[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 1.22, Mon Dec 12 21:48:22 2005 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 51  Line 52 
52    
53  =head3 new  =head3 new
54    
55  C<< my $spl = SproutLoad->new($sprout, $fig, $genomeFile, $subsysFile); >>  C<< my $spl = SproutLoad->new($sprout, $fig, $genomeFile, $subsysFile, $options); >>
56    
57  Construct a new Sprout Loader object, specifying the two participating databases and  Construct a new Sprout Loader object, specifying the two participating databases and
58  the name of the files containing the list of genomes and subsystems to use.  the name of the files containing the list of genomes and subsystems to use.
# Line 82  Line 83 
83  to a list of subsystem names. If nothing is specified, all known subsystems will be  to a list of subsystem names. If nothing is specified, all known subsystems will be
84  considered trusted. Only subsystem data related to the trusted subsystems is loaded.  considered trusted. Only subsystem data related to the trusted subsystems is loaded.
85    
86    =item options
87    
88    Reference to a hash of command-line options.
89    
90  =back  =back
91    
92  =cut  =cut
93    
94  sub new {  sub new {
95      # Get the parameters.      # Get the parameters.
96      my ($class, $sprout, $fig, $genomeFile, $subsysFile) = @_;      my ($class, $sprout, $fig, $genomeFile, $subsysFile, $options) = @_;
97      # Load the list of genomes into a hash.      # Load the list of genomes into a hash.
98      my %genomes;      my %genomes;
99      if (! defined($genomeFile) || $genomeFile eq '') {      if (! defined($genomeFile) || $genomeFile eq '') {
# Line 158  Line 163 
163                    sprout => $sprout,                    sprout => $sprout,
164                    loadDirectory => $directory,                    loadDirectory => $directory,
165                    erdb => $sprout->{_erdb},                    erdb => $sprout->{_erdb},
166                    loaders => []                    loaders => [],
167                      options => $options
168                   };                   };
169      # Bless and return it.      # Bless and return it.
170      bless $retVal, $class;      bless $retVal, $class;
# Line 362  Line 368 
368                              # We store this evidence in the hash if the usage                              # We store this evidence in the hash if the usage
369                              # is nonzero or no prior evidence has been found. This                              # is nonzero or no prior evidence has been found. This
370                              # insures that if there is duplicate evidence, we                              # insures that if there is duplicate evidence, we
371                              # at least keep the meaningful ones. Only evidence is                              # at least keep the meaningful ones. Only evidence in
372                              # the hash makes it to the output.                              # the hash makes it to the output.
373                              if ($usage || ! exists $evidenceMap{$evidenceKey}) {                              if ($usage || ! exists $evidenceMap{$evidenceKey}) {
374                                  $evidenceMap{$evidenceKey} = $evidenceData;                                  $evidenceMap{$evidenceKey} = $evidenceData;
# Line 377  Line 383 
383                          $loadIsEvidencedBy->Put($coupleID, $evidenceID);                          $loadIsEvidencedBy->Put($coupleID, $evidenceID);
384                          # Connect it to the features.                          # Connect it to the features.
385                          $loadUsesAsEvidence->Put($evidenceID, $peg3, 1);                          $loadUsesAsEvidence->Put($evidenceID, $peg3, 1);
386                          $loadUsesAsEvidence->Put($evidenceID, $peg4, 1);                          $loadUsesAsEvidence->Put($evidenceID, $peg4, 2);
387                      }                      }
388                  }                  }
389              }              }
# Line 420  Line 426 
426      my ($self) = @_;      my ($self) = @_;
427      # Get the FIG object.      # Get the FIG object.
428      my $fig = $self->{fig};      my $fig = $self->{fig};
429        # Find out if this is a limited run.
430        my $limited = $self->{options}->{limitedFeatures};
431      # Get the table of genome IDs.      # Get the table of genome IDs.
432      my $genomeHash = $self->{genomes};      my $genomeHash = $self->{genomes};
433      my $genomeCount = (keys %{$genomeHash});      my $genomeCount = (keys %{$genomeHash});
434      my $featureCount = $genomeCount * 4000;      my $featureCount = $genomeCount * 4000;
435      # Create load objects for each of the tables we're loading.      # Create load objects for each of the tables we're loading.
436      my $loadFeature = $self->_TableLoader('Feature', $featureCount);      my $loadFeature = $self->_TableLoader('Feature', $featureCount);
     my $loadFeatureAlias = $self->_TableLoader('FeatureAlias', $featureCount * 6);  
     my $loadFeatureLink = $self->_TableLoader('FeatureLink', $featureCount * 10);  
     my $loadFeatureTranslation = $self->_TableLoader('FeatureTranslation', $featureCount);  
     my $loadFeatureUpstream = $self->_TableLoader('FeatureUpstream', $featureCount);  
437      my $loadIsLocatedIn = $self->_TableLoader('IsLocatedIn', $featureCount);      my $loadIsLocatedIn = $self->_TableLoader('IsLocatedIn', $featureCount);
438        my $loadFeatureAlias = $self->_TableLoader('FeatureAlias', $featureCount * 6);
439        my ($loadFeatureLink, $loadFeatureTranslation, $loadFeatureUpstream);
440        if (! $limited) {
441            $loadFeatureLink = $self->_TableLoader('FeatureLink', $featureCount * 10);
442            $loadFeatureTranslation = $self->_TableLoader('FeatureTranslation', $featureCount);
443            $loadFeatureUpstream = $self->_TableLoader('FeatureUpstream', $featureCount);
444        }
445      # 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
446      # locations.      # locations.
447      my $chunkSize = $self->{sprout}->MaxSegment();      my $chunkSize = $self->{sprout}->MaxSegment();
# Line 445  Line 456 
456          for my $featureData (@{$features}) {          for my $featureData (@{$features}) {
457              $loadFeature->Add("featureIn");              $loadFeature->Add("featureIn");
458              # Split the tuple.              # Split the tuple.
459              my ($featureID, $locations, $aliases, $type) = @{$featureData};              my ($featureID, $locations, undef, $type) = @{$featureData};
460              # Create the feature record.              # Create the feature record.
461              $loadFeature->Put($featureID, 1, $type);              $loadFeature->Put($featureID, 1, $type);
462              # Create the aliases.              # Create the aliases.
463              for my $alias (split /\s*,\s*/, $aliases) {              for my $alias ($fig->feature_aliases($featureID)) {
464                  $loadFeatureAlias->Put($featureID, $alias);                  $loadFeatureAlias->Put($featureID, $alias);
465              }              }
466                # The next stuff is for a full load only.
467                if (! $limited) {
468              # Get the links.              # Get the links.
469              my @links = $fig->fid_links($featureID);              my @links = $fig->fid_links($featureID);
470              for my $link (@links) {              for my $link (@links) {
# Line 470  Line 483 
483                      $loadFeatureUpstream->Put($featureID, $upstream);                      $loadFeatureUpstream->Put($featureID, $upstream);
484                  }                  }
485              }              }
486                }
487              # 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
488              # 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
489              # the maximum segment size. This simplifies the genes_in_region processing              # the maximum segment size. This simplifies the genes_in_region processing
490              # for Sprout.              # for Sprout.
491              my @locationList = split /\s*,\s*/, $locations;              my @locationList = split /\s*,\s*/, $locations;
492                # Create the location position indicator.
493                my $i = 1;
494              # Loop through the locations.              # Loop through the locations.
495              for my $location (@locationList) {              for my $location (@locationList) {
496                  # Parse the location.                  # Parse the location.
497                  my $locObject = BasicLocation->new($location);                  my $locObject = BasicLocation->new("$genomeID:$location");
498                  # Split it into a list of chunks.                  # Split it into a list of chunks.
499                  my @locOList = ();                  my @locOList = ();
500                  while (my $peeling = $locObject->Peel($chunkSize)) {                  while (my $peeling = $locObject->Peel($chunkSize)) {
# Line 488  Line 504 
504                  push @locOList, $locObject;                  push @locOList, $locObject;
505                  # Loop through the chunks, creating IsLocatedIn records. The variable                  # Loop through the chunks, creating IsLocatedIn records. The variable
506                  # "$i" will be used to keep the location index.                  # "$i" will be used to keep the location index.
                 my $i = 1;  
507                  for my $locChunk (@locOList) {                  for my $locChunk (@locOList) {
508                      $loadIsLocatedIn->Put($featureID, $locChunk->Contig, $locChunk->Left,                      $loadIsLocatedIn->Put($featureID, $locChunk->Contig, $locChunk->Left,
509                                            $locChunk->Dir, $locChunk->Length, $i);                                            $locChunk->Dir, $locChunk->Length, $i);
# Line 585  Line 600 
600    
601      Subsystem      Subsystem
602      Role      Role
603        RoleEC
604      SSCell      SSCell
605      ContainsFeature      ContainsFeature
606      IsGenomeOf      IsGenomeOf
# Line 592  Line 608 
608      OccursInSubsystem      OccursInSubsystem
609      ParticipatesIn      ParticipatesIn
610      HasSSCell      HasSSCell
611        ConsistsOfRoles
612        RoleSubset
613        HasRoleSubset
614        ConsistsOfGenomes
615        GenomeSubset
616        HasGenomeSubset
617        Catalyzes
618        Diagram
619        RoleOccursIn
620    
621  =over 4  =over 4
622    
# Line 601  Line 626 
626    
627  =back  =back
628    
 B<TO DO>  
   
 Generate RoleName table?  
   
629  =cut  =cut
630  #: Return Type $%;  #: Return Type $%;
631  sub LoadSubsystemData {  sub LoadSubsystemData {
# Line 621  Line 642 
642      my $subsysCount = @subsysIDs;      my $subsysCount = @subsysIDs;
643      my $genomeCount = (keys %{$genomeHash});      my $genomeCount = (keys %{$genomeHash});
644      my $featureCount = $genomeCount * 4000;      my $featureCount = $genomeCount * 4000;
645        # Get the map list.
646        my @maps = $fig->all_maps;
647        my $mapCount = @maps;
648      # Create load objects for each of the tables we're loading.      # Create load objects for each of the tables we're loading.
649        my $loadDiagram = $self->_TableLoader('Diagram', $mapCount);
650        my $loadRoleOccursIn = $self->_TableLoader('RoleOccursIn', $featureCount * 6);
651      my $loadSubsystem = $self->_TableLoader('Subsystem', $subsysCount);      my $loadSubsystem = $self->_TableLoader('Subsystem', $subsysCount);
652      my $loadRole = $self->_TableLoader('Role', $featureCount * 6);      my $loadRole = $self->_TableLoader('Role', $featureCount * 6);
653        my $loadRoleEC = $self->_TableLoader('RoleEC', $featureCount * 6);
654        my $loadCatalyzes = $self->_TableLoader('Catalyzes', $genomeCount * $featureCount);
655      my $loadSSCell = $self->_TableLoader('SSCell', $featureCount * $genomeCount);      my $loadSSCell = $self->_TableLoader('SSCell', $featureCount * $genomeCount);
656      my $loadContainsFeature = $self->_TableLoader('ContainsFeature', $featureCount * $subsysCount);      my $loadContainsFeature = $self->_TableLoader('ContainsFeature', $featureCount * $subsysCount);
657      my $loadIsGenomeOf = $self->_TableLoader('IsGenomeOf', $featureCount * $genomeCount);      my $loadIsGenomeOf = $self->_TableLoader('IsGenomeOf', $featureCount * $genomeCount);
# Line 631  Line 659 
659      my $loadOccursInSubsystem = $self->_TableLoader('OccursInSubsystem', $featureCount * 6);      my $loadOccursInSubsystem = $self->_TableLoader('OccursInSubsystem', $featureCount * 6);
660      my $loadParticipatesIn = $self->_TableLoader('ParticipatesIn', $subsysCount * $genomeCount);      my $loadParticipatesIn = $self->_TableLoader('ParticipatesIn', $subsysCount * $genomeCount);
661      my $loadHasSSCell = $self->_TableLoader('HasSSCell', $featureCount * $genomeCount);      my $loadHasSSCell = $self->_TableLoader('HasSSCell', $featureCount * $genomeCount);
662        my $loadRoleSubset = $self->_TableLoader('RoleSubset', $subsysCount * 50);
663        my $loadGenomeSubset = $self->_TableLoader('GenomeSubset', $subsysCount * 50);
664        my $loadConsistsOfRoles = $self->_TableLoader('ConsistsOfRoles', $featureCount * $genomeCount);
665        my $loadConsistsOfGenomes = $self->_TableLoader('ConsistsOfGenomes', $featureCount * $genomeCount);
666        my $loadHasRoleSubset = $self->_TableLoader('HasRoleSubset', $subsysCount * 50);
667        my $loadHasGenomeSubset = $self->_TableLoader('HasGenomeSubset', $subsysCount * 50);
668        # Create load objects for each of the tables we're loading.
669      Trace("Beginning subsystem data load.") if T(2);      Trace("Beginning subsystem data load.") if T(2);
670        # This hash will contain the role for each EC. When we're done, this
671        # information will be used to generate the Catalyzes table.
672        my %ecToRoles = ();
673      # Loop through the subsystems. Our first task will be to create the      # Loop through the subsystems. Our first task will be to create the
674      # roles. We do this by looping through the subsystems and creating a      # roles. We do this by looping through the subsystems and creating a
675      # role hash. The hash tracks each role ID so that we don't create      # role hash. The hash tracks each role ID so that we don't create
676      # duplicates. As we move along, we'll connect the roles and subsystems.      # duplicates. As we move along, we'll connect the roles and subsystems
677        # and memorize up the reactions.
678        my ($genomeID, $roleID);
679      my %roleData = ();      my %roleData = ();
680      for my $subsysID (@subsysIDs) {      for my $subsysID (@subsysIDs) {
681          Trace("Creating subsystem $subsysID.") if T(3);          Trace("Creating subsystem $subsysID.") if T(3);
682          $loadSubsystem->Add("subsystemIn");          $loadSubsystem->Add("subsystemIn");
683            # Get the subsystem object.
684            my $sub = $fig->get_subsystem($subsysID);
685          # Create the subsystem record.          # Create the subsystem record.
686          $loadSubsystem->Put($subsysID);          my $curator = $sub->get_curator();
687          # Get the subsystem's roles.          my $notes = $sub->get_notes();
688          my @roles = $fig->subsystem_to_roles($subsysID);          $loadSubsystem->Put($subsysID, $curator, $notes);
689          # Connect the roles to the subsystem. If a role is new, we create          # Connect it to its roles. Each role is a column in the subsystem spreadsheet.
690          # a role record for it.          for (my $col = 0; defined($roleID = $sub->get_role($col)); $col++) {
691          for my $roleID (@roles) {              # Connect to this role.
692              $loadOccursInSubsystem->Add("roleIn");              $loadOccursInSubsystem->Add("roleIn");
693              $loadOccursInSubsystem->Put($roleID, $subsysID);              $loadOccursInSubsystem->Put($roleID, $subsysID, $col);
694                # If it's a new role, add it to the role table.
695              if (! exists $roleData{$roleID}) {              if (! exists $roleData{$roleID}) {
696                  $loadRole->Put($roleID);                  # Get the role's abbreviation.
697                    my $abbr = $sub->get_role_abbr($col);
698                    # Add the role.
699                    $loadRole->Put($roleID, $abbr);
700                  $roleData{$roleID} = 1;                  $roleData{$roleID} = 1;
701                    # Check for an EC number.
702                    if ($roleID =~ /\(EC ([^.]+\.[^.]+\.[^.]+\.[^)]+)\)\s*$/) {
703                        my $ec = $1;
704                        $loadRoleEC->Put($roleID, $ec);
705                        $ecToRoles{$ec} = $roleID;
706              }              }
707          }          }
708          # Now all roles for this subsystem have been filled in. We create the          }
709          # spreadsheet by matches roles to genomes. To do this, we need to          # Now we create the spreadsheet for the subsystem by matching roles to
710          # get the genomes on the sheet.          # genomes. Each genome is a row and each role is a column. We may need
711            # to actually create the roles as we find them.
712          Trace("Creating subsystem $subsysID spreadsheet.") if T(3);          Trace("Creating subsystem $subsysID spreadsheet.") if T(3);
713          my @genomes = map { $_->[0] } @{$fig->subsystem_genomes($subsysID)};          for (my $row = 0; defined($genomeID = $sub->get_genome($row)); $row++) {
714          for my $genomeID (@genomes) {              # Only proceed if this is one of our genomes.
             # Only process this genome if it's one of ours.  
715              if (exists $genomeHash->{$genomeID}) {              if (exists $genomeHash->{$genomeID}) {
716                  # Connect the genome to the subsystem.                  # Count the PEGs and cells found for verification purposes.
717                  $loadParticipatesIn->Put($genomeID, $subsysID);                  my $pegCount = 0;
718                    my $cellCount = 0;
719                    # Create a list for the PEGs we find. This list will be used
720                    # to generate cluster numbers.
721                    my @pegsFound = ();
722                    # Create a hash that maps spreadsheet IDs to PEGs. We will
723                    # use this to generate the ContainsFeature data after we have
724                    # the cluster numbers.
725                    my %cellPegs = ();
726                    # Get the genome's variant code for this subsystem.
727                    my $variantCode = $sub->get_variant_code($row);
728                  # 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
729                  # part of the spreadsheet cell ID.                  # part of the spreadsheet cell ID.
730                  for (my $i = 0; $i <= $#roles; $i++) {                  for (my $col = 0; defined($roleID = $sub->get_role($col)); $col++) {
                     my $role = $roles[$i];  
731                      # Get the features in the spreadsheet cell for this genome and role.                      # Get the features in the spreadsheet cell for this genome and role.
732                      my @pegs = $fig->pegs_in_subsystem_cell($subsysID, $genomeID, $i);                      my @pegs = $sub->get_pegs_from_cell($row, $col);
733                      # Only proceed if features exist.                      # Only proceed if features exist.
734                      if (@pegs > 0) {                      if (@pegs > 0) {
735                          # Create the spreadsheet cell.                          # Create the spreadsheet cell.
736                          my $cellID = "$subsysID:$genomeID:$i";                          $cellCount++;
737                            my $cellID = "$subsysID:$genomeID:$col";
738                          $loadSSCell->Put($cellID);                          $loadSSCell->Put($cellID);
739                          $loadIsGenomeOf->Put($genomeID, $cellID);                          $loadIsGenomeOf->Put($genomeID, $cellID);
740                          $loadIsRoleOf->Put($role, $cellID);                          $loadIsRoleOf->Put($roleID, $cellID);
741                          $loadHasSSCell->Put($subsysID, $cellID);                          $loadHasSSCell->Put($subsysID, $cellID);
742                          # Attach the features to it.                          # Remember its features.
743                          for my $pegID (@pegs) {                          push @pegsFound, @pegs;
744                              $loadContainsFeature->Put($cellID, $pegID);                          $cellPegs{$cellID} = \@pegs;
745                            $pegCount += @pegs;
746                        }
747                    }
748                    # If we found some cells for this genome, we need to compute clusters and
749                    # denote it participates in the subsystem.
750                    if ($pegCount > 0) {
751                        Trace("$pegCount PEGs in $cellCount cells for $genomeID.") if T(3);
752                        $loadParticipatesIn->Put($genomeID, $subsysID, $variantCode);
753                        # Partition the PEGs found into clusters.
754                        my @clusters = $fig->compute_clusters(\@pegsFound, $sub);
755                        # Create a hash mapping PEG IDs to cluster numbers.
756                        # We default to -1 for all of them.
757                        my %clusterOf = map { $_ => -1 } @pegsFound;
758                        for (my $i = 0; $i <= $#clusters; $i++) {
759                            my $subList = $clusters[$i];
760                            for my $peg (@{$subList}) {
761                                $clusterOf{$peg} = $i;
762                            }
763                        }
764                        # Create the ContainsFeature data.
765                        for my $cellID (keys %cellPegs) {
766                            my $cellList = $cellPegs{$cellID};
767                            for my $cellPeg (@$cellList) {
768                                $loadContainsFeature->Put($cellID, $cellPeg, $clusterOf{$cellPeg});
769                          }                          }
770                      }                      }
771                  }                  }
772              }              }
773          }          }
774            # Now we need to generate the subsets. The subset names must be concatenated to
775            # the subsystem name to make them unique keys. There are two types of subsets:
776            # genome subsets and role subsets. We do the role subsets first.
777            my @subsetNames = $sub->get_subset_names();
778            for my $subsetID (@subsetNames) {
779                # Create the subset record.
780                my $actualID = "$subsysID:$subsetID";
781                $loadRoleSubset->Put($actualID);
782                # Connect the subset to the subsystem.
783                $loadHasRoleSubset->Put($subsysID, $actualID);
784                # Connect the subset to its roles.
785                my @roles = $sub->get_subset($subsetID);
786                for my $roleID (@roles) {
787                    $loadConsistsOfRoles->Put($actualID, $roleID);
788      }      }
     # Finish the load.  
     my $retVal = $self->_FinishAll();  
     return $retVal;  
789  }  }
790            # Next the genome subsets.
791  =head3 LoadDiagramData          @subsetNames = $sub->get_subset_namesR();
792            for my $subsetID (@subsetNames) {
793  C<< my $stats = $spl->LoadDiagramData(); >>              # Create the subset record.
794                my $actualID = "$subsysID:$subsetID";
795  Load the diagram data from FIG into Sprout.              $loadGenomeSubset->Put($actualID);
796                # Connect the subset to the subsystem.
797  Diagrams are used to organize functional roles. The diagram shows the              $loadHasGenomeSubset->Put($subsysID, $actualID);
798  connections between chemicals that interact with a subsystem.              # Connect the subset to its genomes.
799                my @genomes = $sub->get_subsetR($subsetID);
800  The following relations are loaded by this method.              for my $genomeID (@genomes) {
801                    $loadConsistsOfGenomes->Put($actualID, $genomeID);
802      Diagram              }
803      RoleOccursIn          }
804        }
805  =over 4      # Now we loop through the diagrams. We need to create the diagram records
806        # and link each diagram to its roles. Note that only roles which occur
807  =item RETURNS      # in subsystems (and therefore appear in the %ecToRoles hash) are
808        # included.
809  Returns a statistics object for the loads.      for my $map (@maps) {
   
 =back  
   
 =cut  
 #: Return Type $%;  
 sub LoadDiagramData {  
     # Get this object instance.  
     my ($self) = @_;  
     # Get the FIG object.  
     my $fig = $self->{fig};  
     # Get the map list.  
     my @maps = $fig->all_maps;  
     my $mapCount = @maps;  
     my $genomeCount = (keys %{$self->{genomes}});  
     my $featureCount = $genomeCount * 4000;  
     # Create load objects for each of the tables we're loading.  
     my $loadDiagram = $self->_TableLoader('Diagram', $mapCount);  
     my $loadRoleOccursIn = $self->_TableLoader('RoleOccursIn', $featureCount * 6);  
     Trace("Beginning diagram data load.") if T(2);  
     # Loop through the diagrams.  
     for my $map ($fig->all_maps) {  
810          Trace("Loading diagram $map.") if T(3);          Trace("Loading diagram $map.") if T(3);
811          # Get the diagram's descriptive name.          # Get the diagram's descriptive name.
812          my $name = $fig->map_name($map);          my $name = $fig->map_name($map);
# Line 740  Line 815 
815          # A hash is used to prevent duplicates.          # A hash is used to prevent duplicates.
816          my %roleHash = ();          my %roleHash = ();
817          for my $role ($fig->map_to_ecs($map)) {          for my $role ($fig->map_to_ecs($map)) {
818              if (! $roleHash{$role}) {              if (exists $ecToRoles{$role} && ! $roleHash{$role}) {
819                  $loadRoleOccursIn->Put($role, $map);                  $loadRoleOccursIn->Put($ecToRoles{$role}, $map);
820                  $roleHash{$role} = 1;                  $roleHash{$role} = 1;
821              }              }
822          }          }
823      }      }
824        # Before we leave, we must create the Catalyzes table. We start with the reactions,
825        # then use the "ecToRoles" table to convert EC numbers to role IDs.
826        my @reactions = $fig->all_reactions();
827        for my $reactionID (@reactions) {
828            # Get this reaction's list of roles. The results will be EC numbers.
829            my @roles = $fig->catalyzed_by($reactionID);
830            # Loop through the roles, creating catalyzation records.
831            for my $thisRole (@roles) {
832                if (exists $ecToRoles{$thisRole}) {
833                    $loadCatalyzes->Put($ecToRoles{$thisRole}, $reactionID);
834                }
835            }
836        }
837      # Finish the load.      # Finish the load.
838      my $retVal = $self->_FinishAll();      my $retVal = $self->_FinishAll();
839      return $retVal;      return $retVal;
# Line 899  Line 987 
987              # 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
988              # from showing up for a single PEG's annotations.              # from showing up for a single PEG's annotations.
989              my %seenTimestamps = ();              my %seenTimestamps = ();
990              # Check for a functional assignment.              # Loop through the annotations.
             my $func = $fig->function_of($peg);  
             if ($func) {  
                 # If this is NOT a hypothetical assignment, we create an  
                 # assignment annotation for it.  
                 if (! FIG::hypo($peg)) {  
                     # 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.  
991                  for my $tuple ($fig->feature_annotations($peg, "raw")) {                  for my $tuple ($fig->feature_annotations($peg, "raw")) {
992                      my ($fid, $timestamp, $user, $text) = @{$tuple};                      my ($fid, $timestamp, $user, $text) = @{$tuple};
993                      # Here we fix up the annotation text. "\r" is removed,                      # Here we fix up the annotation text. "\r" is removed,
# Line 927  Line 1001 
1001                      $text =~ s/Set master function/Set FIG function/s;                      $text =~ s/Set master function/Set FIG function/s;
1002                      # Insure the time stamp is valid.                      # Insure the time stamp is valid.
1003                      if ($timestamp =~ /^\d+$/) {                      if ($timestamp =~ /^\d+$/) {
1004                          # Here it's a number. We need to insure it's unique.                      # Here it's a number. We need to insure the one we use to form
1005                          while ($seenTimestamps{$timestamp}) {                      # the key is unique.
1006                              $timestamp++;                      my $keyStamp = $timestamp;
1007                        while ($seenTimestamps{$keyStamp}) {
1008                            $keyStamp++;
1009                          }                          }
1010                          $seenTimestamps{$timestamp} = 1;                      $seenTimestamps{$keyStamp} = 1;
1011                          my $annotationID = "$peg:$timestamp";                      my $annotationID = "$peg:$keyStamp";
1012                          # Insure the user exists.                          # Insure the user exists.
1013                          if (! $users{$user}) {                          if (! $users{$user}) {
1014                              $loadSproutUser->Put($user, "SEED user");                              $loadSproutUser->Put($user, "SEED user");
# Line 940  Line 1016 
1016                              $users{$user} = 1;                              $users{$user} = 1;
1017                          }                          }
1018                          # Generate the annotation.                          # Generate the annotation.
1019                          $loadAnnotation->Put($annotationID, $timestamp, "$user\\n$text");                      $loadAnnotation->Put($annotationID, $timestamp, $text);
1020                          $loadIsTargetOfAnnotation->Put($peg, $annotationID);                          $loadIsTargetOfAnnotation->Put($peg, $annotationID);
1021                          $loadMadeAnnotation->Put($user, $annotationID);                          $loadMadeAnnotation->Put($user, $annotationID);
1022                      } else {                      } else {
# Line 950  Line 1026 
1026                  }                  }
1027              }              }
1028          }          }
     }  
1029      # Finish the load.      # Finish the load.
1030      my $retVal = $self->_FinishAll();      my $retVal = $self->_FinishAll();
1031      return $retVal;      return $retVal;
# Line 1010  Line 1085 
1085              chomp $line;              chomp $line;
1086              my($sourceID, $desc, $url) = split(/\t/,$line);              my($sourceID, $desc, $url) = split(/\t/,$line);
1087              $loadComesFrom->Put($genomeID, $sourceID);              $loadComesFrom->Put($genomeID, $sourceID);
1088              if ($url && ! exists $sourceURL{$genomeID}) {              if ($url && ! exists $sourceURL{$sourceID}) {
1089                  $loadSourceURL->Put($sourceID, $url);                  $loadSourceURL->Put($sourceID, $url);
1090                  $sourceURL{$sourceID} = 1;                  $sourceURL{$sourceID} = 1;
1091              }              }
1092              if ($desc && ! exists $sourceDesc{$sourceID}) {              if ($desc) {
1093                  $loadSource->Put($sourceID, $desc);                  $sourceDesc{$sourceID} = $desc;
1094                  $sourceDesc{$sourceID} = 1;              } elsif (! exists $sourceDesc{$sourceID}) {
1095                    $sourceDesc{$sourceID} = $sourceID;
1096              }              }
1097          }          }
1098          close TMP;          close TMP;
1099      }      }
1100        # Write the source descriptions.
1101        for my $sourceID (keys %sourceDesc) {
1102            $loadSource->Put($sourceID, $sourceDesc{$sourceID});
1103        }
1104      # Finish the load.      # Finish the load.
1105      my $retVal = $self->_FinishAll();      my $retVal = $self->_FinishAll();
1106      return $retVal;      return $retVal;
# Line 1102  Line 1182 
1182      return $retVal;      return $retVal;
1183  }  }
1184    
1185    
1186    =head3 LoadReactionData
1187    
1188    C<< my $stats = $spl->LoadReactionData(); >>
1189    
1190    Load the reaction data from FIG into Sprout.
1191    
1192    Reaction data connects reactions to the compounds that participate in them.
1193    
1194    The following relations are loaded by this method.
1195    
1196        Reaction
1197        ReactionURL
1198        Compound
1199        CompoundName
1200        CompoundCAS
1201        IsAComponentOf
1202    
1203    This method proceeds reaction by reaction rather than genome by genome.
1204    
1205    =over 4
1206    
1207    =item RETURNS
1208    
1209    Returns a statistics object for the loads.
1210    
1211    =back
1212    
1213    =cut
1214    #: Return Type $%;
1215    sub LoadReactionData {
1216        # Get this object instance.
1217        my ($self) = @_;
1218        # Get the FIG object.
1219        my $fig = $self->{fig};
1220        # Get the genome hash.
1221        my $genomeHash = $self->{genomes};
1222        my $genomeCount = (keys %{$genomeHash});
1223        # Create load objects for each of the tables we're loading.
1224        my $loadReaction = $self->_TableLoader('Reaction', $genomeCount * 4000);
1225        my $loadReactionURL = $self->_TableLoader('ReactionURL', $genomeCount * 4000);
1226        my $loadCompound = $self->_TableLoader('Compound', $genomeCount * 4000);
1227        my $loadCompoundName = $self->_TableLoader('CompoundName', $genomeCount * 8000);
1228        my $loadCompoundCAS = $self->_TableLoader('CompoundCAS', $genomeCount * 4000);
1229        my $loadIsAComponentOf = $self->_TableLoader('IsAComponentOf', $genomeCount * 12000);
1230        Trace("Beginning reaction/compound data load.") if T(2);
1231        # First we create the compounds.
1232        my @compounds = $fig->all_compounds();
1233        for my $cid (@compounds) {
1234            # Check for names.
1235            my @names = $fig->names_of_compound($cid);
1236            # Each name will be given a priority number, starting with 1.
1237            my $prio = 1;
1238            for my $name (@names) {
1239                $loadCompoundName->Put($cid, $name, $prio++);
1240            }
1241            # Create the main compound record. Note that the first name
1242            # becomes the label.
1243            my $label = (@names > 0 ? $names[0] : $cid);
1244            $loadCompound->Put($cid, $label);
1245            # Check for a CAS ID.
1246            my $cas = $fig->cas($cid);
1247            if ($cas) {
1248                $loadCompoundCAS->Put($cid, $cas);
1249            }
1250        }
1251        # All the compounds are set up, so we need to loop through the reactions next. First,
1252        # we initialize the discriminator index. This is a single integer used to insure
1253        # duplicate elements in a reaction are not accidentally collapsed.
1254        my $discrim = 0;
1255        my @reactions = $fig->all_reactions();
1256        for my $reactionID (@reactions) {
1257            # Create the reaction record.
1258            $loadReaction->Put($reactionID, $fig->reversible($reactionID));
1259            # Compute the reaction's URL.
1260            my $url = HTML::reaction_link($reactionID);
1261            # Put it in the ReactionURL table.
1262            $loadReactionURL->Put($reactionID, $url);
1263            # Now we need all of the reaction's compounds. We get these in two phases,
1264            # substrates first and then products.
1265            for my $product (0, 1) {
1266                # Get the compounds of the current type for the current reaction. FIG will
1267                # give us 3-tuples: [ID, stoichiometry, main-flag]. At this time we do not
1268                # have location data in SEED, so it defaults to the empty string.
1269                my @compounds = $fig->reaction2comp($reactionID, $product);
1270                for my $compData (@compounds) {
1271                    # Extract the compound data from the current tuple.
1272                    my ($cid, $stoich, $main) = @{$compData};
1273                    # Link the compound to the reaction.
1274                    $loadIsAComponentOf->Put($cid, $reactionID, $discrim++, "", $main,
1275                                             $product, $stoich);
1276                }
1277            }
1278        }
1279        # Finish the load.
1280        my $retVal = $self->_FinishAll();
1281        return $retVal;
1282    }
1283    
1284  =head3 LoadGroupData  =head3 LoadGroupData
1285    
1286  C<< my $stats = $spl->LoadGroupData(); >>  C<< my $stats = $spl->LoadGroupData(); >>
# Line 1225  Line 1404 
1404      # 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
1405      # ignominiously. At some future point, we want to make the loads restartable.      # ignominiously. At some future point, we want to make the loads restartable.
1406      while (my $loader = pop @{$loadList}) {      while (my $loader = pop @{$loadList}) {
1407            # Trace the fact that we're cleaning up.
1408            my $relName = $loader->RelName;
1409            Trace("Finishing load for $relName.") if T(2);
1410          my $stats = $loader->Finish();          my $stats = $loader->Finish();
1411            if ($self->{options}->{dbLoad}) {
1412                # Here we want to use the load file just created to load the database.
1413                Trace("Loading relation $relName.") if T(2);
1414                my $newStats = $self->{sprout}->LoadUpdate(1, [$relName]);
1415                # Accumulate the statistics from the DB load.
1416                $stats->Accumulate($newStats);
1417            }
1418          $retVal->Accumulate($stats);          $retVal->Accumulate($stats);
         my $relName = $loader->RelName;  
1419          Trace("Statistics for $relName:\n" . $stats->Show()) if T(2);          Trace("Statistics for $relName:\n" . $stats->Show()) if T(2);
1420      }      }
1421      # Return the load statistics.      # Return the load statistics.

Legend:
Removed from v.1.7  
changed lines
  Added in v.1.22

MCS Webmaster
ViewVC Help
Powered by ViewVC 1.0.3