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

Diff of /Sprout/SproutLoad.pm

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

revision 1.7, Tue Sep 13 19:05:20 2005 UTC revision 1.70, Wed Sep 27 12:58:18 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 114  Line 120 
120                  # an omitted access code can be defaulted to 1.                  # an omitted access code can be defaulted to 1.
121                  for my $genomeLine (@genomeList) {                  for my $genomeLine (@genomeList) {
122                      my ($genomeID, $accessCode) = split("\t", $genomeLine);                      my ($genomeID, $accessCode) = split("\t", $genomeLine);
123                      if (undef $accessCode) {                          if (! defined($accessCode)) {
124                          $accessCode = 1;                          $accessCode = 1;
125                      }                      }
126                      $genomes{$genomeID} = $accessCode;                      $genomes{$genomeID} = $accessCode;
# Line 124  Line 130 
130              Confess("Invalid genome parameter ($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 usable subsystems. First we get the whole list.
140          %subsystems = map { $_ => 1 } $fig->all_subsystems();              my @subs = $fig->all_subsystems();
141                # Loop through, checking for usability.
142                for my $sub (@subs) {
143                    if ($fig->usable_subsystem($sub)) {
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");          $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                # Open the NMPDR group file for this genome.
270                my $group;
271                if (open(TMP, "<$FIG_Config::organisms/$genomeID/NMPDR") &&
272                    defined($group = <TMP>)) {
273                    # Clean the line ending.
274                    chomp $group;
275                } else {
276                    # No group, so use the default.
277                    $group = $FIG_Config::otherGroup;
278                }
279                close TMP;
280          # Output the genome record.          # Output the genome record.
281          $loadGenome->Put($genomeID, $accessCode, $fig->is_complete($genomeID), $genus,          $loadGenome->Put($genomeID, $accessCode, $fig->is_complete($genomeID), $genus,
282                           $species, $extra, $taxonomy);                               $group, $species, $extra, $taxonomy);
283          # Now we loop through each of the genome's contigs.          # Now we loop through each of the genome's contigs.
284          my @contigs = $fig->all_contigs($genomeID);          my @contigs = $fig->all_contigs($genomeID);
285          for my $contigID (@contigs) {          for my $contigID (@contigs) {
# Line 262  Line 310 
310              }              }
311          }          }
312      }      }
313        }
314      # Finish the loads.      # Finish the loads.
315      my $retVal = $self->_FinishAll();      my $retVal = $self->_FinishAll();
316      # Return the result.      # Return the result.
# Line 302  Line 351 
351      my $fig = $self->{fig};      my $fig = $self->{fig};
352      # Get the genome hash.      # Get the genome hash.
353      my $genomeFilter = $self->{genomes};      my $genomeFilter = $self->{genomes};
354      my $genomeCount = (keys %{$genomeFilter});      # Set up an ID counter for the PCHs.
355      my $featureCount = $genomeCount * 4000;      my $pchID = 0;
356      # Start the loads.      # Start the loads.
357      my $loadCoupling = $self->_TableLoader('Coupling', $featureCount * $genomeCount);      my $loadCoupling = $self->_TableLoader('Coupling');
358      my $loadIsEvidencedBy = $self->_TableLoader('IsEvidencedBy', $featureCount * 8000);      my $loadIsEvidencedBy = $self->_TableLoader('IsEvidencedBy', $self->PrimaryOnly);
359      my $loadPCH = $self->_TableLoader('PCH', $featureCount * 2000);      my $loadPCH = $self->_TableLoader('PCH', $self->PrimaryOnly);
360      my $loadParticipatesInCoupling = $self->_TableLoader('ParticipatesInCoupling', $featureCount * 2000);      my $loadParticipatesInCoupling = $self->_TableLoader('ParticipatesInCoupling', $self->PrimaryOnly);
361      my $loadUsesAsEvidence = $self->_TableLoader('UsesAsEvidence', $featureCount * 8000);      my $loadUsesAsEvidence = $self->_TableLoader('UsesAsEvidence', $self->PrimaryOnly);
362      Trace("Beginning coupling data load.") if T(2);      if ($self->{options}->{loadOnly}) {
363            Trace("Loading from existing files.") if T(2);
364        } else {
365            Trace("Generating coupling data.") if T(2);
366      # Loop through the genomes found.      # Loop through the genomes found.
367      for my $genome (sort keys %{$genomeFilter}) {      for my $genome (sort keys %{$genomeFilter}) {
368          Trace("Generating coupling data for $genome.") if T(3);          Trace("Generating coupling data for $genome.") if T(3);
# Line 334  Line 386 
386              for my $coupleData (@couplings) {              for my $coupleData (@couplings) {
387                  my ($peg2, $score) = @{$coupleData};                  my ($peg2, $score) = @{$coupleData};
388                  # Compute the coupling ID.                  # Compute the coupling ID.
389                  my $coupleID = Sprout::CouplingID($peg1, $peg2);                      my $coupleID = $self->{erdb}->CouplingID($peg1, $peg2);
390                  if (! exists $dupHash{$coupleID}) {                  if (! exists $dupHash{$coupleID}) {
391                      $loadCoupling->Add("couplingIn");                      $loadCoupling->Add("couplingIn");
392                      # Here we have a new coupling to store in the load files.                      # Here we have a new coupling to store in the load files.
# Line 362  Line 414 
414                              # We store this evidence in the hash if the usage                              # We store this evidence in the hash if the usage
415                              # is nonzero or no prior evidence has been found. This                              # is nonzero or no prior evidence has been found. This
416                              # insures that if there is duplicate evidence, we                              # insures that if there is duplicate evidence, we
417                              # at least keep the meaningful ones. Only evidence is                                  # at least keep the meaningful ones. Only evidence in
418                              # the hash makes it to the output.                              # the hash makes it to the output.
419                              if ($usage || ! exists $evidenceMap{$evidenceKey}) {                              if ($usage || ! exists $evidenceMap{$evidenceKey}) {
420                                  $evidenceMap{$evidenceKey} = $evidenceData;                                  $evidenceMap{$evidenceKey} = $evidenceData;
# Line 370  Line 422 
422                          }                          }
423                      }                      }
424                      for my $evidenceID (keys %evidenceMap) {                      for my $evidenceID (keys %evidenceMap) {
425                                # Get the ID for this evidence.
426                                $pchID++;
427                          # Create the evidence record.                          # Create the evidence record.
428                          my ($peg3, $peg4, $usage) = @{$evidenceMap{$evidenceID}};                          my ($peg3, $peg4, $usage) = @{$evidenceMap{$evidenceID}};
429                          $loadPCH->Put($evidenceID, $usage);                              $loadPCH->Put($pchID, $usage);
430                          # Connect it to the coupling.                          # Connect it to the coupling.
431                          $loadIsEvidencedBy->Put($coupleID, $evidenceID);                              $loadIsEvidencedBy->Put($coupleID, $pchID);
432                          # Connect it to the features.                          # Connect it to the features.
433                          $loadUsesAsEvidence->Put($evidenceID, $peg3, 1);                              $loadUsesAsEvidence->Put($pchID, $peg3, 1);
434                          $loadUsesAsEvidence->Put($evidenceID, $peg4, 1);                              $loadUsesAsEvidence->Put($pchID, $peg4, 2);
435                            }
436                      }                      }
437                  }                  }
438              }              }
# Line 404  Line 459 
459      FeatureTranslation      FeatureTranslation
460      FeatureUpstream      FeatureUpstream
461      IsLocatedIn      IsLocatedIn
462        HasFeature
463        HasRoleInSubsystem
464    
465  =over 4  =over 4
466    
# Line 422  Line 479 
479      my $fig = $self->{fig};      my $fig = $self->{fig};
480      # Get the table of genome IDs.      # Get the table of genome IDs.
481      my $genomeHash = $self->{genomes};      my $genomeHash = $self->{genomes};
     my $genomeCount = (keys %{$genomeHash});  
     my $featureCount = $genomeCount * 4000;  
482      # Create load objects for each of the tables we're loading.      # Create load objects for each of the tables we're loading.
483      my $loadFeature = $self->_TableLoader('Feature', $featureCount);      my $loadFeature = $self->_TableLoader('Feature');
484      my $loadFeatureAlias = $self->_TableLoader('FeatureAlias', $featureCount * 6);      my $loadIsLocatedIn = $self->_TableLoader('IsLocatedIn', $self->PrimaryOnly);
485      my $loadFeatureLink = $self->_TableLoader('FeatureLink', $featureCount * 10);      my $loadFeatureAlias = $self->_TableLoader('FeatureAlias');
486      my $loadFeatureTranslation = $self->_TableLoader('FeatureTranslation', $featureCount);      my $loadFeatureLink = $self->_TableLoader('FeatureLink');
487      my $loadFeatureUpstream = $self->_TableLoader('FeatureUpstream', $featureCount);      my $loadFeatureTranslation = $self->_TableLoader('FeatureTranslation');
488      my $loadIsLocatedIn = $self->_TableLoader('IsLocatedIn', $featureCount);      my $loadFeatureUpstream = $self->_TableLoader('FeatureUpstream');
489        my $loadHasFeature = $self->_TableLoader('HasFeature', $self->PrimaryOnly);
490        my $loadHasRoleInSubsystem = $self->_TableLoader('HasRoleInSubsystem', $self->PrimaryOnly);
491      # 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
492      # locations.      # locations.
493      my $chunkSize = $self->{sprout}->MaxSegment();      my $chunkSize = $self->{sprout}->MaxSegment();
494      Trace("Beginning feature data load.") if T(2);      if ($self->{options}->{loadOnly}) {
495            Trace("Loading from existing files.") if T(2);
496        } else {
497            Trace("Generating feature data.") if T(2);
498      # Now we loop through the genomes, generating the data for each one.      # Now we loop through the genomes, generating the data for each one.
499      for my $genomeID (sort keys %{$genomeHash}) {      for my $genomeID (sort keys %{$genomeHash}) {
500          Trace("Loading features for genome $genomeID.") if T(3);          Trace("Loading features for genome $genomeID.") if T(3);
501          $loadFeature->Add("genomeIn");          $loadFeature->Add("genomeIn");
502          # Get the feature list for this genome.          # Get the feature list for this genome.
503          my $features = $fig->all_features_detailed($genomeID);          my $features = $fig->all_features_detailed($genomeID);
504                # Sort and count the list.
505                my @featureTuples = sort { $a->[0] cmp $b->[0] } @{$features};
506                my $count = scalar @featureTuples;
507                Trace("$count features found for genome $genomeID.") if T(3);
508                # Set up for our duplicate-feature check.
509                my $oldFeatureID = "";
510          # Loop through the features.          # Loop through the features.
511          for my $featureData (@{$features}) {              for my $featureTuple (@featureTuples) {
             $loadFeature->Add("featureIn");  
512              # Split the tuple.              # Split the tuple.
513              my ($featureID, $locations, $aliases, $type) = @{$featureData};                  my ($featureID, $locations, undef, $type) = @{$featureTuple};
514                    # Check for duplicates.
515                    if ($featureID eq $oldFeatureID) {
516                        Trace("Duplicate feature $featureID found.") if T(1);
517                    } else {
518                        $oldFeatureID = $featureID;
519                        # Count this feature.
520                        $loadFeature->Add("featureIn");
521                        # Get the functional assignment.
522                        my $assignment = $fig->function_of($featureID);
523              # Create the feature record.              # Create the feature record.
524              $loadFeature->Put($featureID, 1, $type);                      $loadFeature->Put($featureID, 1, $type, $assignment);
525                        # Link it to the parent genome.
526                        $loadHasFeature->Put($genomeID, $featureID, $type);
527              # Create the aliases.              # Create the aliases.
528              for my $alias (split /\s*,\s*/, $aliases) {                      for my $alias ($fig->feature_aliases($featureID)) {
529                  $loadFeatureAlias->Put($featureID, $alias);                  $loadFeatureAlias->Put($featureID, $alias);
530              }              }
531              # Get the links.              # Get the links.
# Line 470  Line 546 
546                      $loadFeatureUpstream->Put($featureID, $upstream);                      $loadFeatureUpstream->Put($featureID, $upstream);
547                  }                  }
548              }              }
549                        # Now we need to find the subsystems this feature participates in.
550                        my @subsystems = $fig->peg_to_subsystems($featureID);
551                        for my $subsystem (@subsystems) {
552                            $loadHasRoleInSubsystem->Put($featureID, $genomeID, $type, $subsystem);
553                        }
554              # 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
555              # 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
556              # the maximum segment size. This simplifies the genes_in_region processing              # the maximum segment size. This simplifies the genes_in_region processing
557              # for Sprout.              # for Sprout.
558              my @locationList = split /\s*,\s*/, $locations;              my @locationList = split /\s*,\s*/, $locations;
559                        # Create the location position indicator.
560                        my $i = 1;
561              # Loop through the locations.              # Loop through the locations.
562              for my $location (@locationList) {              for my $location (@locationList) {
563                  # Parse the location.                  # Parse the location.
564                  my $locObject = BasicLocation->new($location);                          my $locObject = BasicLocation->new("$genomeID:$location");
565                  # Split it into a list of chunks.                  # Split it into a list of chunks.
566                  my @locOList = ();                  my @locOList = ();
567                  while (my $peeling = $locObject->Peel($chunkSize)) {                  while (my $peeling = $locObject->Peel($chunkSize)) {
# Line 488  Line 571 
571                  push @locOList, $locObject;                  push @locOList, $locObject;
572                  # Loop through the chunks, creating IsLocatedIn records. The variable                  # Loop through the chunks, creating IsLocatedIn records. The variable
573                  # "$i" will be used to keep the location index.                  # "$i" will be used to keep the location index.
                 my $i = 1;  
574                  for my $locChunk (@locOList) {                  for my $locChunk (@locOList) {
575                      $loadIsLocatedIn->Put($featureID, $locChunk->Contig, $locChunk->Left,                      $loadIsLocatedIn->Put($featureID, $locChunk->Contig, $locChunk->Left,
576                                            $locChunk->Dir, $locChunk->Length, $i);                                            $locChunk->Dir, $locChunk->Length, $i);
# Line 497  Line 579 
579              }              }
580          }          }
581      }      }
582            }
583        }
584      # Finish the loads.      # Finish the loads.
585      my $retVal = $self->_FinishAll();      my $retVal = $self->_FinishAll();
586      return $retVal;      return $retVal;
# Line 533  Line 617 
617      my $fig = $self->{fig};      my $fig = $self->{fig};
618      # Get the table of genome IDs.      # Get the table of genome IDs.
619      my $genomeHash = $self->{genomes};      my $genomeHash = $self->{genomes};
     my $genomeCount = (keys %{$genomeHash});  
     my $featureCount = $genomeCount * 4000;  
620      # Create load objects for each of the tables we're loading.      # Create load objects for each of the tables we're loading.
621      my $loadIsBidirectionalBestHitOf = $self->_TableLoader('IsBidirectionalBestHitOf',      my $loadIsBidirectionalBestHitOf = $self->_TableLoader('IsBidirectionalBestHitOf');
622                                                             $featureCount * $genomeCount);      if ($self->{options}->{loadOnly}) {
623      Trace("Beginning BBH load.") if T(2);          Trace("Loading from existing files.") if T(2);
624        } else {
625            Trace("Generating BBH data.") if T(2);
626      # Now we loop through the genomes, generating the data for each one.      # Now we loop through the genomes, generating the data for each one.
627      for my $genomeID (sort keys %{$genomeHash}) {      for my $genomeID (sort keys %{$genomeHash}) {
628          $loadIsBidirectionalBestHitOf->Add("genomeIn");          $loadIsBidirectionalBestHitOf->Add("genomeIn");
629          Trace("Processing features for genome $genomeID.") if T(3);          Trace("Processing features for genome $genomeID.") if T(3);
630          # Get the feature list for this genome.          # Get the feature list for this genome.
631          my $features = $fig->all_features_detailed($genomeID);          my $features = $fig->all_features_detailed($genomeID);
632                # Count the BBHs we find.
633                my $bbhCount = 0;
634          # Loop through the features.          # Loop through the features.
635          for my $featureData (@{$features}) {          for my $featureData (@{$features}) {
636              # Split the tuple.              # Split the tuple.
# Line 560  Line 646 
646                  if ($genomeHash->{$targetGenomeID}) {                  if ($genomeHash->{$targetGenomeID}) {
647                      $loadIsBidirectionalBestHitOf->Put($featureID, $targetID, $targetGenomeID,                      $loadIsBidirectionalBestHitOf->Put($featureID, $targetID, $targetGenomeID,
648                                                         $score);                                                         $score);
649                            $bbhCount++;
650                  }                  }
651              }              }
652          }          }
653                Trace("$bbhCount BBHs found for $genomeID.") if T(3);
654            }
655      }      }
656      # Finish the loads.      # Finish the loads.
657      my $retVal = $self->_FinishAll();      my $retVal = $self->_FinishAll();
# Line 584  Line 673 
673  The following relations are loaded by this method.  The following relations are loaded by this method.
674    
675      Subsystem      Subsystem
676        SubsystemClass
677      Role      Role
678        RoleEC
679      SSCell      SSCell
680      ContainsFeature      ContainsFeature
681      IsGenomeOf      IsGenomeOf
# Line 592  Line 683 
683      OccursInSubsystem      OccursInSubsystem
684      ParticipatesIn      ParticipatesIn
685      HasSSCell      HasSSCell
686        ConsistsOfRoles
687        RoleSubset
688        HasRoleSubset
689        ConsistsOfGenomes
690        GenomeSubset
691        HasGenomeSubset
692        Catalyzes
693        Diagram
694        RoleOccursIn
695    
696  =over 4  =over 4
697    
# Line 601  Line 701 
701    
702  =back  =back
703    
 B<TO DO>  
   
 Generate RoleName table?  
   
704  =cut  =cut
705  #: Return Type $%;  #: Return Type $%;
706  sub LoadSubsystemData {  sub LoadSubsystemData {
# Line 618  Line 714 
714      # Get the subsystem hash. This lists the subsystems we'll process.      # Get the subsystem hash. This lists the subsystems we'll process.
715      my $subsysHash = $self->{subsystems};      my $subsysHash = $self->{subsystems};
716      my @subsysIDs = sort keys %{$subsysHash};      my @subsysIDs = sort keys %{$subsysHash};
717      my $subsysCount = @subsysIDs;      # Get the map list.
718      my $genomeCount = (keys %{$genomeHash});      my @maps = $fig->all_maps;
     my $featureCount = $genomeCount * 4000;  
719      # Create load objects for each of the tables we're loading.      # Create load objects for each of the tables we're loading.
720      my $loadSubsystem = $self->_TableLoader('Subsystem', $subsysCount);      my $loadDiagram = $self->_TableLoader('Diagram', $self->PrimaryOnly);
721      my $loadRole = $self->_TableLoader('Role', $featureCount * 6);      my $loadRoleOccursIn = $self->_TableLoader('RoleOccursIn', $self->PrimaryOnly);
722      my $loadSSCell = $self->_TableLoader('SSCell', $featureCount * $genomeCount);      my $loadSubsystem = $self->_TableLoader('Subsystem');
723      my $loadContainsFeature = $self->_TableLoader('ContainsFeature', $featureCount * $subsysCount);      my $loadRole = $self->_TableLoader('Role', $self->PrimaryOnly);
724      my $loadIsGenomeOf = $self->_TableLoader('IsGenomeOf', $featureCount * $genomeCount);      my $loadRoleEC = $self->_TableLoader('RoleEC', $self->PrimaryOnly);
725      my $loadIsRoleOf = $self->_TableLoader('IsRoleOf', $featureCount * $genomeCount);      my $loadCatalyzes = $self->_TableLoader('Catalyzes', $self->PrimaryOnly);
726      my $loadOccursInSubsystem = $self->_TableLoader('OccursInSubsystem', $featureCount * 6);      my $loadSSCell = $self->_TableLoader('SSCell', $self->PrimaryOnly);
727      my $loadParticipatesIn = $self->_TableLoader('ParticipatesIn', $subsysCount * $genomeCount);      my $loadContainsFeature = $self->_TableLoader('ContainsFeature', $self->PrimaryOnly);
728      my $loadHasSSCell = $self->_TableLoader('HasSSCell', $featureCount * $genomeCount);      my $loadIsGenomeOf = $self->_TableLoader('IsGenomeOf', $self->PrimaryOnly);
729      Trace("Beginning subsystem data load.") if T(2);      my $loadIsRoleOf = $self->_TableLoader('IsRoleOf', $self->PrimaryOnly);
730        my $loadOccursInSubsystem = $self->_TableLoader('OccursInSubsystem', $self->PrimaryOnly);
731        my $loadParticipatesIn = $self->_TableLoader('ParticipatesIn', $self->PrimaryOnly);
732        my $loadHasSSCell = $self->_TableLoader('HasSSCell', $self->PrimaryOnly);
733        my $loadRoleSubset = $self->_TableLoader('RoleSubset', $self->PrimaryOnly);
734        my $loadGenomeSubset = $self->_TableLoader('GenomeSubset', $self->PrimaryOnly);
735        my $loadConsistsOfRoles = $self->_TableLoader('ConsistsOfRoles', $self->PrimaryOnly);
736        my $loadConsistsOfGenomes = $self->_TableLoader('ConsistsOfGenomes', $self->PrimaryOnly);
737        my $loadHasRoleSubset = $self->_TableLoader('HasRoleSubset', $self->PrimaryOnly);
738        my $loadHasGenomeSubset = $self->_TableLoader('HasGenomeSubset', $self->PrimaryOnly);
739        my $loadSubsystemClass = $self->_TableLoader('SubsystemClass', $self->PrimaryOnly);
740        if ($self->{options}->{loadOnly}) {
741            Trace("Loading from existing files.") if T(2);
742        } else {
743            Trace("Generating subsystem data.") if T(2);
744            # This hash will contain the role for each EC. When we're done, this
745            # information will be used to generate the Catalyzes table.
746            my %ecToRoles = ();
747      # Loop through the subsystems. Our first task will be to create the      # Loop through the subsystems. Our first task will be to create the
748      # roles. We do this by looping through the subsystems and creating a      # roles. We do this by looping through the subsystems and creating a
749      # 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
750      # duplicates. As we move along, we'll connect the roles and subsystems.          # duplicates. As we move along, we'll connect the roles and subsystems
751            # and memorize up the reactions.
752            my ($genomeID, $roleID);
753      my %roleData = ();      my %roleData = ();
754      for my $subsysID (@subsysIDs) {      for my $subsysID (@subsysIDs) {
755                # Get the subsystem object.
756                my $sub = $fig->get_subsystem($subsysID);
757                # Only proceed if the subsystem has a spreadsheet.
758                if (! $sub->{empty_ss}) {
759          Trace("Creating subsystem $subsysID.") if T(3);          Trace("Creating subsystem $subsysID.") if T(3);
760          $loadSubsystem->Add("subsystemIn");          $loadSubsystem->Add("subsystemIn");
761          # Create the subsystem record.          # Create the subsystem record.
762          $loadSubsystem->Put($subsysID);                  my $curator = $sub->get_curator();
763          # Get the subsystem's roles.                  my $notes = $sub->get_notes();
764          my @roles = $fig->subsystem_to_roles($subsysID);                  $loadSubsystem->Put($subsysID, $curator, $notes);
765          # Connect the roles to the subsystem. If a role is new, we create                  my $classList = $fig->subsystem_classification($subsysID);
766          # a role record for it.                  my @classes = @$classList;
767          for my $roleID (@roles) {                  if (@classes) {
768                        for my $class (@classes) {
769                            $loadSubsystemClass->Put($subsysID, $class);
770                        }
771                    }
772                    # Connect it to its roles. Each role is a column in the subsystem spreadsheet.
773                    for (my $col = 0; defined($roleID = $sub->get_role($col)); $col++) {
774                        # Connect to this role.
775              $loadOccursInSubsystem->Add("roleIn");              $loadOccursInSubsystem->Add("roleIn");
776              $loadOccursInSubsystem->Put($roleID, $subsysID);                      $loadOccursInSubsystem->Put($roleID, $subsysID, $col);
777                        # If it's a new role, add it to the role table.
778              if (! exists $roleData{$roleID}) {              if (! exists $roleData{$roleID}) {
779                  $loadRole->Put($roleID);                          # Get the role's abbreviation.
780                            my $abbr = $sub->get_role_abbr($col);
781                            # Add the role.
782                            $loadRole->Put($roleID, $abbr);
783                  $roleData{$roleID} = 1;                  $roleData{$roleID} = 1;
784                            # Check for an EC number.
785                            if ($roleID =~ /\(EC ([^.]+\.[^.]+\.[^.]+\.[^)]+)\)\s*$/) {
786                                my $ec = $1;
787                                $loadRoleEC->Put($roleID, $ec);
788                                $ecToRoles{$ec} = $roleID;
789                            }
790              }              }
791          }          }
792          # 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
793          # 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
794          # get the genomes on the sheet.                  # to actually create the roles as we find them.
795          Trace("Creating subsystem $subsysID spreadsheet.") if T(3);          Trace("Creating subsystem $subsysID spreadsheet.") if T(3);
796          my @genomes = map { $_->[0] } @{$fig->subsystem_genomes($subsysID)};                  for (my $row = 0; defined($genomeID = $sub->get_genome($row)); $row++) {
797          for my $genomeID (@genomes) {                      # Only proceed if this is one of our genomes.
             # Only process this genome if it's one of ours.  
798              if (exists $genomeHash->{$genomeID}) {              if (exists $genomeHash->{$genomeID}) {
799                  # Connect the genome to the subsystem.                          # Count the PEGs and cells found for verification purposes.
800                  $loadParticipatesIn->Put($genomeID, $subsysID);                          my $pegCount = 0;
801                            my $cellCount = 0;
802                            # Create a list for the PEGs we find. This list will be used
803                            # to generate cluster numbers.
804                            my @pegsFound = ();
805                            # Create a hash that maps spreadsheet IDs to PEGs. We will
806                            # use this to generate the ContainsFeature data after we have
807                            # the cluster numbers.
808                            my %cellPegs = ();
809                            # Get the genome's variant code for this subsystem.
810                            my $variantCode = $sub->get_variant_code($row);
811                  # 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
812                  # part of the spreadsheet cell ID.                  # part of the spreadsheet cell ID.
813                  for (my $i = 0; $i <= $#roles; $i++) {                          for (my $col = 0; defined($roleID = $sub->get_role($col)); $col++) {
                     my $role = $roles[$i];  
814                      # Get the features in the spreadsheet cell for this genome and role.                      # Get the features in the spreadsheet cell for this genome and role.
815                      my @pegs = $fig->pegs_in_subsystem_cell($subsysID, $genomeID, $i);                              my @pegs = grep { !$fig->is_deleted_fid($_) } $sub->get_pegs_from_cell($row, $col);
816                      # Only proceed if features exist.                      # Only proceed if features exist.
817                      if (@pegs > 0) {                      if (@pegs > 0) {
818                          # Create the spreadsheet cell.                          # Create the spreadsheet cell.
819                          my $cellID = "$subsysID:$genomeID:$i";                                  $cellCount++;
820                                    my $cellID = "$subsysID:$genomeID:$col";
821                          $loadSSCell->Put($cellID);                          $loadSSCell->Put($cellID);
822                          $loadIsGenomeOf->Put($genomeID, $cellID);                          $loadIsGenomeOf->Put($genomeID, $cellID);
823                          $loadIsRoleOf->Put($role, $cellID);                                  $loadIsRoleOf->Put($roleID, $cellID);
824                          $loadHasSSCell->Put($subsysID, $cellID);                          $loadHasSSCell->Put($subsysID, $cellID);
825                          # Attach the features to it.                                  # Remember its features.
826                          for my $pegID (@pegs) {                                  push @pegsFound, @pegs;
827                              $loadContainsFeature->Put($cellID, $pegID);                                  $cellPegs{$cellID} = \@pegs;
828                                    $pegCount += @pegs;
829                                }
830                            }
831                            # If we found some cells for this genome, we need to compute clusters and
832                            # denote it participates in the subsystem.
833                            if ($pegCount > 0) {
834                                Trace("$pegCount PEGs in $cellCount cells for $genomeID.") if T(3);
835                                $loadParticipatesIn->Put($genomeID, $subsysID, $variantCode);
836                                # Create a hash mapping PEG IDs to cluster numbers.
837                                # We default to -1 for all of them.
838                                my %clusterOf = map { $_ => -1 } @pegsFound;
839                                # Partition the PEGs found into clusters.
840                                my @clusters = $fig->compute_clusters([keys %clusterOf], $sub);
841                                for (my $i = 0; $i <= $#clusters; $i++) {
842                                    my $subList = $clusters[$i];
843                                    for my $peg (@{$subList}) {
844                                        $clusterOf{$peg} = $i;
845                                    }
846                                }
847                                # Create the ContainsFeature data.
848                                for my $cellID (keys %cellPegs) {
849                                    my $cellList = $cellPegs{$cellID};
850                                    for my $cellPeg (@$cellList) {
851                                        $loadContainsFeature->Put($cellID, $cellPeg, $clusterOf{$cellPeg});
852                          }                          }
853                      }                      }
854                  }                  }
855              }              }
856          }          }
857                    # Now we need to generate the subsets. The subset names must be concatenated to
858                    # the subsystem name to make them unique keys. There are two types of subsets:
859                    # genome subsets and role subsets. We do the role subsets first.
860                    my @subsetNames = $sub->get_subset_names();
861                    for my $subsetID (@subsetNames) {
862                        # Create the subset record.
863                        my $actualID = "$subsysID:$subsetID";
864                        $loadRoleSubset->Put($actualID);
865                        # Connect the subset to the subsystem.
866                        $loadHasRoleSubset->Put($subsysID, $actualID);
867                        # Connect the subset to its roles.
868                        my @roles = $sub->get_subsetC_roles($subsetID);
869                        for my $roleID (@roles) {
870                            $loadConsistsOfRoles->Put($actualID, $roleID);
871      }      }
     # Finish the load.  
     my $retVal = $self->_FinishAll();  
     return $retVal;  
872  }  }
873                    # Next the genome subsets.
874  =head3 LoadDiagramData                  @subsetNames = $sub->get_subset_namesR();
875                    for my $subsetID (@subsetNames) {
876  C<< my $stats = $spl->LoadDiagramData(); >>                      # Create the subset record.
877                        my $actualID = "$subsysID:$subsetID";
878  Load the diagram data from FIG into Sprout.                      $loadGenomeSubset->Put($actualID);
879                        # Connect the subset to the subsystem.
880  Diagrams are used to organize functional roles. The diagram shows the                      $loadHasGenomeSubset->Put($subsysID, $actualID);
881  connections between chemicals that interact with a subsystem.                      # Connect the subset to its genomes.
882                        my @genomes = $sub->get_subsetR($subsetID);
883  The following relations are loaded by this method.                      for my $genomeID (@genomes) {
884                            $loadConsistsOfGenomes->Put($actualID, $genomeID);
885      Diagram                      }
886      RoleOccursIn                  }
887                }
888  =over 4          }
889            # Now we loop through the diagrams. We need to create the diagram records
890  =item RETURNS          # and link each diagram to its roles. Note that only roles which occur
891            # in subsystems (and therefore appear in the %ecToRoles hash) are
892  Returns a statistics object for the loads.          # included.
893            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) {  
894          Trace("Loading diagram $map.") if T(3);          Trace("Loading diagram $map.") if T(3);
895          # Get the diagram's descriptive name.          # Get the diagram's descriptive name.
896          my $name = $fig->map_name($map);          my $name = $fig->map_name($map);
# Line 740  Line 899 
899          # A hash is used to prevent duplicates.          # A hash is used to prevent duplicates.
900          my %roleHash = ();          my %roleHash = ();
901          for my $role ($fig->map_to_ecs($map)) {          for my $role ($fig->map_to_ecs($map)) {
902              if (! $roleHash{$role}) {                  if (exists $ecToRoles{$role} && ! $roleHash{$role}) {
903                  $loadRoleOccursIn->Put($role, $map);                      $loadRoleOccursIn->Put($ecToRoles{$role}, $map);
904                  $roleHash{$role} = 1;                  $roleHash{$role} = 1;
905              }              }
906          }          }
907      }      }
908            # Before we leave, we must create the Catalyzes table. We start with the reactions,
909            # then use the "ecToRoles" table to convert EC numbers to role IDs.
910            my @reactions = $fig->all_reactions();
911            for my $reactionID (@reactions) {
912                # Get this reaction's list of roles. The results will be EC numbers.
913                my @roles = $fig->catalyzed_by($reactionID);
914                # Loop through the roles, creating catalyzation records.
915                for my $thisRole (@roles) {
916                    if (exists $ecToRoles{$thisRole}) {
917                        $loadCatalyzes->Put($ecToRoles{$thisRole}, $reactionID);
918                    }
919                }
920            }
921        }
922      # Finish the load.      # Finish the load.
923      my $retVal = $self->_FinishAll();      my $retVal = $self->_FinishAll();
924      return $retVal;      return $retVal;
# Line 787  Line 960 
960      my $fig = $self->{fig};      my $fig = $self->{fig};
961      # Get the genome hash.      # Get the genome hash.
962      my $genomeHash = $self->{genomes};      my $genomeHash = $self->{genomes};
     my $genomeCount = (keys %{$genomeHash});  
963      # Create load objects for each of the tables we're loading.      # Create load objects for each of the tables we're loading.
964      my $loadProperty = $self->_TableLoader('Property', $genomeCount * 1500);      my $loadProperty = $self->_TableLoader('Property');
965      my $loadHasProperty = $self->_TableLoader('HasProperty', $genomeCount * 1500);      my $loadHasProperty = $self->_TableLoader('HasProperty', $self->PrimaryOnly);
966      Trace("Beginning property data load.") if T(2);      if ($self->{options}->{loadOnly}) {
967            Trace("Loading from existing files.") if T(2);
968        } else {
969            Trace("Generating property data.") if T(2);
970      # Create a hash for storing property IDs.      # Create a hash for storing property IDs.
971      my %propertyKeys = ();      my %propertyKeys = ();
972      my $nextID = 1;      my $nextID = 1;
973      # Loop through the genomes.      # Loop through the genomes.
974      for my $genomeID (keys %{$genomeHash}) {          for my $genomeID (sort keys %{$genomeHash}) {
975          $loadProperty->Add("genomeIn");          $loadProperty->Add("genomeIn");
976                Trace("Generating properties for $genomeID.") if T(3);
977          # 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
978          # tuples returned by "all_features_detailed". We use "all_features_detailed"          # tuples returned by "all_features_detailed". We use "all_features_detailed"
979          # rather than "all_features" because we want all features regardless of type.          # rather than "all_features" because we want all features regardless of type.
980          my @features = map { $_->[0] } @{$fig->all_features_detailed($genomeID)};          my @features = map { $_->[0] } @{$fig->all_features_detailed($genomeID)};
981                my $featureCount = 0;
982                my $propertyCount = 0;
983          # Loop through the features, creating HasProperty records.          # Loop through the features, creating HasProperty records.
984          for my $fid (@features) {          for my $fid (@features) {
             $loadProperty->Add("featureIn");  
985              # 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
986              # to insure we do not get any genome attributes.              # to insure we do not get any genome attributes.
987              my @attributeList = $fig->get_attributes($fid, '', '', '');              my @attributeList = $fig->get_attributes($fid, '', '', '');
988                    if (scalar @attributeList) {
989                        $featureCount++;
990                    }
991              # Loop through the attributes.              # Loop through the attributes.
992              for my $tuple (@attributeList) {              for my $tuple (@attributeList) {
993                        $propertyCount++;
994                  # Get this attribute value's data. Note that we throw away the FID,                  # Get this attribute value's data. Note that we throw away the FID,
995                  # since it will always be the same as the value if "$fid".                  # since it will always be the same as the value if "$fid".
996                  my (undef, $key, $value, $url) = @{$tuple};                  my (undef, $key, $value, $url) = @{$tuple};
# Line 831  Line 1012 
1012                  $loadHasProperty->Put($fid, $propertyID, $url);                  $loadHasProperty->Put($fid, $propertyID, $url);
1013              }              }
1014          }          }
1015                # Update the statistics.
1016                Trace("$propertyCount attributes processed for $featureCount features.") if T(3);
1017                $loadHasProperty->Add("featuresIn", $featureCount);
1018                $loadHasProperty->Add("propertiesIn", $propertyCount);
1019            }
1020      }      }
1021      # Finish the load.      # Finish the load.
1022      my $retVal = $self->_FinishAll();      my $retVal = $self->_FinishAll();
# Line 871  Line 1057 
1057      my $fig = $self->{fig};      my $fig = $self->{fig};
1058      # Get the genome hash.      # Get the genome hash.
1059      my $genomeHash = $self->{genomes};      my $genomeHash = $self->{genomes};
     my $genomeCount = (keys %{$genomeHash});  
1060      # Create load objects for each of the tables we're loading.      # Create load objects for each of the tables we're loading.
1061      my $loadAnnotation = $self->_TableLoader('Annotation', $genomeCount * 4000);      my $loadAnnotation = $self->_TableLoader('Annotation');
1062      my $loadIsTargetOfAnnotation = $self->_TableLoader('IsTargetOfAnnotation', $genomeCount * 4000);      my $loadIsTargetOfAnnotation = $self->_TableLoader('IsTargetOfAnnotation', $self->PrimaryOnly);
1063      my $loadSproutUser = $self->_TableLoader('SproutUser', 100);      my $loadSproutUser = $self->_TableLoader('SproutUser', $self->PrimaryOnly);
1064      my $loadUserAccess = $self->_TableLoader('UserAccess', 1000);      my $loadUserAccess = $self->_TableLoader('UserAccess', $self->PrimaryOnly);
1065      my $loadMadeAnnotation = $self->_TableLoader('MadeAnnotation', $genomeCount * 4000);      my $loadMadeAnnotation = $self->_TableLoader('MadeAnnotation', $self->PrimaryOnly);
1066      Trace("Beginning annotation data load.") if T(2);      if ($self->{options}->{loadOnly}) {
1067            Trace("Loading from existing files.") if T(2);
1068        } else {
1069            Trace("Generating annotation data.") if T(2);
1070      # 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
1071      # user records.      # user records.
1072      my %users = ( FIG => 1, master => 1 );      my %users = ( FIG => 1, master => 1 );
# Line 892  Line 1080 
1080      # Loop through the genomes.      # Loop through the genomes.
1081      for my $genomeID (sort keys %{$genomeHash}) {      for my $genomeID (sort keys %{$genomeHash}) {
1082          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);  
1083              # 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
1084              # from showing up for a single PEG's annotations.              # from showing up for a single PEG's annotations.
1085              my %seenTimestamps = ();              my %seenTimestamps = ();
1086              # Check for a functional assignment.              # Get the genome's annotations.
1087              my $func = $fig->function_of($peg);              my @annotations = $fig->read_all_annotations($genomeID);
1088              if ($func) {              Trace("Processing annotations.") if T(2);
1089                  # If this is NOT a hypothetical assignment, we create an              for my $tuple (@annotations) {
1090                  # assignment annotation for it.                  # Get the annotation tuple.
1091                  if (! FIG::hypo($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};  
1092                      # Here we fix up the annotation text. "\r" is removed,                      # Here we fix up the annotation text. "\r" is removed,
1093                      # and "\t" and "\n" are escaped. Note we use the "s"                  # and "\t" and "\n" are escaped. Note we use the "gs"
1094                      # modifier so that new-lines inside the text do not                      # modifier so that new-lines inside the text do not
1095                      # stop the substitution search.                      # stop the substitution search.
1096                      $text =~ s/\r//gs;                      $text =~ s/\r//gs;
# Line 927  Line 1100 
1100                      $text =~ s/Set master function/Set FIG function/s;                      $text =~ s/Set master function/Set FIG function/s;
1101                      # Insure the time stamp is valid.                      # Insure the time stamp is valid.
1102                      if ($timestamp =~ /^\d+$/) {                      if ($timestamp =~ /^\d+$/) {
1103                          # Here it's a number. We need to insure it's unique.                      # Here it's a number. We need to insure the one we use to form
1104                          while ($seenTimestamps{$timestamp}) {                      # the key is unique.
1105                              $timestamp++;                      my $keyStamp = $timestamp;
1106                        while ($seenTimestamps{"$peg:$keyStamp"}) {
1107                            $keyStamp++;
1108                          }                          }
1109                          $seenTimestamps{$timestamp} = 1;                      my $annotationID = "$peg:$keyStamp";
1110                          my $annotationID = "$peg:$timestamp";                      $seenTimestamps{$annotationID} = 1;
1111                          # Insure the user exists.                          # Insure the user exists.
1112                          if (! $users{$user}) {                          if (! $users{$user}) {
1113                              $loadSproutUser->Put($user, "SEED user");                              $loadSproutUser->Put($user, "SEED user");
# Line 940  Line 1115 
1115                              $users{$user} = 1;                              $users{$user} = 1;
1116                          }                          }
1117                          # Generate the annotation.                          # Generate the annotation.
1118                          $loadAnnotation->Put($annotationID, $timestamp, "$user\\n$text");                      $loadAnnotation->Put($annotationID, $timestamp, $text);
1119                          $loadIsTargetOfAnnotation->Put($peg, $annotationID);                          $loadIsTargetOfAnnotation->Put($peg, $annotationID);
1120                          $loadMadeAnnotation->Put($user, $annotationID);                          $loadMadeAnnotation->Put($user, $annotationID);
1121                      } else {                      } else {
# Line 950  Line 1125 
1125                  }                  }
1126              }              }
1127          }          }
     }  
1128      # Finish the load.      # Finish the load.
1129      my $retVal = $self->_FinishAll();      my $retVal = $self->_FinishAll();
1130      return $retVal;      return $retVal;
# Line 991  Line 1165 
1165      my $fig = $self->{fig};      my $fig = $self->{fig};
1166      # Get the genome hash.      # Get the genome hash.
1167      my $genomeHash = $self->{genomes};      my $genomeHash = $self->{genomes};
     my $genomeCount = (keys %{$genomeHash});  
1168      # Create load objects for each of the tables we're loading.      # Create load objects for each of the tables we're loading.
1169      my $loadComesFrom = $self->_TableLoader('ComesFrom', $genomeCount * 4);      my $loadComesFrom = $self->_TableLoader('ComesFrom', $self->PrimaryOnly);
1170      my $loadSource = $self->_TableLoader('Source', $genomeCount * 4);      my $loadSource = $self->_TableLoader('Source');
1171      my $loadSourceURL = $self->_TableLoader('SourceURL', $genomeCount * 8);      my $loadSourceURL = $self->_TableLoader('SourceURL');
1172      Trace("Beginning source data load.") if T(2);      if ($self->{options}->{loadOnly}) {
1173            Trace("Loading from existing files.") if T(2);
1174        } else {
1175            Trace("Generating annotation data.") if T(2);
1176      # Create hashes to collect the Source information.      # Create hashes to collect the Source information.
1177      my %sourceURL = ();      my %sourceURL = ();
1178      my %sourceDesc = ();      my %sourceDesc = ();
# Line 1010  Line 1186 
1186              chomp $line;              chomp $line;
1187              my($sourceID, $desc, $url) = split(/\t/,$line);              my($sourceID, $desc, $url) = split(/\t/,$line);
1188              $loadComesFrom->Put($genomeID, $sourceID);              $loadComesFrom->Put($genomeID, $sourceID);
1189              if ($url && ! exists $sourceURL{$genomeID}) {                  if ($url && ! exists $sourceURL{$sourceID}) {
1190                  $loadSourceURL->Put($sourceID, $url);                  $loadSourceURL->Put($sourceID, $url);
1191                  $sourceURL{$sourceID} = 1;                  $sourceURL{$sourceID} = 1;
1192              }              }
1193              if ($desc && ! exists $sourceDesc{$sourceID}) {                  if ($desc) {
1194                  $loadSource->Put($sourceID, $desc);                      $sourceDesc{$sourceID} = $desc;
1195                  $sourceDesc{$sourceID} = 1;                  } elsif (! exists $sourceDesc{$sourceID}) {
1196                        $sourceDesc{$sourceID} = $sourceID;
1197              }              }
1198          }          }
1199          close TMP;          close TMP;
1200      }      }
1201            # Write the source descriptions.
1202            for my $sourceID (keys %sourceDesc) {
1203                $loadSource->Put($sourceID, $sourceDesc{$sourceID});
1204            }
1205        }
1206      # Finish the load.      # Finish the load.
1207      my $retVal = $self->_FinishAll();      my $retVal = $self->_FinishAll();
1208      return $retVal;      return $retVal;
# Line 1060  Line 1242 
1242      my $fig = $self->{fig};      my $fig = $self->{fig};
1243      # Get the genome hash.      # Get the genome hash.
1244      my $genomeHash = $self->{genomes};      my $genomeHash = $self->{genomes};
     my $genomeCount = (keys %{$genomeHash});  
1245      # Convert the genome hash. We'll get the genus and species for each genome and make      # Convert the genome hash. We'll get the genus and species for each genome and make
1246      # it the key.      # it the key.
1247      my %speciesHash = map { $fig->genus_species($_) => $_ } (keys %{$genomeHash});      my %speciesHash = map { $fig->genus_species($_) => $_ } (keys %{$genomeHash});
1248      # Create load objects for each of the tables we're loading.      # Create load objects for each of the tables we're loading.
1249      my $loadExternalAliasFunc = $self->_TableLoader('ExternalAliasFunc', $genomeCount * 4000);      my $loadExternalAliasFunc = $self->_TableLoader('ExternalAliasFunc');
1250      my $loadExternalAliasOrg = $self->_TableLoader('ExternalAliasOrg', $genomeCount * 4000);      my $loadExternalAliasOrg = $self->_TableLoader('ExternalAliasOrg');
1251      Trace("Beginning external data load.") if T(2);      if ($self->{options}->{loadOnly}) {
1252            Trace("Loading from existing files.") if T(2);
1253        } else {
1254            Trace("Generating external data.") if T(2);
1255      # We loop through the files one at a time. First, the organism file.      # We loop through the files one at a time. First, the organism file.
1256      Open(\*ORGS, "<$FIG_Config::global/ext_org.table");          Open(\*ORGS, "sort +0 -1 -u -t\"\t\" $FIG_Config::global/ext_org.table |");
1257      my $orgLine;      my $orgLine;
1258      while (defined($orgLine = <ORGS>)) {      while (defined($orgLine = <ORGS>)) {
1259          # Clean the input line.          # Clean the input line.
# Line 1081  Line 1265 
1265      close ORGS;      close ORGS;
1266      # Now the function file.      # Now the function file.
1267      my $funcLine;      my $funcLine;
1268      Open(\*FUNCS, "<$FIG_Config::global/ext_func.table");          Open(\*FUNCS, "sort +0 -1 -u -t\"\t\" $FIG_Config::global/ext_func.table |");
1269      while (defined($funcLine = <FUNCS>)) {      while (defined($funcLine = <FUNCS>)) {
1270          # Clean the line ending.          # Clean the line ending.
1271          chomp $funcLine;          chomp $funcLine;
# Line 1097  Line 1281 
1281              $loadExternalAliasFunc->Put(@funcFields[0,1]);              $loadExternalAliasFunc->Put(@funcFields[0,1]);
1282          }          }
1283      }      }
1284        }
1285        # Finish the load.
1286        my $retVal = $self->_FinishAll();
1287        return $retVal;
1288    }
1289    
1290    
1291    =head3 LoadReactionData
1292    
1293    C<< my $stats = $spl->LoadReactionData(); >>
1294    
1295    Load the reaction data from FIG into Sprout.
1296    
1297    Reaction data connects reactions to the compounds that participate in them.
1298    
1299    The following relations are loaded by this method.
1300    
1301        Reaction
1302        ReactionURL
1303        Compound
1304        CompoundName
1305        CompoundCAS
1306        IsAComponentOf
1307    
1308    This method proceeds reaction by reaction rather than genome by genome.
1309    
1310    =over 4
1311    
1312    =item RETURNS
1313    
1314    Returns a statistics object for the loads.
1315    
1316    =back
1317    
1318    =cut
1319    #: Return Type $%;
1320    sub LoadReactionData {
1321        # Get this object instance.
1322        my ($self) = @_;
1323        # Get the FIG object.
1324        my $fig = $self->{fig};
1325        # Create load objects for each of the tables we're loading.
1326        my $loadReaction = $self->_TableLoader('Reaction');
1327        my $loadReactionURL = $self->_TableLoader('ReactionURL', $self->PrimaryOnly);
1328        my $loadCompound = $self->_TableLoader('Compound', $self->PrimaryOnly);
1329        my $loadCompoundName = $self->_TableLoader('CompoundName', $self->PrimaryOnly);
1330        my $loadCompoundCAS = $self->_TableLoader('CompoundCAS', $self->PrimaryOnly);
1331        my $loadIsAComponentOf = $self->_TableLoader('IsAComponentOf', $self->PrimaryOnly);
1332        if ($self->{options}->{loadOnly}) {
1333            Trace("Loading from existing files.") if T(2);
1334        } else {
1335            Trace("Generating annotation data.") if T(2);
1336            # First we create the compounds.
1337            my @compounds = $fig->all_compounds();
1338            for my $cid (@compounds) {
1339                # Check for names.
1340                my @names = $fig->names_of_compound($cid);
1341                # Each name will be given a priority number, starting with 1.
1342                my $prio = 1;
1343                for my $name (@names) {
1344                    $loadCompoundName->Put($cid, $name, $prio++);
1345                }
1346                # Create the main compound record. Note that the first name
1347                # becomes the label.
1348                my $label = (@names > 0 ? $names[0] : $cid);
1349                $loadCompound->Put($cid, $label);
1350                # Check for a CAS ID.
1351                my $cas = $fig->cas($cid);
1352                if ($cas) {
1353                    $loadCompoundCAS->Put($cid, $cas);
1354                }
1355            }
1356            # All the compounds are set up, so we need to loop through the reactions next. First,
1357            # we initialize the discriminator index. This is a single integer used to insure
1358            # duplicate elements in a reaction are not accidentally collapsed.
1359            my $discrim = 0;
1360            my @reactions = $fig->all_reactions();
1361            for my $reactionID (@reactions) {
1362                # Create the reaction record.
1363                $loadReaction->Put($reactionID, $fig->reversible($reactionID));
1364                # Compute the reaction's URL.
1365                my $url = HTML::reaction_link($reactionID);
1366                # Put it in the ReactionURL table.
1367                $loadReactionURL->Put($reactionID, $url);
1368                # Now we need all of the reaction's compounds. We get these in two phases,
1369                # substrates first and then products.
1370                for my $product (0, 1) {
1371                    # Get the compounds of the current type for the current reaction. FIG will
1372                    # give us 3-tuples: [ID, stoichiometry, main-flag]. At this time we do not
1373                    # have location data in SEED, so it defaults to the empty string.
1374                    my @compounds = $fig->reaction2comp($reactionID, $product);
1375                    for my $compData (@compounds) {
1376                        # Extract the compound data from the current tuple.
1377                        my ($cid, $stoich, $main) = @{$compData};
1378                        # Link the compound to the reaction.
1379                        $loadIsAComponentOf->Put($cid, $reactionID, $discrim++, "", $main,
1380                                                 $product, $stoich);
1381                    }
1382                }
1383            }
1384        }
1385      # Finish the load.      # Finish the load.
1386      my $retVal = $self->_FinishAll();      my $retVal = $self->_FinishAll();
1387      return $retVal;      return $retVal;
# Line 1112  Line 1397 
1397    
1398      GenomeGroups      GenomeGroups
1399    
1400  There is no direct support for genome groups in FIG, so we access the SEED  Currently, we do not use groups. We used to use them for NMPDR groups,
1401    butThere is no direct support for genome groups in FIG, so we access the SEED
1402  files directly.  files directly.
1403    
1404  =over 4  =over 4
# Line 1132  Line 1418 
1418      my $fig = $self->{fig};      my $fig = $self->{fig};
1419      # Get the genome hash.      # Get the genome hash.
1420      my $genomeHash = $self->{genomes};      my $genomeHash = $self->{genomes};
     my $genomeCount = (keys %{$genomeHash});  
1421      # Create a load object for the table we're loading.      # Create a load object for the table we're loading.
1422      my $loadGenomeGroups = $self->_TableLoader('GenomeGroups', $genomeCount * 4);      my $loadGenomeGroups = $self->_TableLoader('GenomeGroups');
1423      Trace("Beginning group data load.") if T(2);      if ($self->{options}->{loadOnly}) {
1424            Trace("Loading from existing files.") if T(2);
1425        } else {
1426            Trace("Generating group data.") if T(2);
1427            # Currently there are no groups.
1428        }
1429        # Finish the load.
1430        my $retVal = $self->_FinishAll();
1431        return $retVal;
1432    }
1433    
1434    =head3 LoadSynonymData
1435    
1436    C<< my $stats = $spl->LoadSynonymData(); >>
1437    
1438    Load the synonym groups into Sprout.
1439    
1440    The following relations are loaded by this method.
1441    
1442        SynonymGroup
1443        IsSynonymGroupFor
1444    
1445    The source information for these relations is taken from the C<maps_to_id> method
1446    of the B<FIG> object. Unfortunately, to make this work, we need to use direct
1447    SQL against the FIG database.
1448    
1449    =over 4
1450    
1451    =item RETURNS
1452    
1453    Returns a statistics object for the loads.
1454    
1455    =back
1456    
1457    =cut
1458    #: Return Type $%;
1459    sub LoadSynonymData {
1460        # Get this object instance.
1461        my ($self) = @_;
1462        # Get the FIG object.
1463        my $fig = $self->{fig};
1464        # Get the genome hash.
1465        my $genomeHash = $self->{genomes};
1466        # Create a load object for the table we're loading.
1467        my $loadSynonymGroup = $self->_TableLoader('SynonymGroup');
1468        my $loadIsSynonymGroupFor = $self->_TableLoader('IsSynonymGroupFor');
1469        if ($self->{options}->{loadOnly}) {
1470            Trace("Loading from existing files.") if T(2);
1471        } else {
1472            Trace("Generating synonym group data.") if T(2);
1473            # Get the database handle.
1474            my $dbh = $fig->db_handle();
1475            # Ask for the synonyms.
1476            my $sth = $dbh->prepare_command("SELECT maps_to, syn_id FROM peg_synonyms ORDER BY maps_to");
1477            my $result = $sth->execute();
1478            if (! defined($result)) {
1479                Confess("Database error in Synonym load: " . $sth->errstr());
1480            } else {
1481                # Remember the current synonym.
1482                my $current_syn = "";
1483                # Count the features.
1484                my $featureCount = 0;
1485                # Loop through the synonym/peg pairs.
1486                while (my @row = $sth->fetchrow()) {
1487                    # Get the synonym ID and feature ID.
1488                    my ($syn_id, $peg) = @row;
1489                    # Insure it's for one of our genomes.
1490                    my $genomeID = FIG::genome_of($peg);
1491                    if (exists $genomeHash->{$genomeID}) {
1492                        # Verify the synonym.
1493                        if ($syn_id ne $current_syn) {
1494                            # It's new, so put it in the group table.
1495                            $loadSynonymGroup->Put($syn_id);
1496                            $current_syn = $syn_id;
1497                        }
1498                        # Connect the synonym to the peg.
1499                        $loadIsSynonymGroupFor->Put($syn_id, $peg);
1500                        # Count this feature.
1501                        $featureCount++;
1502                        if ($featureCount % 1000 == 0) {
1503                            Trace("$featureCount features processed.") if T(3);
1504                        }
1505                    }
1506                }
1507            }
1508        }
1509        # Finish the load.
1510        my $retVal = $self->_FinishAll();
1511        return $retVal;
1512    }
1513    
1514    =head3 LoadFamilyData
1515    
1516    C<< my $stats = $spl->LoadFamilyData(); >>
1517    
1518    Load the protein families into Sprout.
1519    
1520    The following relations are loaded by this method.
1521    
1522        Family
1523        IsFamilyForFeature
1524    
1525    The source information for these relations is taken from the C<families_for_protein>,
1526    C<family_function>, and C<sz_family> methods of the B<FIG> object.
1527    
1528    =over 4
1529    
1530    =item RETURNS
1531    
1532    Returns a statistics object for the loads.
1533    
1534    =back
1535    
1536    =cut
1537    #: Return Type $%;
1538    sub LoadFamilyData {
1539        # Get this object instance.
1540        my ($self) = @_;
1541        # Get the FIG object.
1542        my $fig = $self->{fig};
1543        # Get the genome hash.
1544        my $genomeHash = $self->{genomes};
1545        # Create load objects for the tables we're loading.
1546        my $loadFamily = $self->_TableLoader('Family');
1547        my $loadIsFamilyForFeature = $self->_TableLoader('IsFamilyForFeature');
1548        if ($self->{options}->{loadOnly}) {
1549            Trace("Loading from existing files.") if T(2);
1550        } else {
1551            Trace("Generating family data.") if T(2);
1552            # Create a hash for the family IDs.
1553            my %familyHash = ();
1554      # Loop through the genomes.      # Loop through the genomes.
1555      my $line;          for my $genomeID (sort keys %{$genomeHash}) {
1556      for my $genomeID (keys %{$genomeHash}) {              Trace("Processing features for $genomeID.") if T(2);
1557          Trace("Processing $genomeID.") if T(3);              # Loop through this genome's PEGs.
1558          # Open the NMPDR group file for this genome.              for my $fid ($fig->all_features($genomeID, "peg")) {
1559          if (open(TMP, "<$FIG_Config::organisms/$genomeID/NMPDR") &&                  $loadIsFamilyForFeature->Add("features", 1);
1560              defined($line = <TMP>)) {                  # Get this feature's families.
1561              # Clean the line ending.                  my @families = $fig->families_for_protein($fid);
1562              chomp $line;                  # Loop through the families, connecting them to the feature.
1563              # Add the group to the table. Note that there can only be one group                  for my $family (@families) {
1564              # per genome.                      $loadIsFamilyForFeature->Put($family, $fid);
1565              $loadGenomeGroups->Put($genomeID, $line);                      # If this is a new family, create a record for it.
1566                        if (! exists $familyHash{$family}) {
1567                            $familyHash{$family} = 1;
1568                            $loadFamily->Add("families", 1);
1569                            my $size = $fig->sz_family($family);
1570                            my $func = $fig->family_function($family);
1571                            $loadFamily->Put($family, $size, $func);
1572                        }
1573                    }
1574                }
1575          }          }
         close TMP;  
1576      }      }
1577      # Finish the load.      # Finish the load.
1578      my $retVal = $self->_FinishAll();      my $retVal = $self->_FinishAll();
1579      return $retVal;      return $retVal;
1580  }  }
1581    
1582    
1583    
1584  =head2 Internal Utility Methods  =head2 Internal Utility Methods
1585    
1586  =head3 TableLoader  =head3 TableLoader
# Line 1172  Line 1597 
1597    
1598  Name of the table (relation) being loaded.  Name of the table (relation) being loaded.
1599    
1600  =item rowCount (optional)  =item ignore
1601    
1602  Estimated maximum number of rows in the table.  TRUE if the table should be ignored entirely, else FALSE.
1603    
1604  =item RETURN  =item RETURN
1605    
# Line 1186  Line 1611 
1611    
1612  sub _TableLoader {  sub _TableLoader {
1613      # Get the parameters.      # Get the parameters.
1614      my ($self, $tableName, $rowCount) = @_;      my ($self, $tableName, $ignore) = @_;
1615      # Create the load object.      # Create the load object.
1616      my $retVal = ERDBLoad->new($self->{erdb}, $tableName, $self->{loadDirectory}, $rowCount);      my $retVal = ERDBLoad->new($self->{erdb}, $tableName, $self->{loadDirectory}, $self->LoadOnly,
1617                                   $ignore);
1618      # Cache it in the loader list.      # Cache it in the loader list.
1619      push @{$self->{loaders}}, $retVal;      push @{$self->{loaders}}, $retVal;
1620      # Return it to the caller.      # Return it to the caller.
# Line 1222  Line 1648 
1648      my $retVal = Stats->new();      my $retVal = Stats->new();
1649      # Get the loader list.      # Get the loader list.
1650      my $loadList = $self->{loaders};      my $loadList = $self->{loaders};
1651        # Create a hash to hold the statistics objects, keyed on relation name.
1652        my %loaderHash = ();
1653      # 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
1654      # 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.
1655      while (my $loader = pop @{$loadList}) {      while (my $loader = pop @{$loadList}) {
1656            # Get the relation name.
1657            my $relName = $loader->RelName;
1658            # Check the ignore flag.
1659            if ($loader->Ignore) {
1660                Trace("Relation $relName not loaded.") if T(2);
1661            } else {
1662                # Here we really need to finish.
1663                Trace("Finishing $relName.") if T(2);
1664          my $stats = $loader->Finish();          my $stats = $loader->Finish();
1665                $loaderHash{$relName} = $stats;
1666            }
1667        }
1668        # Now we loop through again, actually loading the tables. We want to finish before
1669        # loading so that if something goes wrong at this point, all the load files are usable
1670        # and we don't have to redo all that work.
1671        for my $relName (sort keys %loaderHash) {
1672            # Get the statistics for this relation.
1673            my $stats = $loaderHash{$relName};
1674            # Check for a database load.
1675            if ($self->{options}->{dbLoad}) {
1676                # Here we want to use the load file just created to load the database.
1677                Trace("Loading relation $relName.") if T(2);
1678                my $newStats = $self->{sprout}->LoadUpdate(1, [$relName]);
1679                # Accumulate the statistics from the DB load.
1680                $stats->Accumulate($newStats);
1681            }
1682          $retVal->Accumulate($stats);          $retVal->Accumulate($stats);
         my $relName = $loader->RelName;  
1683          Trace("Statistics for $relName:\n" . $stats->Show()) if T(2);          Trace("Statistics for $relName:\n" . $stats->Show()) if T(2);
1684      }      }
1685      # Return the load statistics.      # Return the load statistics.

Legend:
Removed from v.1.7  
changed lines
  Added in v.1.70

MCS Webmaster
ViewVC Help
Powered by ViewVC 1.0.3