user} = 1;
1074                          }                          }
1075                          # Generate the annotation.                          # Generate the annotation.
1076                          $loadAnnotation->Put($annotationID, $timestamp, "$user\\n$text");                      $loadAnnotation->Put($annotationID, $timestamp, $text);
1077                          $loadIsTargetOfAnnotation->Put($peg, $annotationID);                          $loadIsTargetOfAnnotation->Put($peg, $annotationID);
1078                          $loadMadeAnnotation->Put($user, $annotationID);                          $loadMadeAnnotation->Put($user, $annotationID);
1079                      } else {                      } else {
# Line 932  Line 1083 
1083                  }                  }
1084              }              }
1085          }          }
     }  
1086      # Finish the load.      # Finish the load.
1087      my $retVal = $self->_FinishAll();      my $retVal = $self->_FinishAll();
1088      return $retVal;      return $retVal;
# Line 973  Line 1123 
1123      my $fig = $self->{fig};      my $fig = $self->{fig};
1124      # Get the genome hash.      # Get the genome hash.
1125      my $genomeHash = $self->{genomes};      my $genomeHash = $self->{genomes};
     my $genomeCount = (keys %{$genomeHash});  
1126      # Create load objects for each of the tables we're loading.      # Create load objects for each of the tables we're loading.
1127      my $loadComesFrom = $self->_TableLoader('ComesFrom', $genomeCount * 4);      my $loadComesFrom = $self->_TableLoader('ComesFrom', $self->PrimaryOnly);
1128      my $loadSource = $self->_TableLoader('Source', $genomeCount * 4);      my $loadSource = $self->_TableLoader('Source');
1129      my $loadSourceURL = $self->_TableLoader('SourceURL', $genomeCount * 8);      my $loadSourceURL = $self->_TableLoader('SourceURL');
1130      Trace("Beginning source data load.") if T(2);      if ($self->{options}->{loadOnly}) {
1131            Trace("Loading from existing files.") if T(2);
1132        } else {
1133            Trace("Generating annotation data.") if T(2);
1134      # Create hashes to collect the Source information.      # Create hashes to collect the Source information.
1135      my %sourceURL = ();      my %sourceURL = ();
1136      my %sourceDesc = ();      my %sourceDesc = ();
1137      # Loop through the genomes.      # Loop through the genomes.
1138      my $line;      my $line;
1139      for my $genomeID (%{$genomeHash}) {          for my $genomeID (sort keys %{$genomeHash}) {
1140          Trace("Processing $genomeID.") if T(3);          Trace("Processing $genomeID.") if T(3);
1141          # Open the project file.          # Open the project file.
1142          if ((open(TMP, "<$FIG_Config::organisms/$genomeID/PROJECT")) &&          if ((open(TMP, "<$FIG_Config::organisms/$genomeID/PROJECT")) &&
1143              defined($line = <TMP>)) {              defined($line = <TMP>)) {
1144              chomp $line;              chomp $line;
1145              my($sourceID, $desc, $url) = split(/\t/,$_);                  my($sourceID, $desc, $url) = split(/\t/,$line);
1146              $loadComesFrom->Put($genomeID, $sourceID);              $loadComesFrom->Put($genomeID, $sourceID);
1147              if ($url && ! exists $sourceURL{$genomeID}) {                  if ($url && ! exists $sourceURL{$sourceID}) {
1148                  $loadSourceURL->Put($sourceID, $url);                  $loadSourceURL->Put($sourceID, $url);
1149                  $sourceURL{$sourceID} = 1;                  $sourceURL{$sourceID} = 1;
1150              }              }
1151              if ($desc && ! exists $sourceDesc{$sourceID}) {                  if ($desc) {
1152                  $loadSource->Put($sourceID, $desc);                      $sourceDesc{$sourceID} = $desc;
1153                  $sourceDesc{$sourceID} = 1;                  } elsif (! exists $sourceDesc{$sourceID}) {
1154                        $sourceDesc{$sourceID} = $sourceID;
1155              }              }
1156          }          }
1157          close TMP;          close TMP;
1158      }      }
1159            # Write the source descriptions.
1160            for my $sourceID (keys %sourceDesc) {
1161                $loadSource->Put($sourceID, $sourceDesc{$sourceID});
1162            }
1163        }
1164        # Finish the load.
1165        my $retVal = $self->_FinishAll();
1166        return $retVal;
1167    }
1168    
1169    =head3 LoadExternalData
1170    
1171    C<< my $stats = $spl->LoadExternalData(); >>
1172    
1173    Load the external data from FIG into Sprout.
1174    
1175    External data contains information about external feature IDs.
1176    
1177    The following relations are loaded by this method.
1178    
1179        ExternalAliasFunc
1180        ExternalAliasOrg
1181    
1182    The support for external IDs in FIG is hidden beneath layers of other data, so
1183    we access the SEED files directly to create these tables. This is also one of
1184    the few load methods that does not proceed genome by genome.
1185    
1186    =over 4
1187    
1188    =item RETURNS
1189    
1190    Returns a statistics object for the loads.
1191    
1192    =back
1193    
1194    =cut
1195    #: Return Type $%;
1196    sub LoadExternalData {
1197        # Get this object instance.
1198        my ($self) = @_;
1199        # Get the FIG object.
1200        my $fig = $self->{fig};
1201        # Get the genome hash.
1202        my $genomeHash = $self->{genomes};
1203        # Convert the genome hash. We'll get the genus and species for each genome and make
1204        # it the key.
1205        my %speciesHash = map { $fig->genus_species($_) => $_ } (keys %{$genomeHash});
1206        # Create load objects for each of the tables we're loading.
1207        my $loadExternalAliasFunc = $self->_TableLoader('ExternalAliasFunc');
1208        my $loadExternalAliasOrg = $self->_TableLoader('ExternalAliasOrg');
1209        if ($self->{options}->{loadOnly}) {
1210            Trace("Loading from existing files.") if T(2);
1211        } else {
1212            Trace("Generating external data.") if T(2);
1213            # We loop through the files one at a time. First, the organism file.
1214            Open(\*ORGS, "<$FIG_Config::global/ext_org.table");
1215            my $orgLine;
1216            while (defined($orgLine = <ORGS>)) {
1217                # Clean the input line.
1218                chomp $orgLine;
1219                # Parse the organism name.
1220                my ($protID, $name) = split /\s*\t\s*/, $orgLine;
1221                $loadExternalAliasOrg->Put($protID, $name);
1222            }
1223            close ORGS;
1224            # Now the function file.
1225            my $funcLine;
1226            Open(\*FUNCS, "<$FIG_Config::global/ext_func.table");
1227            while (defined($funcLine = <FUNCS>)) {
1228                # Clean the line ending.
1229                chomp $funcLine;
1230                # Only proceed if the line is non-blank.
1231                if ($funcLine) {
1232                    # Split it into fields.
1233                    my @funcFields = split /\s*\t\s*/, $funcLine;
1234                    # If there's an EC number, append it to the description.
1235                    if ($#funcFields >= 2 && $funcFields[2] =~ /^(EC .*\S)/) {
1236                        $funcFields[1] .= " $1";
1237                    }
1238                    # Output the function line.
1239                    $loadExternalAliasFunc->Put(@funcFields[0,1]);
1240                }
1241            }
1242        }
1243      # Finish the load.      # Finish the load.
1244      my $retVal = $self->_FinishAll();      my $retVal = $self->_FinishAll();
1245      return $retVal;      return $retVal;
1246  }  }
1247    
1248    
1249    =head3 LoadReactionData
1250    
1251    C<< my $stats = $spl->LoadReactionData(); >>
1252    
1253    Load the reaction data from FIG into Sprout.
1254    
1255    Reaction data connects reactions to the compounds that participate in them.
1256    
1257    The following relations are loaded by this method.
1258    
1259        Reaction
1260        ReactionURL
1261        Compound
1262        CompoundName
1263        CompoundCAS
1264        IsAComponentOf
1265    
1266    This method proceeds reaction by reaction rather than genome by genome.
1267    
1268    =over 4
1269    
1270    =item RETURNS
1271    
1272    Returns a statistics object for the loads.
1273    
1274    =back
1275    
1276    =cut
1277    #: Return Type $%;
1278    sub LoadReactionData {
1279        # Get this object instance.
1280        my ($self) = @_;
1281        # Get the FIG object.
1282        my $fig = $self->{fig};
1283        # Create load objects for each of the tables we're loading.
1284        my $loadReaction = $self->_TableLoader('Reaction');
1285        my $loadReactionURL = $self->_TableLoader('ReactionURL', $self->PrimaryOnly);
1286        my $loadCompound = $self->_TableLoader('Compound', $self->PrimaryOnly);
1287        my $loadCompoundName = $self->_TableLoader('CompoundName', $self->PrimaryOnly);
1288        my $loadCompoundCAS = $self->_TableLoader('CompoundCAS', $self->PrimaryOnly);
1289        my $loadIsAComponentOf = $self->_TableLoader('IsAComponentOf', $self->PrimaryOnly);
1290        if ($self->{options}->{loadOnly}) {
1291            Trace("Loading from existing files.") if T(2);
1292        } else {
1293            Trace("Generating annotation data.") if T(2);
1294            # First we create the compounds.
1295            my @compounds = $fig->all_compounds();
1296            for my $cid (@compounds) {
1297                # Check for names.
1298                my @names = $fig->names_of_compound($cid);
1299                # Each name will be given a priority number, starting with 1.
1300                my $prio = 1;
1301                for my $name (@names) {
1302                    $loadCompoundName->Put($cid, $name, $prio++);
1303                }
1304                # Create the main compound record. Note that the first name
1305                # becomes the label.
1306                my $label = (@names > 0 ? $names[0] : $cid);
1307                $loadCompound->Put($cid, $label);
1308                # Check for a CAS ID.
1309                my $cas = $fig->cas($cid);
1310                if ($cas) {
1311                    $loadCompoundCAS->Put($cid, $cas);
1312                }
1313            }
1314            # All the compounds are set up, so we need to loop through the reactions next. First,
1315            # we initialize the discriminator index. This is a single integer used to insure
1316            # duplicate elements in a reaction are not accidentally collapsed.
1317            my $discrim = 0;
1318            my @reactions = $fig->all_reactions();
1319            for my $reactionID (@reactions) {
1320                # Create the reaction record.
1321                $loadReaction->Put($reactionID, $fig->reversible($reactionID));
1322                # Compute the reaction's URL.
1323                my $url = HTML::reaction_link($reactionID);
1324                # Put it in the ReactionURL table.
1325                $loadReactionURL->Put($reactionID, $url);
1326                # Now we need all of the reaction's compounds. We get these in two phases,
1327                # substrates first and then products.
1328                for my $product (0, 1) {
1329                    # Get the compounds of the current type for the current reaction. FIG will
1330                    # give us 3-tuples: [ID, stoichiometry, main-flag]. At this time we do not
1331                    # have location data in SEED, so it defaults to the empty string.
1332                    my @compounds = $fig->reaction2comp($reactionID, $product);
1333                    for my $compData (@compounds) {
1334                        # Extract the compound data from the current tuple.
1335                        my ($cid, $stoich, $main) = @{$compData};
1336                        # Link the compound to the reaction.
1337                        $loadIsAComponentOf->Put($cid, $reactionID, $discrim++, "", $main,
1338                                                 $product, $stoich);
1339                    }
1340                }
1341            }
1342        }
1343        # Finish the load.
1344        my $retVal = $self->_FinishAll();
1345        return $retVal;
1346    }
1347    
1348  =head3 LoadGroupData  =head3 LoadGroupData
1349    
1350  C<< my $stats = $spl->LoadGroupData(); >>  C<< my $stats = $spl->LoadGroupData(); >>
# Line 1039  Line 1375 
1375      my $fig = $self->{fig};      my $fig = $self->{fig};
1376      # Get the genome hash.      # Get the genome hash.
1377      my $genomeHash = $self->{genomes};      my $genomeHash = $self->{genomes};
     my $genomeCount = (keys %{$genomeHash});  
1378      # Create a load object for the table we're loading.      # Create a load object for the table we're loading.
1379      my $loadGenomeGroups = $self->_TableLoader('GenomeGroups', $genomeCount * 4);      my $loadGenomeGroups = $self->_TableLoader('GenomeGroups');
1380      Trace("Beginning group data load.") if T(2);      if ($self->{options}->{loadOnly}) {
1381            Trace("Loading from existing files.") if T(2);
1382        } else {
1383            Trace("Generating group data.") if T(2);
1384      # Loop through the genomes.      # Loop through the genomes.
1385      my $line;      my $line;
1386      for my $genomeID (%{$genomeHash}) {          for my $genomeID (keys %{$genomeHash}) {
1387          Trace("Processing $genomeID.") if T(3);          Trace("Processing $genomeID.") if T(3);
1388          # Open the NMPDR group file for this genome.          # Open the NMPDR group file for this genome.
1389          if (open(TMP, "<$FIG_Config::organisms/$genomeID/NMPDR") &&          if (open(TMP, "<$FIG_Config::organisms/$genomeID/NMPDR") &&
1390              defined($line = <TMP>)) {              defined($line = <TMP>)) {
1391              # Clean the line ending.              # Clean the line ending.
1392              chomp;                  chomp $line;
1393              # Add the group to the table. Note that there can only be one group              # Add the group to the table. Note that there can only be one group
1394              # per genome.              # per genome.
1395              $loadGenomeGroups->Put($genomeID, $line);              $loadGenomeGroups->Put($genomeID, $line);
1396          }          }
1397          close TMP;          close TMP;
1398      }      }
1399        }
1400        # Finish the load.
1401        my $retVal = $self->_FinishAll();
1402        return $retVal;
1403    }
1404    
1405    =head3 LoadSynonymData
1406    
1407    C<< my $stats = $spl->LoadSynonymData(); >>
1408    
1409    Load the synonym groups into Sprout.
1410    
1411    The following relations are loaded by this method.
1412    
1413        SynonymGroup
1414        IsSynonymGroupFor
1415    
1416    The source information for these relations is taken from the C<maps_to_id> method
1417    of the B<FIG> object. The process starts from the features, so it is possible
1418    that there will be duplicates in the SynonymGroup load file, since the relationship
1419    is one-to-many toward the features. The automatic sort on primary entity relations
1420    will fix this for us.
1421    
1422    =over 4
1423    
1424    =item RETURNS
1425    
1426    Returns a statistics object for the loads.
1427    
1428    =back
1429    
1430    =cut
1431    #: Return Type $%;
1432    sub LoadSynonymData {
1433        # Get this object instance.
1434        my ($self) = @_;
1435        # Get the FIG object.
1436        my $fig = $self->{fig};
1437        # Get the genome hash.
1438        my $genomeHash = $self->{genomes};
1439        # Create a load object for the table we're loading.
1440        my $loadSynonymGroup = $self->_TableLoader('SynonymGroup');
1441        my $loadIsSynonymGroupFor = $self->_TableLoader('IsSynonymGroupFor');
1442        if ($self->{options}->{loadOnly}) {
1443            Trace("Loading from existing files.") if T(2);
1444        } else {
1445            Trace("Generating synonym group data.") if T(2);
1446            # Loop through the genomes.
1447            for my $genomeID (sort keys %{$genomeHash}) {
1448                Trace("Processing $genomeID.") if T(3);
1449                # Get all of the features for this genome. The only method that does this is
1450                # all_features_detailed, which returns extra baggage that we discard.
1451                my $featureData = $fig->all_features_detailed($genomeID);
1452                my @fids = map { $_->[0] } @{$featureData};
1453                Trace(scalar(@fids) . " features found for genome $genomeID.") if T(3);
1454                # Loop through the feature IDs.
1455                for my $fid (@fids) {
1456                    # Get the group for this feature.
1457                    my $synonym = $fig->maps_to_id($fid);
1458                    # Only proceed if the synonym is a real group.
1459                    if ($synonym ne $fid) {
1460                        $loadSynonymGroup->Put($synonym);
1461                        $loadIsSynonymGroupFor->Put($synonym, $fid);
1462                    }
1463                }
1464            }
1465        }
1466      # Finish the load.      # Finish the load.
1467      my $retVal = $self->_FinishAll();      my $retVal = $self->_FinishAll();
1468      return $retVal;      return $retVal;
1469  }  }
1470    
1471    
1472  =head2 Internal Utility Methods  =head2 Internal Utility Methods
1473    
1474  =head3 TableLoader  =head3 TableLoader
# Line 1079  Line 1485 
1485    
1486  Name of the table (relation) being loaded.  Name of the table (relation) being loaded.
1487    
1488  =item rowCount (optional)  =item ignore
1489    
1490  Estimated maximum number of rows in the table.  TRUE if the table should be ignored entirely, else FALSE.
1491    
1492  =item RETURN  =item RETURN
1493    
# Line 1093  Line 1499 
1499    
1500  sub _TableLoader {  sub _TableLoader {
1501      # Get the parameters.      # Get the parameters.
1502      my ($self, $tableName, $rowCount) = @_;      my ($self, $tableName, $ignore) = @_;
1503      # Create the load object.      # Create the load object.
1504      my $retVal = ERDBLoad->new($self->{erdb}, $tableName, $self->{loadDirectory}, $rowCount);      my $retVal = ERDBLoad->new($self->{erdb}, $tableName, $self->{loadDirectory}, $self->LoadOnly,
1505                                   $ignore);
1506      # Cache it in the loader list.      # Cache it in the loader list.
1507      push @{$self->{loaders}}, $retVal;      push @{$self->{loaders}}, $retVal;
1508      # Return it to the caller.      # Return it to the caller.
# Line 1129  Line 1536 
1536      my $retVal = Stats->new();      my $retVal = Stats->new();
1537      # Get the loader list.      # Get the loader list.
1538      my $loadList = $self->{loaders};      my $loadList = $self->{loaders};
1539        # Create a hash to hold the statistics objects, keyed on relation name.
1540        my %loaderHash = ();
1541      # Loop through the list, finishing the loads. Note that if the finish fails, we die      # Loop through the list, finishing the loads. Note that if the finish fails, we die
1542      # ignominiously. At some future point, we want to make the loads restartable.      # ignominiously. At some future point, we want to make the loads more restartable.
1543      while (my $loader = pop @{$loadList}) {      while (my $loader = pop @{$loadList}) {
1544            # Get the relation name.
1545            my $relName = $loader->RelName;
1546            # Check the ignore flag.
1547            if ($loader->Ignore) {
1548                Trace("Relation $relName not loaded.") if T(2);
1549            } else {
1550                # Here we really need to finish.
1551                Trace("Finishing $relName.") if T(2);
1552          my $stats = $loader->Finish();          my $stats = $loader->Finish();
1553                $loaderHash{$relName} = $stats;
1554            }
1555        }
1556        # Now we loop through again, actually loading the tables. We want to finish before
1557        # loading so that if something goes wrong at this point, all the load files are usable
1558        # and we don't have to redo all that work.
1559        for my $relName (sort keys %loaderHash) {
1560            # Get the statistics for this relation.
1561            my $stats = $loaderHash{$relName};
1562            # Check for a database load.
1563            if ($self->{options}->{dbLoad}) {
1564                # Here we want to use the load file just created to load the database.
1565                Trace("Loading relation $relName.") if T(2);
1566                my $newStats = $self->{sprout}->LoadUpdate(1, [$relName]);
1567                # Accumulate the statistics from the DB load.
1568                $stats->Accumulate($newStats);
1569            }
1570          $retVal->Accumulate($stats);          $retVal->Accumulate($stats);
         my $relName = $loader->RelName;  
1571          Trace("Statistics for $relName:\n" . $stats->Show()) if T(2);          Trace("Statistics for $relName:\n" . $stats->Show()) if T(2);
1572      }      }
1573        }
1574      # Return the load statistics.      # Return the load statistics.
1575      return $retVal;      return $retVal;
1576  }  }

Legend:
Removed from v.1.5  
changed lines
  Added in v.1.48

MCS Webmaster
ViewVC Help
Powered by ViewVC 1.0.3