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

Diff of /Sprout/SproutLoad.pm

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

revision 1.18, Wed Oct 12 03:17:58 2005 UTC revision 1.74, Sat Oct 21 06:44:17 2006 UTC
# Line 30  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 80  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  =item options
86    
# Line 94  Line 93 
93  sub new {  sub new {
94      # Get the parameters.      # Get the parameters.
95      my ($class, $sprout, $fig, $genomeFile, $subsysFile, $options) = @_;      my ($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 119  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} = $accessCode;
# Line 129  Line 130 
130              Confess("Invalid genome parameter ($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 153  Line 163 
163              Confess("Invalid subsystem parameter in SproutLoad constructor.");              Confess("Invalid subsystem parameter in SproutLoad constructor.");
164          }          }
165      }      }
166            # Go through the subsys hash again, creating the keyword list for each subsystem.
167            for my $subsystem (keys %subsystems) {
168                my $name = $subsystem;
169                $name =~ s/_/ /g;
170                my $classes = $fig->subsystem_classification($subsystem);
171                my @classList = map { " $_" } @{$classes};
172                $name .= join("", @classList);
173                $subsystems{$subsystem} = $name;
174            }
175        }
176      # Get the data directory from the Sprout object.      # Get the data directory from the Sprout object.
177      my ($directory) = $sprout->LoadInfo();      my ($directory) = $sprout->LoadInfo();
178      # Create the Sprout load object.      # Create the Sprout load object.
# Line 162  Line 182 
182                    subsystems => \%subsystems,                    subsystems => \%subsystems,
183                    sprout => $sprout,                    sprout => $sprout,
184                    loadDirectory => $directory,                    loadDirectory => $directory,
185                    erdb => $sprout->{_erdb},                    erdb => $sprout,
186                    loaders => [],                    loaders => [],
187                    options => $options                    options => $options
188                   };                   };
# Line 171  Line 191 
191      return $retVal;      return $retVal;
192  }  }
193    
194    =head3 LoadOnly
195    
196    C<< my $flag = $spl->LoadOnly; >>
197    
198    Return TRUE if we are in load-only mode, else FALSE.
199    
200    =cut
201    
202    sub LoadOnly {
203        my ($self) = @_;
204        return $self->{options}->{loadOnly};
205    }
206    
207    =head3 PrimaryOnly
208    
209    C<< my $flag = $spl->PrimaryOnly; >>
210    
211    Return TRUE if only the main entity is to be loaded, else FALSE.
212    
213    =cut
214    
215    sub PrimaryOnly {
216        my ($self) = @_;
217        return $self->{options}->{primaryOnly};
218    }
219    
220  =head3 LoadGenomeData  =head3 LoadGenomeData
221    
222  C<< my $stats = $spl->LoadGenomeData(); >>  C<< my $stats = $spl->LoadGenomeData(); >>
# Line 198  Line 244 
244    
245  =back  =back
246    
 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.)  
   
247  =cut  =cut
248  #: Return Type $%;  #: Return Type $%;
249  sub LoadGenomeData {  sub LoadGenomeData {
# Line 216  Line 254 
254      # Get the genome count.      # Get the genome count.
255      my $genomeHash = $self->{genomes};      my $genomeHash = $self->{genomes};
256      my $genomeCount = (keys %{$genomeHash});      my $genomeCount = (keys %{$genomeHash});
     Trace("Beginning genome data load.") if T(2);  
257      # Create load objects for each of the tables we're loading.      # Create load objects for each of the tables we're loading.
258      my $loadGenome = $self->_TableLoader('Genome', $genomeCount);      my $loadGenome = $self->_TableLoader('Genome');
259      my $loadHasContig = $self->_TableLoader('HasContig', $genomeCount * 300);      my $loadHasContig = $self->_TableLoader('HasContig', $self->PrimaryOnly);
260      my $loadContig = $self->_TableLoader('Contig', $genomeCount * 300);      my $loadContig = $self->_TableLoader('Contig', $self->PrimaryOnly);
261      my $loadIsMadeUpOf = $self->_TableLoader('IsMadeUpOf', $genomeCount * 60000);      my $loadIsMadeUpOf = $self->_TableLoader('IsMadeUpOf', $self->PrimaryOnly);
262      my $loadSequence = $self->_TableLoader('Sequence', $genomeCount * 60000);      my $loadSequence = $self->_TableLoader('Sequence', $self->PrimaryOnly);
263        if ($self->{options}->{loadOnly}) {
264            Trace("Loading from existing files.") if T(2);
265        } else {
266            Trace("Generating genome data.") if T(2);
267      # Now we loop through the genomes, generating the data for each one.      # Now we loop through the genomes, generating the data for each one.
268      for my $genomeID (sort keys %{$genomeHash}) {      for my $genomeID (sort keys %{$genomeHash}) {
269          Trace("Loading data for genome $genomeID.") if T(3);              Trace("Generating data for genome $genomeID.") if T(3);
270          $loadGenome->Add("genomeIn");          $loadGenome->Add("genomeIn");
271          # The access code comes in via the genome hash.          # The access code comes in via the genome hash.
272          my $accessCode = $genomeHash->{$genomeID};          my $accessCode = $genomeHash->{$genomeID};
273          # 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.  
274          my ($genus, $species, @extraData) = split / /, $self->{fig}->genus_species($genomeID);          my ($genus, $species, @extraData) = split / /, $self->{fig}->genus_species($genomeID);
275          my $extra = join " ", @extraData, "[$genomeID]";              my $extra = join " ", @extraData;
276          # Get the full taxonomy.          # Get the full taxonomy.
277          my $taxonomy = $fig->taxonomy_of($genomeID);          my $taxonomy = $fig->taxonomy_of($genomeID);
278                # Open the NMPDR group file for this genome.
279                my $group;
280                if (open(TMP, "<$FIG_Config::organisms/$genomeID/NMPDR") &&
281                    defined($group = <TMP>)) {
282                    # Clean the line ending.
283                    chomp $group;
284                } else {
285                    # No group, so use the default.
286                    $group = $FIG_Config::otherGroup;
287                }
288                close TMP;
289          # Output the genome record.          # Output the genome record.
290          $loadGenome->Put($genomeID, $accessCode, $fig->is_complete($genomeID), $genus,          $loadGenome->Put($genomeID, $accessCode, $fig->is_complete($genomeID), $genus,
291                           $species, $extra, $taxonomy);                               $group, $species, $extra, $taxonomy);
292          # Now we loop through each of the genome's contigs.          # Now we loop through each of the genome's contigs.
293          my @contigs = $fig->all_contigs($genomeID);          my @contigs = $fig->all_contigs($genomeID);
294          for my $contigID (@contigs) {          for my $contigID (@contigs) {
# Line 268  Line 319 
319              }              }
320          }          }
321      }      }
322        }
323      # Finish the loads.      # Finish the loads.
324      my $retVal = $self->_FinishAll();      my $retVal = $self->_FinishAll();
325      # Return the result.      # Return the result.
# Line 308  Line 360 
360      my $fig = $self->{fig};      my $fig = $self->{fig};
361      # Get the genome hash.      # Get the genome hash.
362      my $genomeFilter = $self->{genomes};      my $genomeFilter = $self->{genomes};
363      my $genomeCount = (keys %{$genomeFilter});      # Set up an ID counter for the PCHs.
364      my $featureCount = $genomeCount * 4000;      my $pchID = 0;
365      # Start the loads.      # Start the loads.
366      my $loadCoupling = $self->_TableLoader('Coupling', $featureCount * $genomeCount);      my $loadCoupling = $self->_TableLoader('Coupling');
367      my $loadIsEvidencedBy = $self->_TableLoader('IsEvidencedBy', $featureCount * 8000);      my $loadIsEvidencedBy = $self->_TableLoader('IsEvidencedBy', $self->PrimaryOnly);
368      my $loadPCH = $self->_TableLoader('PCH', $featureCount * 2000);      my $loadPCH = $self->_TableLoader('PCH', $self->PrimaryOnly);
369      my $loadParticipatesInCoupling = $self->_TableLoader('ParticipatesInCoupling', $featureCount * 2000);      my $loadParticipatesInCoupling = $self->_TableLoader('ParticipatesInCoupling', $self->PrimaryOnly);
370      my $loadUsesAsEvidence = $self->_TableLoader('UsesAsEvidence', $featureCount * 8000);      my $loadUsesAsEvidence = $self->_TableLoader('UsesAsEvidence', $self->PrimaryOnly);
371      Trace("Beginning coupling data load.") if T(2);      if ($self->{options}->{loadOnly}) {
372            Trace("Loading from existing files.") if T(2);
373        } else {
374            Trace("Generating coupling data.") if T(2);
375      # Loop through the genomes found.      # Loop through the genomes found.
376      for my $genome (sort keys %{$genomeFilter}) {      for my $genome (sort keys %{$genomeFilter}) {
377          Trace("Generating coupling data for $genome.") if T(3);          Trace("Generating coupling data for $genome.") if T(3);
# Line 340  Line 395 
395              for my $coupleData (@couplings) {              for my $coupleData (@couplings) {
396                  my ($peg2, $score) = @{$coupleData};                  my ($peg2, $score) = @{$coupleData};
397                  # Compute the coupling ID.                  # Compute the coupling ID.
398                  my $coupleID = Sprout::CouplingID($peg1, $peg2);                      my $coupleID = $self->{erdb}->CouplingID($peg1, $peg2);
399                  if (! exists $dupHash{$coupleID}) {                  if (! exists $dupHash{$coupleID}) {
400                      $loadCoupling->Add("couplingIn");                      $loadCoupling->Add("couplingIn");
401                      # 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 376  Line 431 
431                          }                          }
432                      }                      }
433                      for my $evidenceID (keys %evidenceMap) {                      for my $evidenceID (keys %evidenceMap) {
434                                # Get the ID for this evidence.
435                                $pchID++;
436                          # Create the evidence record.                          # Create the evidence record.
437                          my ($peg3, $peg4, $usage) = @{$evidenceMap{$evidenceID}};                          my ($peg3, $peg4, $usage) = @{$evidenceMap{$evidenceID}};
438                          $loadPCH->Put($evidenceID, $usage);                              $loadPCH->Put($pchID, $usage);
439                          # Connect it to the coupling.                          # Connect it to the coupling.
440                          $loadIsEvidencedBy->Put($coupleID, $evidenceID);                              $loadIsEvidencedBy->Put($coupleID, $pchID);
441                          # Connect it to the features.                          # Connect it to the features.
442                          $loadUsesAsEvidence->Put($evidenceID, $peg3, 1);                              $loadUsesAsEvidence->Put($pchID, $peg3, 1);
443                          $loadUsesAsEvidence->Put($evidenceID, $peg4, 2);                              $loadUsesAsEvidence->Put($pchID, $peg4, 2);
444                            }
445                      }                      }
446                  }                  }
447              }              }
# Line 410  Line 468 
468      FeatureTranslation      FeatureTranslation
469      FeatureUpstream      FeatureUpstream
470      IsLocatedIn      IsLocatedIn
471        HasFeature
472        HasRoleInSubsystem
473    
474  =over 4  =over 4
475    
# Line 424  Line 484 
484  sub LoadFeatureData {  sub LoadFeatureData {
485      # Get this object instance.      # Get this object instance.
486      my ($self) = @_;      my ($self) = @_;
487      # Get the FIG object.      # Get the FIG and Sprout objects.
488      my $fig = $self->{fig};      my $fig = $self->{fig};
489      # Find out if this is a limited run.      my $sprout = $self->{sprout};
     my $limited = $self->{options}->{limitedFeatures};  
490      # Get the table of genome IDs.      # Get the table of genome IDs.
491      my $genomeHash = $self->{genomes};      my $genomeHash = $self->{genomes};
     my $genomeCount = (keys %{$genomeHash});  
     my $featureCount = $genomeCount * 4000;  
492      # Create load objects for each of the tables we're loading.      # Create load objects for each of the tables we're loading.
493      my $loadFeature = $self->_TableLoader('Feature', $featureCount);      my $loadFeature = $self->_TableLoader('Feature');
494      my $loadIsLocatedIn = $self->_TableLoader('IsLocatedIn', $featureCount);      my $loadIsLocatedIn = $self->_TableLoader('IsLocatedIn', $self->PrimaryOnly);
495      my $loadFeatureAlias = $self->_TableLoader('FeatureAlias', $featureCount * 6);      my $loadFeatureAlias = $self->_TableLoader('FeatureAlias');
496      my ($loadFeatureLink, $loadFeatureTranslation, $loadFeatureUpstream);      my $loadFeatureLink = $self->_TableLoader('FeatureLink');
497      if (! $limited) {      my $loadFeatureTranslation = $self->_TableLoader('FeatureTranslation');
498          $loadFeatureLink = $self->_TableLoader('FeatureLink', $featureCount * 10);      my $loadFeatureUpstream = $self->_TableLoader('FeatureUpstream');
499          $loadFeatureTranslation = $self->_TableLoader('FeatureTranslation', $featureCount);      my $loadHasFeature = $self->_TableLoader('HasFeature', $self->PrimaryOnly);
500          $loadFeatureUpstream = $self->_TableLoader('FeatureUpstream', $featureCount);      my $loadHasRoleInSubsystem = $self->_TableLoader('HasRoleInSubsystem', $self->PrimaryOnly);
501      }      # Get the subsystem hash.
502        my $subHash = $self->{subsystems};
503      # 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
504      # locations.      # locations.
505      my $chunkSize = $self->{sprout}->MaxSegment();      my $chunkSize = $self->{sprout}->MaxSegment();
506      Trace("Beginning feature data load.") if T(2);      if ($self->{options}->{loadOnly}) {
507            Trace("Loading from existing files.") if T(2);
508        } else {
509            Trace("Generating feature data.") if T(2);
510      # Now we loop through the genomes, generating the data for each one.      # Now we loop through the genomes, generating the data for each one.
511      for my $genomeID (sort keys %{$genomeHash}) {      for my $genomeID (sort keys %{$genomeHash}) {
512          Trace("Loading features for genome $genomeID.") if T(3);          Trace("Loading features for genome $genomeID.") if T(3);
513          $loadFeature->Add("genomeIn");          $loadFeature->Add("genomeIn");
514          # Get the feature list for this genome.          # Get the feature list for this genome.
515          my $features = $fig->all_features_detailed($genomeID);          my $features = $fig->all_features_detailed($genomeID);
516                # Sort and count the list.
517                my @featureTuples = sort { $a->[0] cmp $b->[0] } @{$features};
518                my $count = scalar @featureTuples;
519                Trace("$count features found for genome $genomeID.") if T(3);
520                # Set up for our duplicate-feature check.
521                my $oldFeatureID = "";
522          # Loop through the features.          # Loop through the features.
523          for my $featureData (@{$features}) {              for my $featureTuple (@featureTuples) {
             $loadFeature->Add("featureIn");  
524              # Split the tuple.              # Split the tuple.
525              my ($featureID, $locations, undef, $type) = @{$featureData};                  my ($featureID, $locations, undef, $type) = @{$featureTuple};
526              # Create the feature record.                  # Check for duplicates.
527              $loadFeature->Put($featureID, 1, $type);                  if ($featureID eq $oldFeatureID) {
528                        Trace("Duplicate feature $featureID found.") if T(1);
529                    } else {
530                        $oldFeatureID = $featureID;
531                        # Count this feature.
532                        $loadFeature->Add("featureIn");
533                        # Get the functional assignment.
534                        my $assignment = $fig->function_of($featureID);
535                        # Begin building the keywords.
536                        my $keywords = "$assignment $genomeID";
537                        # Link this feature to the parent genome.
538                        $loadHasFeature->Put($genomeID, $featureID, $type);
539              # Create the aliases.              # Create the aliases.
540              for my $alias ($fig->feature_aliases($featureID)) {              for my $alias ($fig->feature_aliases($featureID)) {
541                  $loadFeatureAlias->Put($featureID, $alias);                  $loadFeatureAlias->Put($featureID, $alias);
542                            $keywords .= " $alias";
543              }              }
             # The next stuff is for a full load only.  
             if (! $limited) {  
544                  # Get the links.                  # Get the links.
545                  my @links = $fig->fid_links($featureID);                  my @links = $fig->fid_links($featureID);
546                  for my $link (@links) {                  for my $link (@links) {
# Line 483  Line 559 
559                          $loadFeatureUpstream->Put($featureID, $upstream);                          $loadFeatureUpstream->Put($featureID, $upstream);
560                      }                      }
561                  }                  }
562                        # Now we need to find the subsystems this feature participates in.
563                        # We also add the subsystems to the keyword list. Before we do that,
564                        # we must convert underscores to spaces and tack on the classifications.
565                        my @subsystems = $fig->peg_to_subsystems($featureID);
566                        for my $subsystem (@subsystems) {
567                            # Only proceed if we like this subsystem.
568                            if (exists $subHash->{$subsystem}) {
569                                # Store the has-role link.
570                                $loadHasRoleInSubsystem->Put($featureID, $subsystem, $genomeID, $type);
571                                # Save the subsystem's keyword data.
572                                my $subKeywords = $subHash->{$subsystem};
573                                $keywords .= " $subKeywords";
574                            }
575                        }
576                        # The final task is to add virulence and essentiality attributes.
577                        if ($fig->virulent($featureID)) {
578                            $keywords .= " virulent";
579                        }
580                        if ($fig->essential($featureID)) {
581                            $keywords .= " essential";
582              }              }
583                        # Clean the keyword list.
584                        my $cleanWords = $sprout->CleanKeywords($keywords);
585                        # Create the feature record.
586                        $loadFeature->Put($featureID, 1, $type, $assignment, $cleanWords);
587              # 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
588              # 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
589              # the maximum segment size. This simplifies the genes_in_region processing              # the maximum segment size. This simplifies the genes_in_region processing
# Line 512  Line 612 
612              }              }
613          }          }
614      }      }
     # Finish the loads.  
     my $retVal = $self->_FinishAll();  
     return $retVal;  
 }  
   
 =head3 LoadBBHData  
   
 C<< my $stats = $spl->LoadBBHData(); >>  
   
 Load the bidirectional best hit data from FIG into Sprout.  
   
 Sprout does not store information on similarities. Instead, it has only the  
 bi-directional best hits. Even so, the BBH table is one of the largest in  
 the database.  
   
 The following relations are loaded by this method.  
   
     IsBidirectionalBestHitOf  
   
 =over 4  
   
 =item RETURNS  
   
 Returns a statistics object for the loads.  
   
 =back  
   
 =cut  
 #: Return Type $%;  
 sub LoadBBHData {  
     # Get this object instance.  
     my ($self) = @_;  
     # Get the FIG object.  
     my $fig = $self->{fig};  
     # Get the table of genome IDs.  
     my $genomeHash = $self->{genomes};  
     my $genomeCount = (keys %{$genomeHash});  
     my $featureCount = $genomeCount * 4000;  
     # Create load objects for each of the tables we're loading.  
     my $loadIsBidirectionalBestHitOf = $self->_TableLoader('IsBidirectionalBestHitOf',  
                                                            $featureCount * $genomeCount);  
     Trace("Beginning BBH load.") if T(2);  
     # Now we loop through the genomes, generating the data for each one.  
     for my $genomeID (sort keys %{$genomeHash}) {  
         $loadIsBidirectionalBestHitOf->Add("genomeIn");  
         Trace("Processing features for genome $genomeID.") if T(3);  
         # Get the feature list for this genome.  
         my $features = $fig->all_features_detailed($genomeID);  
         # Loop through the features.  
         for my $featureData (@{$features}) {  
             # Split the tuple.  
             my ($featureID, $locations, $aliases, $type) = @{$featureData};  
             # Get the bi-directional best hits.  
             my @bbhList = $fig->bbhs($featureID);  
             for my $bbhEntry (@bbhList) {  
                 # Get the target feature ID and the score.  
                 my ($targetID, $score) = @{$bbhEntry};  
                 # Check the target feature's genome.  
                 my $targetGenomeID = $fig->genome_of($targetID);  
                 # Only proceed if it's one of our genomes.  
                 if ($genomeHash->{$targetGenomeID}) {  
                     $loadIsBidirectionalBestHitOf->Put($featureID, $targetID, $targetGenomeID,  
                                                        $score);  
                 }  
             }  
615          }          }
616      }      }
617      # Finish the loads.      # Finish the loads.
# Line 599  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 607  Line 644 
644      OccursInSubsystem      OccursInSubsystem
645      ParticipatesIn      ParticipatesIn
646      HasSSCell      HasSSCell
     Catalyzes  
     Reaction  
647      ConsistsOfRoles      ConsistsOfRoles
648      RoleSubset      RoleSubset
649      HasRoleSubset      HasRoleSubset
650      ConsistsOfGenomes      ConsistsOfGenomes
651      GenomeSubset      GenomeSubset
652      HasGenomeSubset      HasGenomeSubset
653        Catalyzes
654        Diagram
655        RoleOccursIn
656    
657  =over 4  =over 4
658    
# Line 624  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 641  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      my $loadReaction = $self->_TableLoader('Reaction', $featureCount * $genomeCount);      my $loadIsRoleOf = $self->_TableLoader('IsRoleOf', $self->PrimaryOnly);
691      my $loadCatalyzes = $self->_TableLoader('Catalyzes', $featureCount * $genomeCount);      my $loadOccursInSubsystem = $self->_TableLoader('OccursInSubsystem', $self->PrimaryOnly);
692      my $loadRoleSubset = $self->_TableLoader('RoleSubset', $subsysCount * 50);      my $loadParticipatesIn = $self->_TableLoader('ParticipatesIn', $self->PrimaryOnly);
693      my $loadGenomeSubset = $self->_TableLoader('GenomeSubset', $subsysCount * 50);      my $loadHasSSCell = $self->_TableLoader('HasSSCell', $self->PrimaryOnly);
694      my $loadConsistsOfRoles = $self->_TableLoader('ConsistsOfRoles', $featureCount * $genomeCount);      my $loadRoleSubset = $self->_TableLoader('RoleSubset', $self->PrimaryOnly);
695      my $loadConsistsOfGenomes = $self->_TableLoader('ConsistsOfGenomes', $featureCount * $genomeCount);      my $loadGenomeSubset = $self->_TableLoader('GenomeSubset', $self->PrimaryOnly);
696      my $loadHasRoleSubset = $self->_TableLoader('HasRoleSubset', $subsysCount * 50);      my $loadConsistsOfRoles = $self->_TableLoader('ConsistsOfRoles', $self->PrimaryOnly);
697      my $loadHasGenomeSubset = $self->_TableLoader('HasGenomeSubset', $subsysCount * 50);      my $loadConsistsOfGenomes = $self->_TableLoader('ConsistsOfGenomes', $self->PrimaryOnly);
698      Trace("Beginning subsystem data load.") if T(2);      my $loadHasRoleSubset = $self->_TableLoader('HasRoleSubset', $self->PrimaryOnly);
699      # The reaction hash will contain a list of reactions for each role. When we're done,      my $loadHasGenomeSubset = $self->_TableLoader('HasGenomeSubset', $self->PrimaryOnly);
700      # a complicated sort and merge will be used to generate the Reaction and Catalyzes      my $loadSubsystemClass = $self->_TableLoader('SubsystemClass', $self->PrimaryOnly);
701      # tables.      if ($self->{options}->{loadOnly}) {
702      my %reactionsToRoles = ();          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
# Line 675  Line 713 
713      my ($genomeID, $roleID);      my ($genomeID, $roleID);
714      my %roleData = ();      my %roleData = ();
715      for my $subsysID (@subsysIDs) {      for my $subsysID (@subsysIDs) {
         Trace("Creating subsystem $subsysID.") if T(3);  
         $loadSubsystem->Add("subsystemIn");  
716          # Get the subsystem object.          # Get the subsystem object.
717          my $sub = $fig->get_subsystem($subsysID);          my $sub = $fig->get_subsystem($subsysID);
718          # Get its reaction hash.              # Only proceed if the subsystem has a spreadsheet.
719          my $reactionHash = $sub->get_reactions();              if (! $sub->{empty_ss}) {
720                    Trace("Creating subsystem $subsysID.") if T(3);
721                    $loadSubsystem->Add("subsystemIn");
722          # Create the subsystem record.          # Create the subsystem record.
723          my $curator = $sub->get_curator();          my $curator = $sub->get_curator();
724          my $notes = $sub->get_notes();          my $notes = $sub->get_notes();
725          $loadSubsystem->Put($subsysID, $curator, $notes);          $loadSubsystem->Put($subsysID, $curator, $notes);
726                    # Now for the classification string. This comes back as a list
727                    # reference and we convert it to a space-delimited string.
728                    my $classList = $fig->subsystem_classification($subsysID);
729                    my $classString = join(" : ", grep { $_ } @$classList);
730                    $loadSubsystemClass->Put($subsysID, $classString);
731          # Connect it to its roles. Each role is a column in the subsystem spreadsheet.          # Connect it to its roles. Each role is a column in the subsystem spreadsheet.
732          for (my $col = 0; defined($roleID = $sub->get_role($col)); $col++) {          for (my $col = 0; defined($roleID = $sub->get_role($col)); $col++) {
733              # Connect to this role.              # Connect to this role.
# Line 697  Line 740 
740                  # Add the role.                  # Add the role.
741                  $loadRole->Put($roleID, $abbr);                  $loadRole->Put($roleID, $abbr);
742                  $roleData{$roleID} = 1;                  $roleData{$roleID} = 1;
743                  # Add the role's reactions.                          # Check for an EC number.
744                  my $reactions = $reactionHash->{$roleID};                          if ($roleID =~ /\(EC ([^.]+\.[^.]+\.[^.]+\.[^)]+)\)\s*$/) {
745                  for my $reactionID (@{$reactions}) {                              my $ec = $1;
746                      if (! exists $reactionsToRoles{$reactionID}) {                              $loadRoleEC->Put($roleID, $ec);
747                          # Here the reaction is brand-new, so we create its reaction                              $ecToRoles{$ec} = $roleID;
                         # record.  
                         $loadReaction->Put($reactionID, $fig->reversible($reactionID));  
                         # We also create a blank list for it in the reaction hash.  
                         $reactionsToRoles{$reactionID} = [];  
                     }  
                     # Add the role to the reaction's role list.  
                     push @{$reactionsToRoles{$reactionID}}, $roleID;  
748                  }                  }
749              }              }
750          }          }
# Line 735  Line 771 
771                  # part of the spreadsheet cell ID.                  # part of the spreadsheet cell ID.
772                  for (my $col = 0; defined($roleID = $sub->get_role($col)); $col++) {                  for (my $col = 0; defined($roleID = $sub->get_role($col)); $col++) {
773                      # Get the features in the spreadsheet cell for this genome and role.                      # Get the features in the spreadsheet cell for this genome and role.
774                      my @pegs = $sub->get_pegs_from_cell($row, $col);                              my @pegs = grep { !$fig->is_deleted_fid($_) } $sub->get_pegs_from_cell($row, $col);
775                      # Only proceed if features exist.                      # Only proceed if features exist.
776                      if (@pegs > 0) {                      if (@pegs > 0) {
777                          # Create the spreadsheet cell.                          # Create the spreadsheet cell.
# Line 756  Line 792 
792                  if ($pegCount > 0) {                  if ($pegCount > 0) {
793                      Trace("$pegCount PEGs in $cellCount cells for $genomeID.") if T(3);                      Trace("$pegCount PEGs in $cellCount cells for $genomeID.") if T(3);
794                      $loadParticipatesIn->Put($genomeID, $subsysID, $variantCode);                      $loadParticipatesIn->Put($genomeID, $subsysID, $variantCode);
                     # Partition the PEGs found into clusters.  
                     my @clusters = $fig->compute_clusters(\@pegsFound, $sub);  
795                      # Create a hash mapping PEG IDs to cluster numbers.                      # Create a hash mapping PEG IDs to cluster numbers.
796                      # We default to -1 for all of them.                      # We default to -1 for all of them.
797                      my %clusterOf = map { $_ => -1 } @pegsFound;                      my %clusterOf = map { $_ => -1 } @pegsFound;
798                                # Partition the PEGs found into clusters.
799                                my @clusters = $fig->compute_clusters([keys %clusterOf], $sub);
800                      for (my $i = 0; $i <= $#clusters; $i++) {                      for (my $i = 0; $i <= $#clusters; $i++) {
801                          my $subList = $clusters[$i];                          my $subList = $clusters[$i];
802                          for my $peg (@{$subList}) {                          for my $peg (@{$subList}) {
# Line 788  Line 824 
824              # Connect the subset to the subsystem.              # Connect the subset to the subsystem.
825              $loadHasRoleSubset->Put($subsysID, $actualID);              $loadHasRoleSubset->Put($subsysID, $actualID);
826              # Connect the subset to its roles.              # Connect the subset to its roles.
827              my @roles = $sub->get_subset($subsetID);                      my @roles = $sub->get_subsetC_roles($subsetID);
828              for my $roleID (@roles) {              for my $roleID (@roles) {
829                  $loadConsistsOfRoles->Put($actualID, $roleID);                  $loadConsistsOfRoles->Put($actualID, $roleID);
830              }              }
# Line 808  Line 844 
844              }              }
845          }          }
846      }      }
     # Before we leave, we must create the Catalyzes table. The data is all stored in  
     # "reactionToRoles" hash.  
     for my $reactionID (keys %reactionsToRoles) {  
         # Get this reaction's list of roles. We sort it so we can merge out duplicates.  
         my @roles = sort @{$reactionsToRoles{$reactionID}};  
         my $lastRole = "";  
         # Loop through the roles, creating catalyzation records.  
         for my $thisRole (@roles) {  
             if ($thisRole ne $lastRole) {  
                 $loadCatalyzes->Put($thisRole, $reactionID);  
             }  
847          }          }
848      }          # Now we loop through the diagrams. We need to create the diagram records
849      # Finish the load.          # and link each diagram to its roles. Note that only roles which occur
850      my $retVal = $self->_FinishAll();          # in subsystems (and therefore appear in the %ecToRoles hash) are
851      return $retVal;          # included.
852  }          for my $map (@maps) {
   
 =head3 LoadDiagramData  
   
 C<< my $stats = $spl->LoadDiagramData(); >>  
   
 Load the diagram data from FIG into Sprout.  
   
 Diagrams are used to organize functional roles. The diagram shows the  
 connections between chemicals that interact with a subsystem.  
   
 The following relations are loaded by this method.  
   
     Diagram  
     RoleOccursIn  
   
 =over 4  
   
 =item RETURNS  
   
 Returns a statistics object for the loads.  
   
 =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) {  
853          Trace("Loading diagram $map.") if T(3);          Trace("Loading diagram $map.") if T(3);
854          # Get the diagram's descriptive name.          # Get the diagram's descriptive name.
855          my $name = $fig->map_name($map);          my $name = $fig->map_name($map);
# Line 874  Line 858 
858          # A hash is used to prevent duplicates.          # A hash is used to prevent duplicates.
859          my %roleHash = ();          my %roleHash = ();
860          for my $role ($fig->map_to_ecs($map)) {          for my $role ($fig->map_to_ecs($map)) {
861              if (! $roleHash{$role}) {                  if (exists $ecToRoles{$role} && ! $roleHash{$role}) {
862                  $loadRoleOccursIn->Put($role, $map);                      $loadRoleOccursIn->Put($ecToRoles{$role}, $map);
863                  $roleHash{$role} = 1;                  $roleHash{$role} = 1;
864              }              }
865          }          }
866      }      }
867            # Before we leave, we must create the Catalyzes table. We start with the reactions,
868            # then use the "ecToRoles" table to convert EC numbers to role IDs.
869            my @reactions = $fig->all_reactions();
870            for my $reactionID (@reactions) {
871                # Get this reaction's list of roles. The results will be EC numbers.
872                my @roles = $fig->catalyzed_by($reactionID);
873                # Loop through the roles, creating catalyzation records.
874                for my $thisRole (@roles) {
875                    if (exists $ecToRoles{$thisRole}) {
876                        $loadCatalyzes->Put($ecToRoles{$thisRole}, $reactionID);
877                    }
878                }
879            }
880        }
881      # Finish the load.      # Finish the load.
882      my $retVal = $self->_FinishAll();      my $retVal = $self->_FinishAll();
883      return $retVal;      return $retVal;
# Line 921  Line 919 
919      my $fig = $self->{fig};      my $fig = $self->{fig};
920      # Get the genome hash.      # Get the genome hash.
921      my $genomeHash = $self->{genomes};      my $genomeHash = $self->{genomes};
     my $genomeCount = (keys %{$genomeHash});  
922      # Create load objects for each of the tables we're loading.      # Create load objects for each of the tables we're loading.
923      my $loadProperty = $self->_TableLoader('Property', $genomeCount * 1500);      my $loadProperty = $self->_TableLoader('Property');
924      my $loadHasProperty = $self->_TableLoader('HasProperty', $genomeCount * 1500);      my $loadHasProperty = $self->_TableLoader('HasProperty', $self->PrimaryOnly);
925      Trace("Beginning property data load.") if T(2);      if ($self->{options}->{loadOnly}) {
926            Trace("Loading from existing files.") if T(2);
927        } else {
928            Trace("Generating property data.") if T(2);
929      # Create a hash for storing property IDs.      # Create a hash for storing property IDs.
930      my %propertyKeys = ();      my %propertyKeys = ();
931      my $nextID = 1;      my $nextID = 1;
932      # Loop through the genomes.      # Loop through the genomes.
933      for my $genomeID (keys %{$genomeHash}) {          for my $genomeID (sort keys %{$genomeHash}) {
934          $loadProperty->Add("genomeIn");          $loadProperty->Add("genomeIn");
935                Trace("Generating properties for $genomeID.") if T(3);
936          # 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
937          # tuples returned by "all_features_detailed". We use "all_features_detailed"          # tuples returned by "all_features_detailed". We use "all_features_detailed"
938          # rather than "all_features" because we want all features regardless of type.          # rather than "all_features" because we want all features regardless of type.
939          my @features = map { $_->[0] } @{$fig->all_features_detailed($genomeID)};          my @features = map { $_->[0] } @{$fig->all_features_detailed($genomeID)};
940                my $featureCount = 0;
941                my $propertyCount = 0;
942          # Loop through the features, creating HasProperty records.          # Loop through the features, creating HasProperty records.
943          for my $fid (@features) {          for my $fid (@features) {
             $loadProperty->Add("featureIn");  
944              # 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
945              # to insure we do not get any genome attributes.              # to insure we do not get any genome attributes.
946              my @attributeList = $fig->get_attributes($fid, '', '', '');              my @attributeList = $fig->get_attributes($fid, '', '', '');
947                    # Add essentiality and virulence attributes.
948                    if ($fig->essential($fid)) {
949                        push @attributeList, [$fid, 'essential', 1, ''];
950                    }
951                    if ($fig->virulent($fid)) {
952                        push @attributeList, [$fid, 'virulent', 1, ''];
953                    }
954                    if (scalar @attributeList) {
955                        $featureCount++;
956                    }
957              # Loop through the attributes.              # Loop through the attributes.
958              for my $tuple (@attributeList) {              for my $tuple (@attributeList) {
959                        $propertyCount++;
960                  # Get this attribute value's data. Note that we throw away the FID,                  # Get this attribute value's data. Note that we throw away the FID,
961                  # since it will always be the same as the value if "$fid".                  # since it will always be the same as the value if "$fid".
962                  my (undef, $key, $value, $url) = @{$tuple};                  my (undef, $key, $value, $url) = @{$tuple};
# Line 965  Line 978 
978                  $loadHasProperty->Put($fid, $propertyID, $url);                  $loadHasProperty->Put($fid, $propertyID, $url);
979              }              }
980          }          }
981                # Update the statistics.
982                Trace("$propertyCount attributes processed for $featureCount features.") if T(3);
983                $loadHasProperty->Add("featuresIn", $featureCount);
984                $loadHasProperty->Add("propertiesIn", $propertyCount);
985            }
986      }      }
987      # Finish the load.      # Finish the load.
988      my $retVal = $self->_FinishAll();      my $retVal = $self->_FinishAll();
# Line 1005  Line 1023 
1023      my $fig = $self->{fig};      my $fig = $self->{fig};
1024      # Get the genome hash.      # Get the genome hash.
1025      my $genomeHash = $self->{genomes};      my $genomeHash = $self->{genomes};
     my $genomeCount = (keys %{$genomeHash});  
1026      # Create load objects for each of the tables we're loading.      # Create load objects for each of the tables we're loading.
1027      my $loadAnnotation = $self->_TableLoader('Annotation', $genomeCount * 4000);      my $loadAnnotation = $self->_TableLoader('Annotation');
1028      my $loadIsTargetOfAnnotation = $self->_TableLoader('IsTargetOfAnnotation', $genomeCount * 4000);      my $loadIsTargetOfAnnotation = $self->_TableLoader('IsTargetOfAnnotation', $self->PrimaryOnly);
1029      my $loadSproutUser = $self->_TableLoader('SproutUser', 100);      my $loadSproutUser = $self->_TableLoader('SproutUser', $self->PrimaryOnly);
1030      my $loadUserAccess = $self->_TableLoader('UserAccess', 1000);      my $loadUserAccess = $self->_TableLoader('UserAccess', $self->PrimaryOnly);
1031      my $loadMadeAnnotation = $self->_TableLoader('MadeAnnotation', $genomeCount * 4000);      my $loadMadeAnnotation = $self->_TableLoader('MadeAnnotation', $self->PrimaryOnly);
1032      Trace("Beginning annotation data load.") if T(2);      if ($self->{options}->{loadOnly}) {
1033            Trace("Loading from existing files.") if T(2);
1034        } else {
1035            Trace("Generating annotation data.") if T(2);
1036      # 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
1037      # user records.      # user records.
1038      my %users = ( FIG => 1, master => 1 );      my %users = ( FIG => 1, master => 1 );
# Line 1026  Line 1046 
1046      # Loop through the genomes.      # Loop through the genomes.
1047      for my $genomeID (sort keys %{$genomeHash}) {      for my $genomeID (sort keys %{$genomeHash}) {
1048          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);  
1049              # 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
1050              # from showing up for a single PEG's annotations.              # from showing up for a single PEG's annotations.
1051              my %seenTimestamps = ();              my %seenTimestamps = ();
1052              # Check for a functional assignment.              # Get the genome's annotations.
1053              my $func = $fig->function_of($peg);              my @annotations = $fig->read_all_annotations($genomeID);
1054              if ($func) {              Trace("Processing annotations.") if T(2);
1055                  # If this is NOT a hypothetical assignment, we create an              for my $tuple (@annotations) {
1056                  # assignment annotation for it.                  # Get the annotation tuple.
1057                  if (! FIG::hypo($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};  
1058                  # Here we fix up the annotation text. "\r" is removed,                  # Here we fix up the annotation text. "\r" is removed,
1059                  # and "\t" and "\n" are escaped. Note we use the "s"                  # and "\t" and "\n" are escaped. Note we use the "gs"
1060                  # modifier so that new-lines inside the text do not                  # modifier so that new-lines inside the text do not
1061                  # stop the substitution search.                  # stop the substitution search.
1062                  $text =~ s/\r//gs;                  $text =~ s/\r//gs;
# Line 1065  Line 1069 
1069                      # Here it's a number. We need to insure the one we use to form                      # Here it's a number. We need to insure the one we use to form
1070                      # the key is unique.                      # the key is unique.
1071                      my $keyStamp = $timestamp;                      my $keyStamp = $timestamp;
1072                      while ($seenTimestamps{$keyStamp}) {                      while ($seenTimestamps{"$peg:$keyStamp"}) {
1073                          $keyStamp++;                          $keyStamp++;
1074                      }                      }
                     $seenTimestamps{$keyStamp} = 1;  
1075                      my $annotationID = "$peg:$keyStamp";                      my $annotationID = "$peg:$keyStamp";
1076                        $seenTimestamps{$annotationID} = 1;
1077                      # Insure the user exists.                      # Insure the user exists.
1078                      if (! $users{$user}) {                      if (! $users{$user}) {
1079                          $loadSproutUser->Put($user, "SEED user");                          $loadSproutUser->Put($user, "SEED user");
# Line 1127  Line 1131 
1131      my $fig = $self->{fig};      my $fig = $self->{fig};
1132      # Get the genome hash.      # Get the genome hash.
1133      my $genomeHash = $self->{genomes};      my $genomeHash = $self->{genomes};
     my $genomeCount = (keys %{$genomeHash});  
1134      # Create load objects for each of the tables we're loading.      # Create load objects for each of the tables we're loading.
1135      my $loadComesFrom = $self->_TableLoader('ComesFrom', $genomeCount * 4);      my $loadComesFrom = $self->_TableLoader('ComesFrom', $self->PrimaryOnly);
1136      my $loadSource = $self->_TableLoader('Source', $genomeCount * 4);      my $loadSource = $self->_TableLoader('Source');
1137      my $loadSourceURL = $self->_TableLoader('SourceURL', $genomeCount * 8);      my $loadSourceURL = $self->_TableLoader('SourceURL');
1138      Trace("Beginning source data load.") if T(2);      if ($self->{options}->{loadOnly}) {
1139            Trace("Loading from existing files.") if T(2);
1140        } else {
1141            Trace("Generating annotation data.") if T(2);
1142      # Create hashes to collect the Source information.      # Create hashes to collect the Source information.
1143      my %sourceURL = ();      my %sourceURL = ();
1144      my %sourceDesc = ();      my %sourceDesc = ();
# Line 1162  Line 1168 
1168      for my $sourceID (keys %sourceDesc) {      for my $sourceID (keys %sourceDesc) {
1169          $loadSource->Put($sourceID, $sourceDesc{$sourceID});          $loadSource->Put($sourceID, $sourceDesc{$sourceID});
1170      }      }
1171        }
1172      # Finish the load.      # Finish the load.
1173      my $retVal = $self->_FinishAll();      my $retVal = $self->_FinishAll();
1174      return $retVal;      return $retVal;
# Line 1201  Line 1208 
1208      my $fig = $self->{fig};      my $fig = $self->{fig};
1209      # Get the genome hash.      # Get the genome hash.
1210      my $genomeHash = $self->{genomes};      my $genomeHash = $self->{genomes};
     my $genomeCount = (keys %{$genomeHash});  
1211      # 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
1212      # it the key.      # it the key.
1213      my %speciesHash = map { $fig->genus_species($_) => $_ } (keys %{$genomeHash});      my %speciesHash = map { $fig->genus_species($_) => $_ } (keys %{$genomeHash});
1214      # Create load objects for each of the tables we're loading.      # Create load objects for each of the tables we're loading.
1215      my $loadExternalAliasFunc = $self->_TableLoader('ExternalAliasFunc', $genomeCount * 4000);      my $loadExternalAliasFunc = $self->_TableLoader('ExternalAliasFunc');
1216      my $loadExternalAliasOrg = $self->_TableLoader('ExternalAliasOrg', $genomeCount * 4000);      my $loadExternalAliasOrg = $self->_TableLoader('ExternalAliasOrg');
1217      Trace("Beginning external data load.") if T(2);      if ($self->{options}->{loadOnly}) {
1218            Trace("Loading from existing files.") if T(2);
1219        } else {
1220            Trace("Generating external data.") if T(2);
1221      # 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.
1222      Open(\*ORGS, "<$FIG_Config::global/ext_org.table");          Open(\*ORGS, "sort +0 -1 -u -t\"\t\" $FIG_Config::global/ext_org.table |");
1223      my $orgLine;      my $orgLine;
1224      while (defined($orgLine = <ORGS>)) {      while (defined($orgLine = <ORGS>)) {
1225          # Clean the input line.          # Clean the input line.
# Line 1222  Line 1231 
1231      close ORGS;      close ORGS;
1232      # Now the function file.      # Now the function file.
1233      my $funcLine;      my $funcLine;
1234      Open(\*FUNCS, "<$FIG_Config::global/ext_func.table");          Open(\*FUNCS, "sort +0 -1 -u -t\"\t\" $FIG_Config::global/ext_func.table |");
1235      while (defined($funcLine = <FUNCS>)) {      while (defined($funcLine = <FUNCS>)) {
1236          # Clean the line ending.          # Clean the line ending.
1237          chomp $funcLine;          chomp $funcLine;
# Line 1238  Line 1247 
1247              $loadExternalAliasFunc->Put(@funcFields[0,1]);              $loadExternalAliasFunc->Put(@funcFields[0,1]);
1248          }          }
1249      }      }
1250        }
1251      # Finish the load.      # Finish the load.
1252      my $retVal = $self->_FinishAll();      my $retVal = $self->_FinishAll();
1253      return $retVal;      return $retVal;
# Line 1254  Line 1264 
1264    
1265  The following relations are loaded by this method.  The following relations are loaded by this method.
1266    
1267        Reaction
1268      ReactionURL      ReactionURL
1269      Compound      Compound
1270      CompoundName      CompoundName
# Line 1277  Line 1288 
1288      my ($self) = @_;      my ($self) = @_;
1289      # Get the FIG object.      # Get the FIG object.
1290      my $fig = $self->{fig};      my $fig = $self->{fig};
     # Get the genome hash.  
     my $genomeHash = $self->{genomes};  
     my $genomeCount = (keys %{$genomeHash});  
1291      # Create load objects for each of the tables we're loading.      # Create load objects for each of the tables we're loading.
1292      my $loadReactionURL = $self->_TableLoader('ReactionURL', $genomeCount * 4000);      my $loadReaction = $self->_TableLoader('Reaction');
1293      my $loadCompound = $self->_TableLoader('Compound', $genomeCount * 4000);      my $loadReactionURL = $self->_TableLoader('ReactionURL', $self->PrimaryOnly);
1294      my $loadCompoundName = $self->_TableLoader('CompoundName', $genomeCount * 8000);      my $loadCompound = $self->_TableLoader('Compound', $self->PrimaryOnly);
1295      my $loadCompoundCAS = $self->_TableLoader('CompoundCAS', $genomeCount * 4000);      my $loadCompoundName = $self->_TableLoader('CompoundName', $self->PrimaryOnly);
1296      my $loadIsAComponentOf = $self->_TableLoader('IsAComponentOf', $genomeCount * 12000);      my $loadCompoundCAS = $self->_TableLoader('CompoundCAS', $self->PrimaryOnly);
1297      Trace("Beginning reaction/compound data load.") if T(2);      my $loadIsAComponentOf = $self->_TableLoader('IsAComponentOf', $self->PrimaryOnly);
1298      # Create a hash to remember the compounds we've generated in the compound table.      if ($self->{options}->{loadOnly}) {
1299      my %compoundHash = ();          Trace("Loading from existing files.") if T(2);
1300      # Loop through the reactions.      } else {
1301            Trace("Generating annotation data.") if T(2);
1302            # First we create the compounds.
1303            my @compounds = $fig->all_compounds();
1304            for my $cid (@compounds) {
1305                # Check for names.
1306                my @names = $fig->names_of_compound($cid);
1307                # Each name will be given a priority number, starting with 1.
1308                my $prio = 1;
1309                for my $name (@names) {
1310                    $loadCompoundName->Put($cid, $name, $prio++);
1311                }
1312                # Create the main compound record. Note that the first name
1313                # becomes the label.
1314                my $label = (@names > 0 ? $names[0] : $cid);
1315                $loadCompound->Put($cid, $label);
1316                # Check for a CAS ID.
1317                my $cas = $fig->cas($cid);
1318                if ($cas) {
1319                    $loadCompoundCAS->Put($cid, $cas);
1320                }
1321            }
1322            # All the compounds are set up, so we need to loop through the reactions next. First,
1323            # we initialize the discriminator index. This is a single integer used to insure
1324            # duplicate elements in a reaction are not accidentally collapsed.
1325            my $discrim = 0;
1326      my @reactions = $fig->all_reactions();      my @reactions = $fig->all_reactions();
1327      for my $reactionID (@reactions) {      for my $reactionID (@reactions) {
1328                # Create the reaction record.
1329                $loadReaction->Put($reactionID, $fig->reversible($reactionID));
1330          # Compute the reaction's URL.          # Compute the reaction's URL.
1331          my $url = HTML::reaction_link($reactionID);          my $url = HTML::reaction_link($reactionID);
1332          # Put it in the ReactionURL table.          # Put it in the ReactionURL table.
# Line 1300  Line 1335 
1335          # substrates first and then products.          # substrates first and then products.
1336          for my $product (0, 1) {          for my $product (0, 1) {
1337              # Get the compounds of the current type for the current reaction. FIG will              # Get the compounds of the current type for the current reaction. FIG will
1338              # give us 3-tuples: [ID, Stoichometry, main-flag]. At this time we do not                  # give us 3-tuples: [ID, stoichiometry, main-flag]. At this time we do not
1339              # have location data in SEED, so it defaults to the empty string.              # have location data in SEED, so it defaults to the empty string.
1340              my @compounds = $fig->reaction2comp($reactionID, $product);              my @compounds = $fig->reaction2comp($reactionID, $product);
1341              for my $compData (@compounds) {              for my $compData (@compounds) {
1342                  # Extract the compound data from the current tuple.                  # Extract the compound data from the current tuple.
1343                  my ($cid, $stoich, $main) = @{$compData};                  my ($cid, $stoich, $main) = @{$compData};
1344                  # Link the compound to the reaction.                  # Link the compound to the reaction.
1345                  $loadIsAComponentOf->Put($cid, $reactionID, "", $main, $product, $stoich);                      $loadIsAComponentOf->Put($cid, $reactionID, $discrim++, "", $main,
1346                  # If this is a new compound, we need to create its table entries.                                               $product, $stoich);
                 if (! exists $compoundHash{$cid}) {  
                     $compoundHash{$cid} = 1;  
                     # Create the main compound record and denote we've done it.  
                     $loadCompound->Put($cid);  
                     # Check for a CAS ID.  
                     my $cas = $fig->cas($cid);  
                     if ($cas) {  
                         $loadCompoundCAS->Put($cid, $cas);  
                     }  
                     # Check for names.  
                     my @names = $fig->names_of_compound($cid);  
                     # Each name will be given a priority number, starting with 1.  
                     my $prio = 0;  
                     for my $name (@names) {  
                         $loadCompoundName->Put($cid, $name, $prio++);  
                     }  
1347                  }                  }
1348              }              }
1349          }          }
# Line 1344  Line 1363 
1363    
1364      GenomeGroups      GenomeGroups
1365    
1366  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,
1367    butThere is no direct support for genome groups in FIG, so we access the SEED
1368  files directly.  files directly.
1369    
1370  =over 4  =over 4
# Line 1364  Line 1384 
1384      my $fig = $self->{fig};      my $fig = $self->{fig};
1385      # Get the genome hash.      # Get the genome hash.
1386      my $genomeHash = $self->{genomes};      my $genomeHash = $self->{genomes};
     my $genomeCount = (keys %{$genomeHash});  
1387      # Create a load object for the table we're loading.      # Create a load object for the table we're loading.
1388      my $loadGenomeGroups = $self->_TableLoader('GenomeGroups', $genomeCount * 4);      my $loadGenomeGroups = $self->_TableLoader('GenomeGroups');
1389      Trace("Beginning group data load.") if T(2);      if ($self->{options}->{loadOnly}) {
1390            Trace("Loading from existing files.") if T(2);
1391        } else {
1392            Trace("Generating group data.") if T(2);
1393            # Currently there are no groups.
1394        }
1395        # Finish the load.
1396        my $retVal = $self->_FinishAll();
1397        return $retVal;
1398    }
1399    
1400    =head3 LoadSynonymData
1401    
1402    C<< my $stats = $spl->LoadSynonymData(); >>
1403    
1404    Load the synonym groups into Sprout.
1405    
1406    The following relations are loaded by this method.
1407    
1408        SynonymGroup
1409        IsSynonymGroupFor
1410    
1411    The source information for these relations is taken from the C<maps_to_id> method
1412    of the B<FIG> object. Unfortunately, to make this work, we need to use direct
1413    SQL against the FIG database.
1414    
1415    =over 4
1416    
1417    =item RETURNS
1418    
1419    Returns a statistics object for the loads.
1420    
1421    =back
1422    
1423    =cut
1424    #: Return Type $%;
1425    sub LoadSynonymData {
1426        # Get this object instance.
1427        my ($self) = @_;
1428        # Get the FIG object.
1429        my $fig = $self->{fig};
1430        # Get the genome hash.
1431        my $genomeHash = $self->{genomes};
1432        # Create a load object for the table we're loading.
1433        my $loadSynonymGroup = $self->_TableLoader('SynonymGroup');
1434        my $loadIsSynonymGroupFor = $self->_TableLoader('IsSynonymGroupFor');
1435        if ($self->{options}->{loadOnly}) {
1436            Trace("Loading from existing files.") if T(2);
1437        } else {
1438            Trace("Generating synonym group data.") if T(2);
1439            # Get the database handle.
1440            my $dbh = $fig->db_handle();
1441            # Ask for the synonyms.
1442            my $sth = $dbh->prepare_command("SELECT maps_to, syn_id FROM peg_synonyms ORDER BY maps_to");
1443            my $result = $sth->execute();
1444            if (! defined($result)) {
1445                Confess("Database error in Synonym load: " . $sth->errstr());
1446            } else {
1447                # Remember the current synonym.
1448                my $current_syn = "";
1449                # Count the features.
1450                my $featureCount = 0;
1451                # Loop through the synonym/peg pairs.
1452                while (my @row = $sth->fetchrow()) {
1453                    # Get the synonym ID and feature ID.
1454                    my ($syn_id, $peg) = @row;
1455                    # Insure it's for one of our genomes.
1456                    my $genomeID = FIG::genome_of($peg);
1457                    if (exists $genomeHash->{$genomeID}) {
1458                        # Verify the synonym.
1459                        if ($syn_id ne $current_syn) {
1460                            # It's new, so put it in the group table.
1461                            $loadSynonymGroup->Put($syn_id);
1462                            $current_syn = $syn_id;
1463                        }
1464                        # Connect the synonym to the peg.
1465                        $loadIsSynonymGroupFor->Put($syn_id, $peg);
1466                        # Count this feature.
1467                        $featureCount++;
1468                        if ($featureCount % 1000 == 0) {
1469                            Trace("$featureCount features processed.") if T(3);
1470                        }
1471                    }
1472                }
1473            }
1474        }
1475        # Finish the load.
1476        my $retVal = $self->_FinishAll();
1477        return $retVal;
1478    }
1479    
1480    =head3 LoadFamilyData
1481    
1482    C<< my $stats = $spl->LoadFamilyData(); >>
1483    
1484    Load the protein families into Sprout.
1485    
1486    The following relations are loaded by this method.
1487    
1488        Family
1489        IsFamilyForFeature
1490    
1491    The source information for these relations is taken from the C<families_for_protein>,
1492    C<family_function>, and C<sz_family> methods of the B<FIG> object.
1493    
1494    =over 4
1495    
1496    =item RETURNS
1497    
1498    Returns a statistics object for the loads.
1499    
1500    =back
1501    
1502    =cut
1503    #: Return Type $%;
1504    sub LoadFamilyData {
1505        # Get this object instance.
1506        my ($self) = @_;
1507        # Get the FIG object.
1508        my $fig = $self->{fig};
1509        # Get the genome hash.
1510        my $genomeHash = $self->{genomes};
1511        # Create load objects for the tables we're loading.
1512        my $loadFamily = $self->_TableLoader('Family');
1513        my $loadIsFamilyForFeature = $self->_TableLoader('IsFamilyForFeature');
1514        if ($self->{options}->{loadOnly}) {
1515            Trace("Loading from existing files.") if T(2);
1516        } else {
1517            Trace("Generating family data.") if T(2);
1518            # Create a hash for the family IDs.
1519            my %familyHash = ();
1520      # Loop through the genomes.      # Loop through the genomes.
1521      my $line;          for my $genomeID (sort keys %{$genomeHash}) {
1522      for my $genomeID (keys %{$genomeHash}) {              Trace("Processing features for $genomeID.") if T(2);
1523          Trace("Processing $genomeID.") if T(3);              # Loop through this genome's PEGs.
1524          # Open the NMPDR group file for this genome.              for my $fid ($fig->all_features($genomeID, "peg")) {
1525          if (open(TMP, "<$FIG_Config::organisms/$genomeID/NMPDR") &&                  $loadIsFamilyForFeature->Add("features", 1);
1526              defined($line = <TMP>)) {                  # Get this feature's families.
1527              # Clean the line ending.                  my @families = $fig->families_for_protein($fid);
1528              chomp $line;                  # Loop through the families, connecting them to the feature.
1529              # Add the group to the table. Note that there can only be one group                  for my $family (@families) {
1530              # per genome.                      $loadIsFamilyForFeature->Put($family, $fid);
1531              $loadGenomeGroups->Put($genomeID, $line);                      # If this is a new family, create a record for it.
1532                        if (! exists $familyHash{$family}) {
1533                            $familyHash{$family} = 1;
1534                            $loadFamily->Add("families", 1);
1535                            my $size = $fig->sz_family($family);
1536                            my $func = $fig->family_function($family);
1537                            $loadFamily->Put($family, $size, $func);
1538                        }
1539                    }
1540                }
1541          }          }
         close TMP;  
1542      }      }
1543      # Finish the load.      # Finish the load.
1544      my $retVal = $self->_FinishAll();      my $retVal = $self->_FinishAll();
1545      return $retVal;      return $retVal;
1546  }  }
1547    
1548    
1549    
1550  =head2 Internal Utility Methods  =head2 Internal Utility Methods
1551    
1552  =head3 TableLoader  =head3 TableLoader
# Line 1404  Line 1563 
1563    
1564  Name of the table (relation) being loaded.  Name of the table (relation) being loaded.
1565    
1566  =item rowCount (optional)  =item ignore
1567    
1568  Estimated maximum number of rows in the table.  TRUE if the table should be ignored entirely, else FALSE.
1569    
1570  =item RETURN  =item RETURN
1571    
# Line 1418  Line 1577 
1577    
1578  sub _TableLoader {  sub _TableLoader {
1579      # Get the parameters.      # Get the parameters.
1580      my ($self, $tableName, $rowCount) = @_;      my ($self, $tableName, $ignore) = @_;
1581      # Create the load object.      # Create the load object.
1582      my $retVal = ERDBLoad->new($self->{erdb}, $tableName, $self->{loadDirectory}, $rowCount);      my $retVal = ERDBLoad->new($self->{erdb}, $tableName, $self->{loadDirectory}, $self->LoadOnly,
1583                                   $ignore);
1584      # Cache it in the loader list.      # Cache it in the loader list.
1585      push @{$self->{loaders}}, $retVal;      push @{$self->{loaders}}, $retVal;
1586      # Return it to the caller.      # Return it to the caller.
# Line 1454  Line 1614 
1614      my $retVal = Stats->new();      my $retVal = Stats->new();
1615      # Get the loader list.      # Get the loader list.
1616      my $loadList = $self->{loaders};      my $loadList = $self->{loaders};
1617        # Create a hash to hold the statistics objects, keyed on relation name.
1618        my %loaderHash = ();
1619      # 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
1620      # 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.
1621      while (my $loader = pop @{$loadList}) {      while (my $loader = pop @{$loadList}) {
1622            # Get the relation name.
1623            my $relName = $loader->RelName;
1624            # Check the ignore flag.
1625            if ($loader->Ignore) {
1626                Trace("Relation $relName not loaded.") if T(2);
1627            } else {
1628                # Here we really need to finish.
1629                Trace("Finishing $relName.") if T(2);
1630          my $stats = $loader->Finish();          my $stats = $loader->Finish();
1631                $loaderHash{$relName} = $stats;
1632            }
1633        }
1634        # Now we loop through again, actually loading the tables. We want to finish before
1635        # loading so that if something goes wrong at this point, all the load files are usable
1636        # and we don't have to redo all that work.
1637        for my $relName (sort keys %loaderHash) {
1638            # Get the statistics for this relation.
1639            my $stats = $loaderHash{$relName};
1640            # Check for a database load.
1641            if ($self->{options}->{dbLoad}) {
1642                # Here we want to use the load file just created to load the database.
1643                Trace("Loading relation $relName.") if T(2);
1644                my $newStats = $self->{sprout}->LoadUpdate(1, [$relName]);
1645                # Accumulate the statistics from the DB load.
1646                $stats->Accumulate($newStats);
1647            }
1648          $retVal->Accumulate($stats);          $retVal->Accumulate($stats);
         my $relName = $loader->RelName;  
1649          Trace("Statistics for $relName:\n" . $stats->Show()) if T(2);          Trace("Statistics for $relName:\n" . $stats->Show()) if T(2);
1650      }      }
1651      # Return the load statistics.      # Return the load statistics.

Legend:
Removed from v.1.18  
changed lines
  Added in v.1.74

MCS Webmaster
ViewVC Help
Powered by ViewVC 1.0.3