[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 1.76, Fri Nov 3 00:43:22 2006 UTC
# Line 10  Line 10 
10      use Sprout;      use Sprout;
11      use Stats;      use Stats;
12      use BasicLocation;      use BasicLocation;
13        use HTML;
14    
15  =head1 Sprout Load Methods  =head1 Sprout Load Methods
16    
# Line 29  Line 30 
30      $stats->Accumulate($spl->LoadFeatureData());      $stats->Accumulate($spl->LoadFeatureData());
31      print $stats->Show();      print $stats->Show();
32    
 This module makes use of the internal Sprout property C<_erdb>.  
   
33  It is worth noting that the FIG object does not need to be a real one. Any object  It is worth noting that the FIG object does not need to be a real one. Any object
34  that implements the FIG methods for data retrieval could be used. So, for example,  that implements the FIG methods for data retrieval could be used. So, for example,
35  this object could be used to copy data from one Sprout database to another, or  this object could be used to copy data from one Sprout database to another, or
# Line 51  Line 50 
50    
51  =head3 new  =head3 new
52    
53  C<< my $spl = SproutLoad->new($sprout, $fig, $genomeFile, $subsysFile); >>  C<< my $spl = SproutLoad->new($sprout, $fig, $genomeFile, $subsysFile, $options); >>
54    
55  Construct a new Sprout Loader object, specifying the two participating databases and  Construct a new Sprout Loader object, specifying the two participating databases and
56  the name of the files containing the list of genomes and subsystems to use.  the name of the files containing the list of genomes and subsystems to use.
# Line 79  Line 78 
78  =item subsysFile  =item subsysFile
79    
80  Either the name of the file containing the list of trusted subsystems or a reference  Either the name of the file containing the list of trusted subsystems or a reference
81  to a list of subsystem names. If nothing is specified, all known subsystems will be  to a list of subsystem names. If nothing is specified, all NMPDR subsystems will be
82  considered trusted. Only subsystem data related to the trusted subsystems is loaded.  considered trusted. (A subsystem is considered NMPDR if it has a file named C<NMPDR>
83    in its data directory.) Only subsystem data related to the NMPDR subsystems is loaded.
84    
85    =item options
86    
87    Reference to a hash of command-line options.
88    
89  =back  =back
90    
# Line 88  Line 92 
92    
93  sub new {  sub new {
94      # Get the parameters.      # Get the parameters.
95      my ($class, $sprout, $fig, $genomeFile, $subsysFile) = @_;      my ($class, $sprout, $fig, $genomeFile, $subsysFile, $options) = @_;
96      # Load the list of genomes into a hash.      # Create the genome hash.
97      my %genomes;      my %genomes = ();
98        # We only need it if load-only is NOT specified.
99        if (! $options->{loadOnly}) {
100      if (! defined($genomeFile) || $genomeFile eq '') {      if (! defined($genomeFile) || $genomeFile eq '') {
101          # Here we want all the complete genomes and an access code of 1.          # Here we want all the complete genomes and an access code of 1.
102          my @genomeList = $fig->genomes(1);          my @genomeList = $fig->genomes(1);
# Line 114  Line 120 
120                  # an omitted access code can be defaulted to 1.                  # an omitted access code can be defaulted to 1.
121                  for my $genomeLine (@genomeList) {                  for my $genomeLine (@genomeList) {
122                      my ($genomeID, $accessCode) = split("\t", $genomeLine);                      my ($genomeID, $accessCode) = split("\t", $genomeLine);
123                      if (undef $accessCode) {                          if (! defined($accessCode)) {
124                          $accessCode = 1;                          $accessCode = 1;
125                      }                      }
126                      $genomes{$genomeID} = $accessCode;                      $genomes{$genomeID} = $accessCode;
# Line 124  Line 130 
130              Confess("Invalid genome parameter ($type) in SproutLoad constructor.");              Confess("Invalid genome parameter ($type) in SproutLoad constructor.");
131          }          }
132      }      }
133        }
134      # Load the list of trusted subsystems.      # Load the list of trusted subsystems.
135      my %subsystems = ();      my %subsystems = ();
136        # We only need it if load-only is NOT specified.
137        if (! $options->{loadOnly}) {
138      if (! defined $subsysFile || $subsysFile eq '') {      if (! defined $subsysFile || $subsysFile eq '') {
139          # Here we want all the subsystems.              # Here we want all the usable subsystems. First we get the whole list.
140          %subsystems = map { $_ => 1 } $fig->all_subsystems();              my @subs = $fig->all_subsystems();
141                # Loop through, checking for the NMPDR file.
142                for my $sub (@subs) {
143                    if ($fig->nmpdr_subsystem($sub)) {
144                        $subsystems{$sub} = 1;
145                    }
146                }
147      } else {      } else {
148          my $type = ref $subsysFile;          my $type = ref $subsysFile;
149          if ($type eq 'ARRAY') {          if ($type eq 'ARRAY') {
# Line 148  Line 163 
163              Confess("Invalid subsystem parameter in SproutLoad constructor.");              Confess("Invalid subsystem parameter in SproutLoad constructor.");
164          }          }
165      }      }
166            # 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                $name .= " " . join(" ", @{$classes});
172                $subsystems{$subsystem} = $name;
173            }
174        }
175      # Get the data directory from the Sprout object.      # Get the data directory from the Sprout object.
176      my ($directory) = $sprout->LoadInfo();      my ($directory) = $sprout->LoadInfo();
177      # Create the Sprout load object.      # Create the Sprout load object.
# Line 157  Line 181 
181                    subsystems => \%subsystems,                    subsystems => \%subsystems,
182                    sprout => $sprout,                    sprout => $sprout,
183                    loadDirectory => $directory,                    loadDirectory => $directory,
184                    erdb => $sprout->{_erdb},                    erdb => $sprout,
185                    loaders => []                    loaders => [],
186                      options => $options
187                   };                   };
188      # Bless and return it.      # Bless and return it.
189      bless $retVal, $class;      bless $retVal, $class;
190      return $retVal;      return $retVal;
191  }  }
192    
193    =head3 LoadOnly
194    
195    C<< my $flag = $spl->LoadOnly; >>
196    
197    Return TRUE if we are in load-only mode, else FALSE.
198    
199    =cut
200    
201    sub LoadOnly {
202        my ($self) = @_;
203        return $self->{options}->{loadOnly};
204    }
205    
206    =head3 PrimaryOnly
207    
208    C<< my $flag = $spl->PrimaryOnly; >>
209    
210    Return TRUE if only the main entity is to be loaded, else FALSE.
211    
212    =cut
213    
214    sub PrimaryOnly {
215        my ($self) = @_;
216        return $self->{options}->{primaryOnly};
217    }
218    
219  =head3 LoadGenomeData  =head3 LoadGenomeData
220    
221  C<< my $stats = $spl->LoadGenomeData(); >>  C<< my $stats = $spl->LoadGenomeData(); >>
# Line 192  Line 243 
243    
244  =back  =back
245    
 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.)  
   
246  =cut  =cut
247  #: Return Type $%;  #: Return Type $%;
248  sub LoadGenomeData {  sub LoadGenomeData {
# Line 210  Line 253 
253      # Get the genome count.      # Get the genome count.
254      my $genomeHash = $self->{genomes};      my $genomeHash = $self->{genomes};
255      my $genomeCount = (keys %{$genomeHash});      my $genomeCount = (keys %{$genomeHash});
     Trace("Beginning genome data load.") if T(2);  
256      # Create load objects for each of the tables we're loading.      # Create load objects for each of the tables we're loading.
257      my $loadGenome = $self->_TableLoader('Genome', $genomeCount);      my $loadGenome = $self->_TableLoader('Genome');
258      my $loadHasContig = $self->_TableLoader('HasContig', $genomeCount * 300);      my $loadHasContig = $self->_TableLoader('HasContig', $self->PrimaryOnly);
259      my $loadContig = $self->_TableLoader('Contig', $genomeCount * 300);      my $loadContig = $self->_TableLoader('Contig', $self->PrimaryOnly);
260      my $loadIsMadeUpOf = $self->_TableLoader('IsMadeUpOf', $genomeCount * 60000);      my $loadIsMadeUpOf = $self->_TableLoader('IsMadeUpOf', $self->PrimaryOnly);
261      my $loadSequence = $self->_TableLoader('Sequence', $genomeCount * 60000);      my $loadSequence = $self->_TableLoader('Sequence', $self->PrimaryOnly);
262        if ($self->{options}->{loadOnly}) {
263            Trace("Loading from existing files.") if T(2);
264        } else {
265            Trace("Generating genome data.") if T(2);
266      # Now we loop through the genomes, generating the data for each one.      # Now we loop through the genomes, generating the data for each one.
267      for my $genomeID (sort keys %{$genomeHash}) {      for my $genomeID (sort keys %{$genomeHash}) {
268          Trace("Loading data for genome $genomeID.") if T(3);              Trace("Generating data for genome $genomeID.") if T(3);
269          $loadGenome->Add("genomeIn");          $loadGenome->Add("genomeIn");
270          # The access code comes in via the genome hash.          # The access code comes in via the genome hash.
271          my $accessCode = $genomeHash->{$genomeID};          my $accessCode = $genomeHash->{$genomeID};
272          # 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.  
273          my ($genus, $species, @extraData) = split / /, $self->{fig}->genus_species($genomeID);          my ($genus, $species, @extraData) = split / /, $self->{fig}->genus_species($genomeID);
274          my $extra = join " ", @extraData, "[$genomeID]";              my $extra = join " ", @extraData;
275          # Get the full taxonomy.          # Get the full taxonomy.
276          my $taxonomy = $fig->taxonomy_of($genomeID);          my $taxonomy = $fig->taxonomy_of($genomeID);
277                # Open the NMPDR group file for this genome.
278                my $group;
279                if (open(TMP, "<$FIG_Config::organisms/$genomeID/NMPDR") &&
280                    defined($group = <TMP>)) {
281                    # Clean the line ending.
282                    chomp $group;
283                } else {
284                    # No group, so use the default.
285                    $group = $FIG_Config::otherGroup;
286                }
287                close TMP;
288          # Output the genome record.          # Output the genome record.
289          $loadGenome->Put($genomeID, $accessCode, $fig->is_complete($genomeID), $genus,          $loadGenome->Put($genomeID, $accessCode, $fig->is_complete($genomeID), $genus,
290                           $species, $extra, $taxonomy);                               $group, $species, $extra, $taxonomy);
291          # Now we loop through each of the genome's contigs.          # Now we loop through each of the genome's contigs.
292          my @contigs = $fig->all_contigs($genomeID);          my @contigs = $fig->all_contigs($genomeID);
293          for my $contigID (@contigs) {          for my $contigID (@contigs) {
# Line 262  Line 318 
318              }              }
319          }          }
320      }      }
321        }
322      # Finish the loads.      # Finish the loads.
323      my $retVal = $self->_FinishAll();      my $retVal = $self->_FinishAll();
324      # Return the result.      # Return the result.
# Line 302  Line 359 
359      my $fig = $self->{fig};      my $fig = $self->{fig};
360      # Get the genome hash.      # Get the genome hash.
361      my $genomeFilter = $self->{genomes};      my $genomeFilter = $self->{genomes};
362      my $genomeCount = (keys %{$genomeFilter});      # Set up an ID counter for the PCHs.
363      my $featureCount = $genomeCount * 4000;      my $pchID = 0;
364      # Start the loads.      # Start the loads.
365      my $loadCoupling = $self->_TableLoader('Coupling', $featureCount * $genomeCount);      my $loadCoupling = $self->_TableLoader('Coupling');
366      my $loadIsEvidencedBy = $self->_TableLoader('IsEvidencedBy', $featureCount * 8000);      my $loadIsEvidencedBy = $self->_TableLoader('IsEvidencedBy', $self->PrimaryOnly);
367      my $loadPCH = $self->_TableLoader('PCH', $featureCount * 2000);      my $loadPCH = $self->_TableLoader('PCH', $self->PrimaryOnly);
368      my $loadParticipatesInCoupling = $self->_TableLoader('ParticipatesInCoupling', $featureCount * 2000);      my $loadParticipatesInCoupling = $self->_TableLoader('ParticipatesInCoupling', $self->PrimaryOnly);
369      my $loadUsesAsEvidence = $self->_TableLoader('UsesAsEvidence', $featureCount * 8000);      my $loadUsesAsEvidence = $self->_TableLoader('UsesAsEvidence', $self->PrimaryOnly);
370      Trace("Beginning coupling data load.") if T(2);      if ($self->{options}->{loadOnly}) {
371            Trace("Loading from existing files.") if T(2);
372        } else {
373            Trace("Generating coupling data.") if T(2);
374      # Loop through the genomes found.      # Loop through the genomes found.
375      for my $genome (sort keys %{$genomeFilter}) {      for my $genome (sort keys %{$genomeFilter}) {
376          Trace("Generating coupling data for $genome.") if T(3);          Trace("Generating coupling data for $genome.") if T(3);
# Line 334  Line 394 
394              for my $coupleData (@couplings) {              for my $coupleData (@couplings) {
395                  my ($peg2, $score) = @{$coupleData};                  my ($peg2, $score) = @{$coupleData};
396                  # Compute the coupling ID.                  # Compute the coupling ID.
397                  my $coupleID = Sprout::CouplingID($peg1, $peg2);                      my $coupleID = $self->{erdb}->CouplingID($peg1, $peg2);
398                  if (! exists $dupHash{$coupleID}) {                  if (! exists $dupHash{$coupleID}) {
399                      $loadCoupling->Add("couplingIn");                      $loadCoupling->Add("couplingIn");
400                      # Here we have a new coupling to store in the load files.                      # Here we have a new coupling to store in the load files.
# Line 362  Line 422 
422                              # We store this evidence in the hash if the usage                              # We store this evidence in the hash if the usage
423                              # is nonzero or no prior evidence has been found. This                              # is nonzero or no prior evidence has been found. This
424                              # insures that if there is duplicate evidence, we                              # insures that if there is duplicate evidence, we
425                              # at least keep the meaningful ones. Only evidence is                                  # at least keep the meaningful ones. Only evidence in
426                              # the hash makes it to the output.                              # the hash makes it to the output.
427                              if ($usage || ! exists $evidenceMap{$evidenceKey}) {                              if ($usage || ! exists $evidenceMap{$evidenceKey}) {
428                                  $evidenceMap{$evidenceKey} = $evidenceData;                                  $evidenceMap{$evidenceKey} = $evidenceData;
# Line 370  Line 430 
430                          }                          }
431                      }                      }
432                      for my $evidenceID (keys %evidenceMap) {                      for my $evidenceID (keys %evidenceMap) {
433                                # Get the ID for this evidence.
434                                $pchID++;
435                          # Create the evidence record.                          # Create the evidence record.
436                          my ($peg3, $peg4, $usage) = @{$evidenceMap{$evidenceID}};                          my ($peg3, $peg4, $usage) = @{$evidenceMap{$evidenceID}};
437                          $loadPCH->Put($evidenceID, $usage);                              $loadPCH->Put($pchID, $usage);
438                          # Connect it to the coupling.                          # Connect it to the coupling.
439                          $loadIsEvidencedBy->Put($coupleID, $evidenceID);                              $loadIsEvidencedBy->Put($coupleID, $pchID);
440                          # Connect it to the features.                          # Connect it to the features.
441                          $loadUsesAsEvidence->Put($evidenceID, $peg3, 1);                              $loadUsesAsEvidence->Put($pchID, $peg3, 1);
442                          $loadUsesAsEvidence->Put($evidenceID, $peg4, 1);                              $loadUsesAsEvidence->Put($pchID, $peg4, 2);
443                            }
444                      }                      }
445                  }                  }
446              }              }
# Line 404  Line 467 
467      FeatureTranslation      FeatureTranslation
468      FeatureUpstream      FeatureUpstream
469      IsLocatedIn      IsLocatedIn
470        HasFeature
471        HasRoleInSubsystem
472        FeatureEssential
473        FeatureVirulent
474        FeatureIEDB
475    
476  =over 4  =over 4
477    
# Line 418  Line 486 
486  sub LoadFeatureData {  sub LoadFeatureData {
487      # Get this object instance.      # Get this object instance.
488      my ($self) = @_;      my ($self) = @_;
489      # Get the FIG object.      # Get the FIG and Sprout objects.
490      my $fig = $self->{fig};      my $fig = $self->{fig};
491        my $sprout = $self->{sprout};
492      # Get the table of genome IDs.      # Get the table of genome IDs.
493      my $genomeHash = $self->{genomes};      my $genomeHash = $self->{genomes};
     my $genomeCount = (keys %{$genomeHash});  
     my $featureCount = $genomeCount * 4000;  
494      # Create load objects for each of the tables we're loading.      # Create load objects for each of the tables we're loading.
495      my $loadFeature = $self->_TableLoader('Feature', $featureCount);      my $loadFeature = $self->_TableLoader('Feature');
496      my $loadFeatureAlias = $self->_TableLoader('FeatureAlias', $featureCount * 6);      my $loadIsLocatedIn = $self->_TableLoader('IsLocatedIn', $self->PrimaryOnly);
497      my $loadFeatureLink = $self->_TableLoader('FeatureLink', $featureCount * 10);      my $loadFeatureAlias = $self->_TableLoader('FeatureAlias');
498      my $loadFeatureTranslation = $self->_TableLoader('FeatureTranslation', $featureCount);      my $loadFeatureLink = $self->_TableLoader('FeatureLink');
499      my $loadFeatureUpstream = $self->_TableLoader('FeatureUpstream', $featureCount);      my $loadFeatureTranslation = $self->_TableLoader('FeatureTranslation');
500      my $loadIsLocatedIn = $self->_TableLoader('IsLocatedIn', $featureCount);      my $loadFeatureUpstream = $self->_TableLoader('FeatureUpstream');
501        my $loadHasFeature = $self->_TableLoader('HasFeature', $self->PrimaryOnly);
502        my $loadHasRoleInSubsystem = $self->_TableLoader('HasRoleInSubsystem', $self->PrimaryOnly);
503        my $loadFeatureEssential = $self->_TableLoader('FeatureEssential');
504        my $loadFeatureVirulent = $self->_TableLoader('FeatureVirulent');
505        my $loadFeatureIEDB = $self->_TableLoader('FeatureIEDB');
506        # Get the subsystem hash.
507        my $subHash = $self->{subsystems};
508      # 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
509      # locations.      # locations.
510      my $chunkSize = $self->{sprout}->MaxSegment();      my $chunkSize = $self->{sprout}->MaxSegment();
511      Trace("Beginning feature data load.") if T(2);      if ($self->{options}->{loadOnly}) {
512            Trace("Loading from existing files.") if T(2);
513        } else {
514            Trace("Generating feature data.") if T(2);
515      # Now we loop through the genomes, generating the data for each one.      # Now we loop through the genomes, generating the data for each one.
516      for my $genomeID (sort keys %{$genomeHash}) {      for my $genomeID (sort keys %{$genomeHash}) {
517          Trace("Loading features for genome $genomeID.") if T(3);          Trace("Loading features for genome $genomeID.") if T(3);
518          $loadFeature->Add("genomeIn");          $loadFeature->Add("genomeIn");
519          # Get the feature list for this genome.          # Get the feature list for this genome.
520          my $features = $fig->all_features_detailed($genomeID);          my $features = $fig->all_features_detailed($genomeID);
521                # Sort and count the list.
522                my @featureTuples = sort { $a->[0] cmp $b->[0] } @{$features};
523                my $count = scalar @featureTuples;
524                Trace("$count features found for genome $genomeID.") if T(3);
525                # Set up for our duplicate-feature check.
526                my $oldFeatureID = "";
527          # Loop through the features.          # Loop through the features.
528          for my $featureData (@{$features}) {              for my $featureTuple (@featureTuples) {
             $loadFeature->Add("featureIn");  
529              # Split the tuple.              # Split the tuple.
530              my ($featureID, $locations, $aliases, $type) = @{$featureData};                  my ($featureID, $locations, undef, $type) = @{$featureTuple};
531              # Create the feature record.                  # Check for duplicates.
532              $loadFeature->Put($featureID, 1, $type);                  if ($featureID eq $oldFeatureID) {
533                        Trace("Duplicate feature $featureID found.") if T(1);
534                    } else {
535                        $oldFeatureID = $featureID;
536                        # Count this feature.
537                        $loadFeature->Add("featureIn");
538                        # Begin building the keywords. We start with the genome ID, the
539                        # feature ID, and the organism name.
540                        my @keywords = ($genomeID, $featureID, $fig->genus_species($genomeID));
541                        # Get the functional assignment and aliases. This
542                        # depends on the feature type.
543                        my $assignment;
544                        if ($type eq "peg") {
545                            $assignment = $fig->function_of($featureID);
546              # Create the aliases.              # Create the aliases.
547              for my $alias (split /\s*,\s*/, $aliases) {                          for my $alias ($fig->feature_aliases($featureID)) {
548                  $loadFeatureAlias->Put($featureID, $alias);                  $loadFeatureAlias->Put($featureID, $alias);
549                                push @keywords, $alias;
550              }              }
551                        } else {
552                            # For other types, the assignment is the first (and ONLY) alias.
553                            ($assignment) = $fig->feature_aliases($featureID);
554                        }
555                        Trace("Assignment for $featureID is: $assignment") if T(4);
556                        # Break the assignment into words and shove it onto the
557                        # keyword list.
558                        push @keywords, split(/\s+/, $assignment);
559                        # Link this feature to the parent genome.
560                        $loadHasFeature->Put($genomeID, $featureID, $type);
561              # Get the links.              # Get the links.
562              my @links = $fig->fid_links($featureID);              my @links = $fig->fid_links($featureID);
563              for my $link (@links) {              for my $link (@links) {
# Line 470  Line 576 
576                      $loadFeatureUpstream->Put($featureID, $upstream);                      $loadFeatureUpstream->Put($featureID, $upstream);
577                  }                  }
578              }              }
579                        # Now we need to find the subsystems this feature participates in.
580                        # We also add the subsystems to the keyword list. Before we do that,
581                        # we must convert underscores to spaces and tack on the classifications.
582                        my @subsystems = $fig->peg_to_subsystems($featureID);
583                        for my $subsystem (@subsystems) {
584                            # Only proceed if we like this subsystem.
585                            if (exists $subHash->{$subsystem}) {
586                                # Store the has-role link.
587                                $loadHasRoleInSubsystem->Put($featureID, $subsystem, $genomeID, $type);
588                                # Save the subsystem's keyword data.
589                                my $subKeywords = $subHash->{$subsystem};
590                                push @keywords, split /\s+/, $subKeywords;
591                                # Now we need to get this feature's role in the subsystem.
592                                my $subObject = $fig->get_subsystem($subsystem);
593                                my @roleColumns = $subObject->get_peg_roles($featureID);
594                                my @allRoles = $subObject->get_roles();
595                                for my $col (@roleColumns) {
596                                    my $role = $allRoles[$col];
597                                    push @keywords, split /\s+/, $role;
598                                    push @keywords, $subObject->get_role_abbr($col);
599                                }
600                            }
601                        }
602                        # There are three special attributes computed from property
603                        # data that we build next. If the special attribute is non-empty,
604                        # its name will be added to the keyword list. First, we get all
605                        # the attributes for this feature. They will come back as
606                        # 4-tuples: [peg, name, value, URL]. We use a 3-tuple instead:
607                        # [name, value, value with URL]. (We don't need the PEG, since
608                        # we already know it.)
609                        my @attributes = map { [$_->[1], $_->[2], Tracer::CombineURL($_->[2], $_->[3])] }
610                                             $fig->get_attributes($featureID);
611                        # Now we process each of the special attributes.
612                        if (SpecialAttribute($featureID, \@attributes,
613                                             1, 2, '^(essential|potential_essential)$',
614                                             $loadFeatureEssential)) {
615                            push @keywords, 'essential';
616                            $loadFeature->Add('essential');
617                        }
618                        if (SpecialAttribute($featureID, \@attributes,
619                                             0, 2, '^virulen',
620                                             $loadFeatureVirulent)) {
621                            push @keywords, 'virulent';
622                            $loadFeature->Add('virulent');
623                        }
624                        if (SpecialAttribute($featureID, \@attributes,
625                                             0, 2, '^iedb_',
626                                             $loadFeatureIEDB)) {
627                            push @keywords, 'iedb';
628                            $loadFeature->Add('iedb');
629                        }
630                        # Now we need to bust up hyphenated words in the keyword
631                        # list.
632                        my $keywordString = "";
633                        for my $keyword (@keywords) {
634                            if (length $keyword >= 4) {
635                                $keywordString .= " $keyword";
636                                if ($keyword =~ /-/) {
637                                    my @words = grep { length($_) >= 4 } split /-/, $keyword;
638                                    $keywordString .= join(" ", "", @words);
639                                }
640                            }
641                        }
642                        # Clean the keyword list.
643                        my $cleanWords = $sprout->CleanKeywords($keywordString);
644                        Trace("Keyword string for $featureID: $cleanWords") if T(4);
645                        # Create the feature record.
646                        $loadFeature->Put($featureID, 1, $type, $assignment, $cleanWords);
647              # 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
648              # 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
649              # the maximum segment size. This simplifies the genes_in_region processing              # the maximum segment size. This simplifies the genes_in_region processing
650              # for Sprout.              # for Sprout.
651              my @locationList = split /\s*,\s*/, $locations;              my @locationList = split /\s*,\s*/, $locations;
652                        # Create the location position indicator.
653                        my $i = 1;
654              # Loop through the locations.              # Loop through the locations.
655              for my $location (@locationList) {              for my $location (@locationList) {
656                  # Parse the location.                  # Parse the location.
657                  my $locObject = BasicLocation->new($location);                          my $locObject = BasicLocation->new("$genomeID:$location");
658                  # Split it into a list of chunks.                  # Split it into a list of chunks.
659                  my @locOList = ();                  my @locOList = ();
660                  while (my $peeling = $locObject->Peel($chunkSize)) {                  while (my $peeling = $locObject->Peel($chunkSize)) {
# Line 488  Line 664 
664                  push @locOList, $locObject;                  push @locOList, $locObject;
665                  # Loop through the chunks, creating IsLocatedIn records. The variable                  # Loop through the chunks, creating IsLocatedIn records. The variable
666                  # "$i" will be used to keep the location index.                  # "$i" will be used to keep the location index.
                 my $i = 1;  
667                  for my $locChunk (@locOList) {                  for my $locChunk (@locOList) {
668                      $loadIsLocatedIn->Put($featureID, $locChunk->Contig, $locChunk->Left,                      $loadIsLocatedIn->Put($featureID, $locChunk->Contig, $locChunk->Left,
669                                            $locChunk->Dir, $locChunk->Length, $i);                                            $locChunk->Dir, $locChunk->Length, $i);
# Line 497  Line 672 
672              }              }
673          }          }
674      }      }
     # 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);  
                 }  
             }  
675          }          }
676      }      }
677      # Finish the loads.      # Finish the loads.
# Line 584  Line 694 
694  The following relations are loaded by this method.  The following relations are loaded by this method.
695    
696      Subsystem      Subsystem
697        SubsystemClass
698      Role      Role
699        RoleEC
700      SSCell      SSCell
701      ContainsFeature      ContainsFeature
702      IsGenomeOf      IsGenomeOf
# Line 592  Line 704 
704      OccursInSubsystem      OccursInSubsystem
705      ParticipatesIn      ParticipatesIn
706      HasSSCell      HasSSCell
707        ConsistsOfRoles
708        RoleSubset
709        HasRoleSubset
710        ConsistsOfGenomes
711        GenomeSubset
712        HasGenomeSubset
713        Catalyzes
714        Diagram
715        RoleOccursIn
716    
717  =over 4  =over 4
718    
# Line 601  Line 722 
722    
723  =back  =back
724    
 B<TO DO>  
   
 Generate RoleName table?  
   
725  =cut  =cut
726  #: Return Type $%;  #: Return Type $%;
727  sub LoadSubsystemData {  sub LoadSubsystemData {
# Line 618  Line 735 
735      # Get the subsystem hash. This lists the subsystems we'll process.      # Get the subsystem hash. This lists the subsystems we'll process.
736      my $subsysHash = $self->{subsystems};      my $subsysHash = $self->{subsystems};
737      my @subsysIDs = sort keys %{$subsysHash};      my @subsysIDs = sort keys %{$subsysHash};
738      my $subsysCount = @subsysIDs;      # Get the map list.
739      my $genomeCount = (keys %{$genomeHash});      my @maps = $fig->all_maps;
     my $featureCount = $genomeCount * 4000;  
740      # Create load objects for each of the tables we're loading.      # Create load objects for each of the tables we're loading.
741      my $loadSubsystem = $self->_TableLoader('Subsystem', $subsysCount);      my $loadDiagram = $self->_TableLoader('Diagram', $self->PrimaryOnly);
742      my $loadRole = $self->_TableLoader('Role', $featureCount * 6);      my $loadRoleOccursIn = $self->_TableLoader('RoleOccursIn', $self->PrimaryOnly);
743      my $loadSSCell = $self->_TableLoader('SSCell', $featureCount * $genomeCount);      my $loadSubsystem = $self->_TableLoader('Subsystem');
744      my $loadContainsFeature = $self->_TableLoader('ContainsFeature', $featureCount * $subsysCount);      my $loadRole = $self->_TableLoader('Role', $self->PrimaryOnly);
745      my $loadIsGenomeOf = $self->_TableLoader('IsGenomeOf', $featureCount * $genomeCount);      my $loadRoleEC = $self->_TableLoader('RoleEC', $self->PrimaryOnly);
746      my $loadIsRoleOf = $self->_TableLoader('IsRoleOf', $featureCount * $genomeCount);      my $loadCatalyzes = $self->_TableLoader('Catalyzes', $self->PrimaryOnly);
747      my $loadOccursInSubsystem = $self->_TableLoader('OccursInSubsystem', $featureCount * 6);      my $loadSSCell = $self->_TableLoader('SSCell', $self->PrimaryOnly);
748      my $loadParticipatesIn = $self->_TableLoader('ParticipatesIn', $subsysCount * $genomeCount);      my $loadContainsFeature = $self->_TableLoader('ContainsFeature', $self->PrimaryOnly);
749      my $loadHasSSCell = $self->_TableLoader('HasSSCell', $featureCount * $genomeCount);      my $loadIsGenomeOf = $self->_TableLoader('IsGenomeOf', $self->PrimaryOnly);
750      Trace("Beginning subsystem data load.") if T(2);      my $loadIsRoleOf = $self->_TableLoader('IsRoleOf', $self->PrimaryOnly);
751        my $loadOccursInSubsystem = $self->_TableLoader('OccursInSubsystem', $self->PrimaryOnly);
752        my $loadParticipatesIn = $self->_TableLoader('ParticipatesIn', $self->PrimaryOnly);
753        my $loadHasSSCell = $self->_TableLoader('HasSSCell', $self->PrimaryOnly);
754        my $loadRoleSubset = $self->_TableLoader('RoleSubset', $self->PrimaryOnly);
755        my $loadGenomeSubset = $self->_TableLoader('GenomeSubset', $self->PrimaryOnly);
756        my $loadConsistsOfRoles = $self->_TableLoader('ConsistsOfRoles', $self->PrimaryOnly);
757        my $loadConsistsOfGenomes = $self->_TableLoader('ConsistsOfGenomes', $self->PrimaryOnly);
758        my $loadHasRoleSubset = $self->_TableLoader('HasRoleSubset', $self->PrimaryOnly);
759        my $loadHasGenomeSubset = $self->_TableLoader('HasGenomeSubset', $self->PrimaryOnly);
760        my $loadSubsystemClass = $self->_TableLoader('SubsystemClass', $self->PrimaryOnly);
761        if ($self->{options}->{loadOnly}) {
762            Trace("Loading from existing files.") if T(2);
763        } else {
764            Trace("Generating subsystem data.") if T(2);
765            # This hash will contain the role for each EC. When we're done, this
766            # information will be used to generate the Catalyzes table.
767            my %ecToRoles = ();
768      # Loop through the subsystems. Our first task will be to create the      # Loop through the subsystems. Our first task will be to create the
769      # roles. We do this by looping through the subsystems and creating a      # roles. We do this by looping through the subsystems and creating a
770      # 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
771      # duplicates. As we move along, we'll connect the roles and subsystems.          # duplicates. As we move along, we'll connect the roles and subsystems
772            # and memorize up the reactions.
773            my ($genomeID, $roleID);
774      my %roleData = ();      my %roleData = ();
775      for my $subsysID (@subsysIDs) {      for my $subsysID (@subsysIDs) {
776                # Get the subsystem object.
777                my $sub = $fig->get_subsystem($subsysID);
778                # Only proceed if the subsystem has a spreadsheet.
779                if (! $sub->{empty_ss}) {
780          Trace("Creating subsystem $subsysID.") if T(3);          Trace("Creating subsystem $subsysID.") if T(3);
781          $loadSubsystem->Add("subsystemIn");          $loadSubsystem->Add("subsystemIn");
782          # Create the subsystem record.          # Create the subsystem record.
783          $loadSubsystem->Put($subsysID);                  my $curator = $sub->get_curator();
784          # Get the subsystem's roles.                  my $notes = $sub->get_notes();
785          my @roles = $fig->subsystem_to_roles($subsysID);                  $loadSubsystem->Put($subsysID, $curator, $notes);
786          # Connect the roles to the subsystem. If a role is new, we create                  # Now for the classification string. This comes back as a list
787          # a role record for it.                  # reference and we convert it to a space-delimited string.
788          for my $roleID (@roles) {                  my $classList = $fig->subsystem_classification($subsysID);
789                    my $classString = join(" : ", grep { $_ } @$classList);
790                    $loadSubsystemClass->Put($subsysID, $classString);
791                    # Connect it to its roles. Each role is a column in the subsystem spreadsheet.
792                    for (my $col = 0; defined($roleID = $sub->get_role($col)); $col++) {
793                        # Connect to this role.
794              $loadOccursInSubsystem->Add("roleIn");              $loadOccursInSubsystem->Add("roleIn");
795              $loadOccursInSubsystem->Put($roleID, $subsysID);                      $loadOccursInSubsystem->Put($roleID, $subsysID, $col);
796                        # If it's a new role, add it to the role table.
797              if (! exists $roleData{$roleID}) {              if (! exists $roleData{$roleID}) {
798                  $loadRole->Put($roleID);                          # Get the role's abbreviation.
799                            my $abbr = $sub->get_role_abbr($col);
800                            # Add the role.
801                            $loadRole->Put($roleID, $abbr);
802                  $roleData{$roleID} = 1;                  $roleData{$roleID} = 1;
803                            # Check for an EC number.
804                            if ($roleID =~ /\(EC ([^.]+\.[^.]+\.[^.]+\.[^)]+)\)\s*$/) {
805                                my $ec = $1;
806                                $loadRoleEC->Put($roleID, $ec);
807                                $ecToRoles{$ec} = $roleID;
808              }              }
809          }          }
810          # Now all roles for this subsystem have been filled in. We create the                  }
811          # spreadsheet by matches roles to genomes. To do this, we need to                  # Now we create the spreadsheet for the subsystem by matching roles to
812          # get the genomes on the sheet.                  # genomes. Each genome is a row and each role is a column. We may need
813                    # to actually create the roles as we find them.
814          Trace("Creating subsystem $subsysID spreadsheet.") if T(3);          Trace("Creating subsystem $subsysID spreadsheet.") if T(3);
815          my @genomes = map { $_->[0] } @{$fig->subsystem_genomes($subsysID)};                  for (my $row = 0; defined($genomeID = $sub->get_genome($row)); $row++) {
816          for my $genomeID (@genomes) {                      # Only proceed if this is one of our genomes.
             # Only process this genome if it's one of ours.  
817              if (exists $genomeHash->{$genomeID}) {              if (exists $genomeHash->{$genomeID}) {
818                  # Connect the genome to the subsystem.                          # Count the PEGs and cells found for verification purposes.
819                  $loadParticipatesIn->Put($genomeID, $subsysID);                          my $pegCount = 0;
820                            my $cellCount = 0;
821                            # Create a list for the PEGs we find. This list will be used
822                            # to generate cluster numbers.
823                            my @pegsFound = ();
824                            # Create a hash that maps spreadsheet IDs to PEGs. We will
825                            # use this to generate the ContainsFeature data after we have
826                            # the cluster numbers.
827                            my %cellPegs = ();
828                            # Get the genome's variant code for this subsystem.
829                            my $variantCode = $sub->get_variant_code($row);
830                  # 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
831                  # part of the spreadsheet cell ID.                  # part of the spreadsheet cell ID.
832                  for (my $i = 0; $i <= $#roles; $i++) {                          for (my $col = 0; defined($roleID = $sub->get_role($col)); $col++) {
                     my $role = $roles[$i];  
833                      # Get the features in the spreadsheet cell for this genome and role.                      # Get the features in the spreadsheet cell for this genome and role.
834                      my @pegs = $fig->pegs_in_subsystem_cell($subsysID, $genomeID, $i);                              my @pegs = grep { !$fig->is_deleted_fid($_) } $sub->get_pegs_from_cell($row, $col);
835                      # Only proceed if features exist.                      # Only proceed if features exist.
836                      if (@pegs > 0) {                      if (@pegs > 0) {
837                          # Create the spreadsheet cell.                          # Create the spreadsheet cell.
838                          my $cellID = "$subsysID:$genomeID:$i";                                  $cellCount++;
839                                    my $cellID = "$subsysID:$genomeID:$col";
840                          $loadSSCell->Put($cellID);                          $loadSSCell->Put($cellID);
841                          $loadIsGenomeOf->Put($genomeID, $cellID);                          $loadIsGenomeOf->Put($genomeID, $cellID);
842                          $loadIsRoleOf->Put($role, $cellID);                                  $loadIsRoleOf->Put($roleID, $cellID);
843                          $loadHasSSCell->Put($subsysID, $cellID);                          $loadHasSSCell->Put($subsysID, $cellID);
844                          # Attach the features to it.                                  # Remember its features.
845                          for my $pegID (@pegs) {                                  push @pegsFound, @pegs;
846                              $loadContainsFeature->Put($cellID, $pegID);                                  $cellPegs{$cellID} = \@pegs;
847                                    $pegCount += @pegs;
848                                }
849                            }
850                            # If we found some cells for this genome, we need to compute clusters and
851                            # denote it participates in the subsystem.
852                            if ($pegCount > 0) {
853                                Trace("$pegCount PEGs in $cellCount cells for $genomeID.") if T(3);
854                                $loadParticipatesIn->Put($genomeID, $subsysID, $variantCode);
855                                # Create a hash mapping PEG IDs to cluster numbers.
856                                # We default to -1 for all of them.
857                                my %clusterOf = map { $_ => -1 } @pegsFound;
858                                # Partition the PEGs found into clusters.
859                                my @clusters = $fig->compute_clusters([keys %clusterOf], $sub);
860                                for (my $i = 0; $i <= $#clusters; $i++) {
861                                    my $subList = $clusters[$i];
862                                    for my $peg (@{$subList}) {
863                                        $clusterOf{$peg} = $i;
864                                    }
865                                }
866                                # Create the ContainsFeature data.
867                                for my $cellID (keys %cellPegs) {
868                                    my $cellList = $cellPegs{$cellID};
869                                    for my $cellPeg (@$cellList) {
870                                        $loadContainsFeature->Put($cellID, $cellPeg, $clusterOf{$cellPeg});
871                          }                          }
872                      }                      }
873                  }                  }
874              }              }
875          }          }
876                    # Now we need to generate the subsets. The subset names must be concatenated to
877                    # the subsystem name to make them unique keys. There are two types of subsets:
878                    # genome subsets and role subsets. We do the role subsets first.
879                    my @subsetNames = $sub->get_subset_names();
880                    for my $subsetID (@subsetNames) {
881                        # Create the subset record.
882                        my $actualID = "$subsysID:$subsetID";
883                        $loadRoleSubset->Put($actualID);
884                        # Connect the subset to the subsystem.
885                        $loadHasRoleSubset->Put($subsysID, $actualID);
886                        # Connect the subset to its roles.
887                        my @roles = $sub->get_subsetC_roles($subsetID);
888                        for my $roleID (@roles) {
889                            $loadConsistsOfRoles->Put($actualID, $roleID);
890      }      }
     # Finish the load.  
     my $retVal = $self->_FinishAll();  
     return $retVal;  
891  }  }
892                    # Next the genome subsets.
893  =head3 LoadDiagramData                  @subsetNames = $sub->get_subset_namesR();
894                    for my $subsetID (@subsetNames) {
895  C<< my $stats = $spl->LoadDiagramData(); >>                      # Create the subset record.
896                        my $actualID = "$subsysID:$subsetID";
897  Load the diagram data from FIG into Sprout.                      $loadGenomeSubset->Put($actualID);
898                        # Connect the subset to the subsystem.
899  Diagrams are used to organize functional roles. The diagram shows the                      $loadHasGenomeSubset->Put($subsysID, $actualID);
900  connections between chemicals that interact with a subsystem.                      # Connect the subset to its genomes.
901                        my @genomes = $sub->get_subsetR($subsetID);
902  The following relations are loaded by this method.                      for my $genomeID (@genomes) {
903                            $loadConsistsOfGenomes->Put($actualID, $genomeID);
904      Diagram                      }
905      RoleOccursIn                  }
906                }
907  =over 4          }
908            # Now we loop through the diagrams. We need to create the diagram records
909  =item RETURNS          # and link each diagram to its roles. Note that only roles which occur
910            # in subsystems (and therefore appear in the %ecToRoles hash) are
911  Returns a statistics object for the loads.          # included.
912            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) {  
913          Trace("Loading diagram $map.") if T(3);          Trace("Loading diagram $map.") if T(3);
914          # Get the diagram's descriptive name.          # Get the diagram's descriptive name.
915          my $name = $fig->map_name($map);          my $name = $fig->map_name($map);
# Line 740  Line 918 
918          # A hash is used to prevent duplicates.          # A hash is used to prevent duplicates.
919          my %roleHash = ();          my %roleHash = ();
920          for my $role ($fig->map_to_ecs($map)) {          for my $role ($fig->map_to_ecs($map)) {
921              if (! $roleHash{$role}) {                  if (exists $ecToRoles{$role} && ! $roleHash{$role}) {
922                  $loadRoleOccursIn->Put($role, $map);                      $loadRoleOccursIn->Put($ecToRoles{$role}, $map);
923                  $roleHash{$role} = 1;                  $roleHash{$role} = 1;
924              }              }
925          }          }
926      }      }
927            # Before we leave, we must create the Catalyzes table. We start with the reactions,
928            # then use the "ecToRoles" table to convert EC numbers to role IDs.
929            my @reactions = $fig->all_reactions();
930            for my $reactionID (@reactions) {
931                # Get this reaction's list of roles. The results will be EC numbers.
932                my @roles = $fig->catalyzed_by($reactionID);
933                # Loop through the roles, creating catalyzation records.
934                for my $thisRole (@roles) {
935                    if (exists $ecToRoles{$thisRole}) {
936                        $loadCatalyzes->Put($ecToRoles{$thisRole}, $reactionID);
937                    }
938                }
939            }
940        }
941      # Finish the load.      # Finish the load.
942      my $retVal = $self->_FinishAll();      my $retVal = $self->_FinishAll();
943      return $retVal;      return $retVal;
# Line 787  Line 979 
979      my $fig = $self->{fig};      my $fig = $self->{fig};
980      # Get the genome hash.      # Get the genome hash.
981      my $genomeHash = $self->{genomes};      my $genomeHash = $self->{genomes};
     my $genomeCount = (keys %{$genomeHash});  
982      # Create load objects for each of the tables we're loading.      # Create load objects for each of the tables we're loading.
983      my $loadProperty = $self->_TableLoader('Property', $genomeCount * 1500);      my $loadProperty = $self->_TableLoader('Property');
984      my $loadHasProperty = $self->_TableLoader('HasProperty', $genomeCount * 1500);      my $loadHasProperty = $self->_TableLoader('HasProperty', $self->PrimaryOnly);
985      Trace("Beginning property data load.") if T(2);      if ($self->{options}->{loadOnly}) {
986            Trace("Loading from existing files.") if T(2);
987        } else {
988            Trace("Generating property data.") if T(2);
989      # Create a hash for storing property IDs.      # Create a hash for storing property IDs.
990      my %propertyKeys = ();      my %propertyKeys = ();
991      my $nextID = 1;      my $nextID = 1;
992      # Loop through the genomes.      # Loop through the genomes.
993      for my $genomeID (keys %{$genomeHash}) {          for my $genomeID (sort keys %{$genomeHash}) {
994          $loadProperty->Add("genomeIn");          $loadProperty->Add("genomeIn");
995                Trace("Generating properties for $genomeID.") if T(3);
996          # 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
997          # tuples returned by "all_features_detailed". We use "all_features_detailed"          # tuples returned by "all_features_detailed". We use "all_features_detailed"
998          # rather than "all_features" because we want all features regardless of type.          # rather than "all_features" because we want all features regardless of type.
999          my @features = map { $_->[0] } @{$fig->all_features_detailed($genomeID)};          my @features = map { $_->[0] } @{$fig->all_features_detailed($genomeID)};
1000                my $featureCount = 0;
1001                my $propertyCount = 0;
1002          # Loop through the features, creating HasProperty records.          # Loop through the features, creating HasProperty records.
1003          for my $fid (@features) {          for my $fid (@features) {
             $loadProperty->Add("featureIn");  
1004              # 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
1005              # to insure we do not get any genome attributes.              # to insure we do not get any genome attributes.
1006              my @attributeList = $fig->get_attributes($fid, '', '', '');              my @attributeList = $fig->get_attributes($fid, '', '', '');
1007                    # Add essentiality and virulence attributes.
1008                    if ($fig->essential($fid)) {
1009                        push @attributeList, [$fid, 'essential', 1, ''];
1010                    }
1011                    if ($fig->virulent($fid)) {
1012                        push @attributeList, [$fid, 'virulent', 1, ''];
1013                    }
1014                    if (scalar @attributeList) {
1015                        $featureCount++;
1016                    }
1017              # Loop through the attributes.              # Loop through the attributes.
1018              for my $tuple (@attributeList) {              for my $tuple (@attributeList) {
1019                        $propertyCount++;
1020                  # Get this attribute value's data. Note that we throw away the FID,                  # Get this attribute value's data. Note that we throw away the FID,
1021                  # since it will always be the same as the value if "$fid".                  # since it will always be the same as the value if "$fid".
1022                  my (undef, $key, $value, $url) = @{$tuple};                  my (undef, $key, $value, $url) = @{$tuple};
# Line 831  Line 1038 
1038                  $loadHasProperty->Put($fid, $propertyID, $url);                  $loadHasProperty->Put($fid, $propertyID, $url);
1039              }              }
1040          }          }
1041                # Update the statistics.
1042                Trace("$propertyCount attributes processed for $featureCount features.") if T(3);
1043                $loadHasProperty->Add("featuresIn", $featureCount);
1044                $loadHasProperty->Add("propertiesIn", $propertyCount);
1045            }
1046      }      }
1047      # Finish the load.      # Finish the load.
1048      my $retVal = $self->_FinishAll();      my $retVal = $self->_FinishAll();
# Line 871  Line 1083 
1083      my $fig = $self->{fig};      my $fig = $self->{fig};
1084      # Get the genome hash.      # Get the genome hash.
1085      my $genomeHash = $self->{genomes};      my $genomeHash = $self->{genomes};
     my $genomeCount = (keys %{$genomeHash});  
1086      # Create load objects for each of the tables we're loading.      # Create load objects for each of the tables we're loading.
1087      my $loadAnnotation = $self->_TableLoader('Annotation', $genomeCount * 4000);      my $loadAnnotation = $self->_TableLoader('Annotation');
1088      my $loadIsTargetOfAnnotation = $self->_TableLoader('IsTargetOfAnnotation', $genomeCount * 4000);      my $loadIsTargetOfAnnotation = $self->_TableLoader('IsTargetOfAnnotation', $self->PrimaryOnly);
1089      my $loadSproutUser = $self->_TableLoader('SproutUser', 100);      my $loadSproutUser = $self->_TableLoader('SproutUser', $self->PrimaryOnly);
1090      my $loadUserAccess = $self->_TableLoader('UserAccess', 1000);      my $loadUserAccess = $self->_TableLoader('UserAccess', $self->PrimaryOnly);
1091      my $loadMadeAnnotation = $self->_TableLoader('MadeAnnotation', $genomeCount * 4000);      my $loadMadeAnnotation = $self->_TableLoader('MadeAnnotation', $self->PrimaryOnly);
1092      Trace("Beginning annotation data load.") if T(2);      if ($self->{options}->{loadOnly}) {
1093            Trace("Loading from existing files.") if T(2);
1094        } else {
1095            Trace("Generating annotation data.") if T(2);
1096      # 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
1097      # user records.      # user records.
1098      my %users = ( FIG => 1, master => 1 );      my %users = ( FIG => 1, master => 1 );
# Line 892  Line 1106 
1106      # Loop through the genomes.      # Loop through the genomes.
1107      for my $genomeID (sort keys %{$genomeHash}) {      for my $genomeID (sort keys %{$genomeHash}) {
1108          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);  
1109              # 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
1110              # from showing up for a single PEG's annotations.              # from showing up for a single PEG's annotations.
1111              my %seenTimestamps = ();              my %seenTimestamps = ();
1112              # Check for a functional assignment.              # Get the genome's annotations.
1113              my $func = $fig->function_of($peg);              my @annotations = $fig->read_all_annotations($genomeID);
1114              if ($func) {              Trace("Processing annotations.") if T(2);
1115                  # If this is NOT a hypothetical assignment, we create an              for my $tuple (@annotations) {
1116                  # assignment annotation for it.                  # Get the annotation tuple.
1117                  if (! FIG::hypo($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};  
1118                      # Here we fix up the annotation text. "\r" is removed,                      # Here we fix up the annotation text. "\r" is removed,
1119                      # and "\t" and "\n" are escaped. Note we use the "s"                  # and "\t" and "\n" are escaped. Note we use the "gs"
1120                      # modifier so that new-lines inside the text do not                      # modifier so that new-lines inside the text do not
1121                      # stop the substitution search.                      # stop the substitution search.
1122                      $text =~ s/\r//gs;                      $text =~ s/\r//gs;
# Line 927  Line 1126 
1126                      $text =~ s/Set master function/Set FIG function/s;                      $text =~ s/Set master function/Set FIG function/s;
1127                      # Insure the time stamp is valid.                      # Insure the time stamp is valid.
1128                      if ($timestamp =~ /^\d+$/) {                      if ($timestamp =~ /^\d+$/) {
1129                          # Here it's a number. We need to insure it's unique.                      # Here it's a number. We need to insure the one we use to form
1130                          while ($seenTimestamps{$timestamp}) {                      # the key is unique.
1131                              $timestamp++;                      my $keyStamp = $timestamp;
1132                        while ($seenTimestamps{"$peg:$keyStamp"}) {
1133                            $keyStamp++;
1134                          }                          }
1135                          $seenTimestamps{$timestamp} = 1;                      my $annotationID = "$peg:$keyStamp";
1136                          my $annotationID = "$peg:$timestamp";                      $seenTimestamps{$annotationID} = 1;
1137                          # Insure the user exists.                          # Insure the user exists.
1138                          if (! $users{$user}) {                          if (! $users{$user}) {
1139                              $loadSproutUser->Put($user, "SEED user");                              $loadSproutUser->Put($user, "SEED user");
# Line 940  Line 1141 
1141                              $users{$user} = 1;                              $users{$user} = 1;
1142                          }                          }
1143                          # Generate the annotation.                          # Generate the annotation.
1144                          $loadAnnotation->Put($annotationID, $timestamp, "$user\\n$text");                      $loadAnnotation->Put($annotationID, $timestamp, $text);
1145                          $loadIsTargetOfAnnotation->Put($peg, $annotationID);                          $loadIsTargetOfAnnotation->Put($peg, $annotationID);
1146                          $loadMadeAnnotation->Put($user, $annotationID);                          $loadMadeAnnotation->Put($user, $annotationID);
1147                      } else {                      } else {
# Line 950  Line 1151 
1151                  }                  }
1152              }              }
1153          }          }
     }  
1154      # Finish the load.      # Finish the load.
1155      my $retVal = $self->_FinishAll();      my $retVal = $self->_FinishAll();
1156      return $retVal;      return $retVal;
# Line 991  Line 1191 
1191      my $fig = $self->{fig};      my $fig = $self->{fig};
1192      # Get the genome hash.      # Get the genome hash.
1193      my $genomeHash = $self->{genomes};      my $genomeHash = $self->{genomes};
     my $genomeCount = (keys %{$genomeHash});  
1194      # Create load objects for each of the tables we're loading.      # Create load objects for each of the tables we're loading.
1195      my $loadComesFrom = $self->_TableLoader('ComesFrom', $genomeCount * 4);      my $loadComesFrom = $self->_TableLoader('ComesFrom', $self->PrimaryOnly);
1196      my $loadSource = $self->_TableLoader('Source', $genomeCount * 4);      my $loadSource = $self->_TableLoader('Source');
1197      my $loadSourceURL = $self->_TableLoader('SourceURL', $genomeCount * 8);      my $loadSourceURL = $self->_TableLoader('SourceURL');
1198      Trace("Beginning source data load.") if T(2);      if ($self->{options}->{loadOnly}) {
1199            Trace("Loading from existing files.") if T(2);
1200        } else {
1201            Trace("Generating annotation data.") if T(2);
1202      # Create hashes to collect the Source information.      # Create hashes to collect the Source information.
1203      my %sourceURL = ();      my %sourceURL = ();
1204      my %sourceDesc = ();      my %sourceDesc = ();
# Line 1010  Line 1212 
1212              chomp $line;              chomp $line;
1213              my($sourceID, $desc, $url) = split(/\t/,$line);              my($sourceID, $desc, $url) = split(/\t/,$line);
1214              $loadComesFrom->Put($genomeID, $sourceID);              $loadComesFrom->Put($genomeID, $sourceID);
1215              if ($url && ! exists $sourceURL{$genomeID}) {                  if ($url && ! exists $sourceURL{$sourceID}) {
1216                  $loadSourceURL->Put($sourceID, $url);                  $loadSourceURL->Put($sourceID, $url);
1217                  $sourceURL{$sourceID} = 1;                  $sourceURL{$sourceID} = 1;
1218              }              }
1219              if ($desc && ! exists $sourceDesc{$sourceID}) {                  if ($desc) {
1220                  $loadSource->Put($sourceID, $desc);                      $sourceDesc{$sourceID} = $desc;
1221                  $sourceDesc{$sourceID} = 1;                  } elsif (! exists $sourceDesc{$sourceID}) {
1222                        $sourceDesc{$sourceID} = $sourceID;
1223              }              }
1224          }          }
1225          close TMP;          close TMP;
1226      }      }
1227            # Write the source descriptions.
1228            for my $sourceID (keys %sourceDesc) {
1229                $loadSource->Put($sourceID, $sourceDesc{$sourceID});
1230            }
1231        }
1232      # Finish the load.      # Finish the load.
1233      my $retVal = $self->_FinishAll();      my $retVal = $self->_FinishAll();
1234      return $retVal;      return $retVal;
# Line 1060  Line 1268 
1268      my $fig = $self->{fig};      my $fig = $self->{fig};
1269      # Get the genome hash.      # Get the genome hash.
1270      my $genomeHash = $self->{genomes};      my $genomeHash = $self->{genomes};
     my $genomeCount = (keys %{$genomeHash});  
1271      # 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
1272      # it the key.      # it the key.
1273      my %speciesHash = map { $fig->genus_species($_) => $_ } (keys %{$genomeHash});      my %speciesHash = map { $fig->genus_species($_) => $_ } (keys %{$genomeHash});
1274      # Create load objects for each of the tables we're loading.      # Create load objects for each of the tables we're loading.
1275      my $loadExternalAliasFunc = $self->_TableLoader('ExternalAliasFunc', $genomeCount * 4000);      my $loadExternalAliasFunc = $self->_TableLoader('ExternalAliasFunc');
1276      my $loadExternalAliasOrg = $self->_TableLoader('ExternalAliasOrg', $genomeCount * 4000);      my $loadExternalAliasOrg = $self->_TableLoader('ExternalAliasOrg');
1277      Trace("Beginning external data load.") if T(2);      if ($self->{options}->{loadOnly}) {
1278            Trace("Loading from existing files.") if T(2);
1279        } else {
1280            Trace("Generating external data.") if T(2);
1281      # 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.
1282      Open(\*ORGS, "<$FIG_Config::global/ext_org.table");          Open(\*ORGS, "sort +0 -1 -u -t\"\t\" $FIG_Config::global/ext_org.table |");
1283      my $orgLine;      my $orgLine;
1284      while (defined($orgLine = <ORGS>)) {      while (defined($orgLine = <ORGS>)) {
1285          # Clean the input line.          # Clean the input line.
# Line 1081  Line 1291 
1291      close ORGS;      close ORGS;
1292      # Now the function file.      # Now the function file.
1293      my $funcLine;      my $funcLine;
1294      Open(\*FUNCS, "<$FIG_Config::global/ext_func.table");          Open(\*FUNCS, "sort +0 -1 -u -t\"\t\" $FIG_Config::global/ext_func.table |");
1295      while (defined($funcLine = <FUNCS>)) {      while (defined($funcLine = <FUNCS>)) {
1296          # Clean the line ending.          # Clean the line ending.
1297          chomp $funcLine;          chomp $funcLine;
# Line 1097  Line 1307 
1307              $loadExternalAliasFunc->Put(@funcFields[0,1]);              $loadExternalAliasFunc->Put(@funcFields[0,1]);
1308          }          }
1309      }      }
1310        }
1311        # Finish the load.
1312        my $retVal = $self->_FinishAll();
1313        return $retVal;
1314    }
1315    
1316    
1317    =head3 LoadReactionData
1318    
1319    C<< my $stats = $spl->LoadReactionData(); >>
1320    
1321    Load the reaction data from FIG into Sprout.
1322    
1323    Reaction data connects reactions to the compounds that participate in them.
1324    
1325    The following relations are loaded by this method.
1326    
1327        Reaction
1328        ReactionURL
1329        Compound
1330        CompoundName
1331        CompoundCAS
1332        IsAComponentOf
1333    
1334    This method proceeds reaction by reaction rather than genome by genome.
1335    
1336    =over 4
1337    
1338    =item RETURNS
1339    
1340    Returns a statistics object for the loads.
1341    
1342    =back
1343    
1344    =cut
1345    #: Return Type $%;
1346    sub LoadReactionData {
1347        # Get this object instance.
1348        my ($self) = @_;
1349        # Get the FIG object.
1350        my $fig = $self->{fig};
1351        # Create load objects for each of the tables we're loading.
1352        my $loadReaction = $self->_TableLoader('Reaction');
1353        my $loadReactionURL = $self->_TableLoader('ReactionURL', $self->PrimaryOnly);
1354        my $loadCompound = $self->_TableLoader('Compound', $self->PrimaryOnly);
1355        my $loadCompoundName = $self->_TableLoader('CompoundName', $self->PrimaryOnly);
1356        my $loadCompoundCAS = $self->_TableLoader('CompoundCAS', $self->PrimaryOnly);
1357        my $loadIsAComponentOf = $self->_TableLoader('IsAComponentOf', $self->PrimaryOnly);
1358        if ($self->{options}->{loadOnly}) {
1359            Trace("Loading from existing files.") if T(2);
1360        } else {
1361            Trace("Generating annotation data.") if T(2);
1362            # First we create the compounds.
1363            my @compounds = $fig->all_compounds();
1364            for my $cid (@compounds) {
1365                # Check for names.
1366                my @names = $fig->names_of_compound($cid);
1367                # Each name will be given a priority number, starting with 1.
1368                my $prio = 1;
1369                for my $name (@names) {
1370                    $loadCompoundName->Put($cid, $name, $prio++);
1371                }
1372                # Create the main compound record. Note that the first name
1373                # becomes the label.
1374                my $label = (@names > 0 ? $names[0] : $cid);
1375                $loadCompound->Put($cid, $label);
1376                # Check for a CAS ID.
1377                my $cas = $fig->cas($cid);
1378                if ($cas) {
1379                    $loadCompoundCAS->Put($cid, $cas);
1380                }
1381            }
1382            # All the compounds are set up, so we need to loop through the reactions next. First,
1383            # we initialize the discriminator index. This is a single integer used to insure
1384            # duplicate elements in a reaction are not accidentally collapsed.
1385            my $discrim = 0;
1386            my @reactions = $fig->all_reactions();
1387            for my $reactionID (@reactions) {
1388                # Create the reaction record.
1389                $loadReaction->Put($reactionID, $fig->reversible($reactionID));
1390                # Compute the reaction's URL.
1391                my $url = HTML::reaction_link($reactionID);
1392                # Put it in the ReactionURL table.
1393                $loadReactionURL->Put($reactionID, $url);
1394                # Now we need all of the reaction's compounds. We get these in two phases,
1395                # substrates first and then products.
1396                for my $product (0, 1) {
1397                    # Get the compounds of the current type for the current reaction. FIG will
1398                    # give us 3-tuples: [ID, stoichiometry, main-flag]. At this time we do not
1399                    # have location data in SEED, so it defaults to the empty string.
1400                    my @compounds = $fig->reaction2comp($reactionID, $product);
1401                    for my $compData (@compounds) {
1402                        # Extract the compound data from the current tuple.
1403                        my ($cid, $stoich, $main) = @{$compData};
1404                        # Link the compound to the reaction.
1405                        $loadIsAComponentOf->Put($cid, $reactionID, $discrim++, "", $main,
1406                                                 $product, $stoich);
1407                    }
1408                }
1409            }
1410        }
1411      # Finish the load.      # Finish the load.
1412      my $retVal = $self->_FinishAll();      my $retVal = $self->_FinishAll();
1413      return $retVal;      return $retVal;
# Line 1112  Line 1423 
1423    
1424      GenomeGroups      GenomeGroups
1425    
1426  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,
1427    butThere is no direct support for genome groups in FIG, so we access the SEED
1428  files directly.  files directly.
1429    
1430  =over 4  =over 4
# Line 1132  Line 1444 
1444      my $fig = $self->{fig};      my $fig = $self->{fig};
1445      # Get the genome hash.      # Get the genome hash.
1446      my $genomeHash = $self->{genomes};      my $genomeHash = $self->{genomes};
     my $genomeCount = (keys %{$genomeHash});  
1447      # Create a load object for the table we're loading.      # Create a load object for the table we're loading.
1448      my $loadGenomeGroups = $self->_TableLoader('GenomeGroups', $genomeCount * 4);      my $loadGenomeGroups = $self->_TableLoader('GenomeGroups');
1449      Trace("Beginning group data load.") if T(2);      if ($self->{options}->{loadOnly}) {
1450            Trace("Loading from existing files.") if T(2);
1451        } else {
1452            Trace("Generating group data.") if T(2);
1453            # Currently there are no groups.
1454        }
1455        # Finish the load.
1456        my $retVal = $self->_FinishAll();
1457        return $retVal;
1458    }
1459    
1460    =head3 LoadSynonymData
1461    
1462    C<< my $stats = $spl->LoadSynonymData(); >>
1463    
1464    Load the synonym groups into Sprout.
1465    
1466    The following relations are loaded by this method.
1467    
1468        SynonymGroup
1469        IsSynonymGroupFor
1470    
1471    The source information for these relations is taken from the C<maps_to_id> method
1472    of the B<FIG> object. Unfortunately, to make this work, we need to use direct
1473    SQL against the FIG database.
1474    
1475    =over 4
1476    
1477    =item RETURNS
1478    
1479    Returns a statistics object for the loads.
1480    
1481    =back
1482    
1483    =cut
1484    #: Return Type $%;
1485    sub LoadSynonymData {
1486        # Get this object instance.
1487        my ($self) = @_;
1488        # Get the FIG object.
1489        my $fig = $self->{fig};
1490        # Get the genome hash.
1491        my $genomeHash = $self->{genomes};
1492        # Create a load object for the table we're loading.
1493        my $loadSynonymGroup = $self->_TableLoader('SynonymGroup');
1494        my $loadIsSynonymGroupFor = $self->_TableLoader('IsSynonymGroupFor');
1495        if ($self->{options}->{loadOnly}) {
1496            Trace("Loading from existing files.") if T(2);
1497        } else {
1498            Trace("Generating synonym group data.") if T(2);
1499            # Get the database handle.
1500            my $dbh = $fig->db_handle();
1501            # Ask for the synonyms.
1502            my $sth = $dbh->prepare_command("SELECT maps_to, syn_id FROM peg_synonyms ORDER BY maps_to");
1503            my $result = $sth->execute();
1504            if (! defined($result)) {
1505                Confess("Database error in Synonym load: " . $sth->errstr());
1506            } else {
1507                # Remember the current synonym.
1508                my $current_syn = "";
1509                # Count the features.
1510                my $featureCount = 0;
1511                # Loop through the synonym/peg pairs.
1512                while (my @row = $sth->fetchrow()) {
1513                    # Get the synonym ID and feature ID.
1514                    my ($syn_id, $peg) = @row;
1515                    # Insure it's for one of our genomes.
1516                    my $genomeID = FIG::genome_of($peg);
1517                    if (exists $genomeHash->{$genomeID}) {
1518                        # Verify the synonym.
1519                        if ($syn_id ne $current_syn) {
1520                            # It's new, so put it in the group table.
1521                            $loadSynonymGroup->Put($syn_id);
1522                            $current_syn = $syn_id;
1523                        }
1524                        # Connect the synonym to the peg.
1525                        $loadIsSynonymGroupFor->Put($syn_id, $peg);
1526                        # Count this feature.
1527                        $featureCount++;
1528                        if ($featureCount % 1000 == 0) {
1529                            Trace("$featureCount features processed.") if T(3);
1530                        }
1531                    }
1532                }
1533            }
1534        }
1535        # Finish the load.
1536        my $retVal = $self->_FinishAll();
1537        return $retVal;
1538    }
1539    
1540    =head3 LoadFamilyData
1541    
1542    C<< my $stats = $spl->LoadFamilyData(); >>
1543    
1544    Load the protein families into Sprout.
1545    
1546    The following relations are loaded by this method.
1547    
1548        Family
1549        IsFamilyForFeature
1550    
1551    The source information for these relations is taken from the C<families_for_protein>,
1552    C<family_function>, and C<sz_family> methods of the B<FIG> object.
1553    
1554    =over 4
1555    
1556    =item RETURNS
1557    
1558    Returns a statistics object for the loads.
1559    
1560    =back
1561    
1562    =cut
1563    #: Return Type $%;
1564    sub LoadFamilyData {
1565        # Get this object instance.
1566        my ($self) = @_;
1567        # Get the FIG object.
1568        my $fig = $self->{fig};
1569        # Get the genome hash.
1570        my $genomeHash = $self->{genomes};
1571        # Create load objects for the tables we're loading.
1572        my $loadFamily = $self->_TableLoader('Family');
1573        my $loadIsFamilyForFeature = $self->_TableLoader('IsFamilyForFeature');
1574        if ($self->{options}->{loadOnly}) {
1575            Trace("Loading from existing files.") if T(2);
1576        } else {
1577            Trace("Generating family data.") if T(2);
1578            # Create a hash for the family IDs.
1579            my %familyHash = ();
1580      # Loop through the genomes.      # Loop through the genomes.
1581      my $line;          for my $genomeID (sort keys %{$genomeHash}) {
1582      for my $genomeID (keys %{$genomeHash}) {              Trace("Processing features for $genomeID.") if T(2);
1583          Trace("Processing $genomeID.") if T(3);              # Loop through this genome's PEGs.
1584          # Open the NMPDR group file for this genome.              for my $fid ($fig->all_features($genomeID, "peg")) {
1585          if (open(TMP, "<$FIG_Config::organisms/$genomeID/NMPDR") &&                  $loadIsFamilyForFeature->Add("features", 1);
1586              defined($line = <TMP>)) {                  # Get this feature's families.
1587              # Clean the line ending.                  my @families = $fig->families_for_protein($fid);
1588              chomp $line;                  # Loop through the families, connecting them to the feature.
1589              # Add the group to the table. Note that there can only be one group                  for my $family (@families) {
1590              # per genome.                      $loadIsFamilyForFeature->Put($family, $fid);
1591              $loadGenomeGroups->Put($genomeID, $line);                      # If this is a new family, create a record for it.
1592                        if (! exists $familyHash{$family}) {
1593                            $familyHash{$family} = 1;
1594                            $loadFamily->Add("families", 1);
1595                            my $size = $fig->sz_family($family);
1596                            my $func = $fig->family_function($family);
1597                            $loadFamily->Put($family, $size, $func);
1598                        }
1599                    }
1600                }
1601            }
1602        }
1603        # Finish the load.
1604        my $retVal = $self->_FinishAll();
1605        return $retVal;
1606    }
1607    
1608    =head3 LoadDrugData
1609    
1610    C<< my $stats = $spl->LoadDrugData(); >>
1611    
1612    Load the drug target data into Sprout.
1613    
1614    The following relations are loaded by this method.
1615    
1616        DrugProject
1617        ContainsTopic
1618        DrugTopic
1619        ContainsAnalysisOf
1620        PDB
1621        IncludesBound
1622        IsBoundIn
1623        BindsWith
1624        Ligand
1625        DescribesProteinForFeature
1626        FeatureConservation
1627    
1628    The source information for these relations is taken from flat files in the
1629    C<$FIG_Config::drug_directory>. The file C<master_tables.list> contains
1630    a list of drug project names paired with file names. The named file (in the
1631    same directory) contains all the data for the project.
1632    
1633    =over 4
1634    
1635    =item RETURNS
1636    
1637    Returns a statistics object for the loads.
1638    
1639    =back
1640    
1641    =cut
1642    #: Return Type $%;
1643    sub LoadDrugData {
1644        # Get this object instance.
1645        my ($self) = @_;
1646        # Get the FIG object.
1647        my $fig = $self->{fig};
1648        # Get the genome hash.
1649        my $genomeHash = $self->{genomes};
1650        # Create load objects for the tables we're loading.
1651        my $loadDrugProject = $self->_TableLoader('DrugProject');
1652        my $loadContainsTopic = $self->_TableLoader('ContainsTopic');
1653        my $loadDrugTopic = $self->_TableLoader('DrugTopic');
1654        my $loadContainsAnalysisOf = $self->_TableLoader('ContainsAnalysisOf');
1655        my $loadPDB = $self->_TableLoader('PDB');
1656        my $loadIncludesBound = $self->_TableLoader('IncludesBound');
1657        my $loadIsBoundIn = $self->_TableLoader('IsBoundIn');
1658        my $loadBindsWith = $self->_TableLoader('BindsWith');
1659        my $loadLigand = $self->_TableLoader('Ligand');
1660        my $loadDescribesProteinForFeature = $self->_TableLoader('DescribesProteinForFeature');
1661        my $loadFeatureConservation = $self->_TableLoader('FeatureConservation');
1662        if ($self->{options}->{loadOnly}) {
1663            Trace("Loading from existing files.") if T(2);
1664        } else {
1665            Trace("Generating drug target data.") if T(2);
1666            # Load the project list. The file comes in as a list of chomped lines,
1667            # and we split them on the TAB character to make the project name the
1668            # key and the file name the value of the resulting hash.
1669            my %projects = map { split /\t/, $_ } Tracer::GetFile("$FIG_Config::drug_directory/master_tables.list");
1670            # Create hashes for the derived objects: PDBs, Features, and Ligands. These objects
1671            # may occur multiple times in a single project file or even in multiple project
1672            # files.
1673            my %ligands = ();
1674            my %pdbs = ();
1675            my %features = ();
1676            my %bindings = ();
1677            # Set up a counter for drug topics. This will be used as the key.
1678            my $topicCounter = 0;
1679            # Loop through the projects. We sort the keys not because we need them sorted, but
1680            # because it makes it easier to infer our progress from trace messages.
1681            for my $project (sort keys %projects) {
1682                Trace("Processing project $project.") if T(3);
1683                # Only proceed if the download file exists.
1684                my $projectFile = "$FIG_Config::drug_directory/$projects{$project}";
1685                if (! -f $projectFile) {
1686                    Trace("Project file $projectFile not found.") if T(0);
1687                } else {
1688                    # Create the project record.
1689                    $loadDrugProject->Put($project);
1690                    # Create a hash for the topics. Each project has one or more topics. The
1691                    # topic is identified by a URL, a category, and an identifier.
1692                    my %topics = ();
1693                    # Now we can open the project file.
1694                    Trace("Reading project file $projectFile.") if T(3);
1695                    Open(\*PROJECT, "<$projectFile");
1696                    # Get the first record, which is a list of column headers. We don't use this
1697                    # for anything, but it may be useful for debugging.
1698                    my $headerLine = <PROJECT>;
1699                    # Loop through the rest of the records.
1700                    while (! eof PROJECT) {
1701                        # Get the current line of data. Note that not all lines will have all
1702                        # the fields. In particular, the CLIBE data is fairly rare.
1703                        my ($authorOrganism, $category, $tag, $refURL, $peg, $conservation,
1704                            $pdbBound, $pdbBoundEval, $pdbFree, $pdbFreeEval, $pdbFreeTitle,
1705                            $protDistInfo, $passAspInfo, $passAspFile, $passWeightInfo,
1706                            $passWeightFile, $clibeInfo, $clibeURL, $clibeTotalEnergy,
1707                            $clibeVanderwaals, $clibeHBonds, $clibeEI, $clibeSolvationE)
1708                           = Tracer::GetLine(\*PROJECT);
1709                        # The tag contains an identifier for the current line of data followed
1710                        # by a text statement that generally matches a property name in the
1711                        # main database. We split it up, since the identifier goes with
1712                        # the PDB data and the text statement is part of the topic.
1713                        my ($lineID, $topicTag) = split /\s*,\s*/, $tag;
1714                        $loadDrugProject->Add("data line");
1715                        # Check for a new topic.
1716                        my $topicData = "$category\t$topicTag\t$refURL";
1717                        if (! exists $topics{$topicData}) {
1718                            # Here we have a new topic. Compute its ID.
1719                            $topicCounter++;
1720                            $topics{$topicData} = $topicCounter;
1721                            # Create its database record.
1722                            $loadDrugTopic->Put($topicCounter, $refURL, $category, $authorOrganism,
1723                                                $topicTag);
1724                            # Connect it to the project.
1725                            $loadContainsTopic->Put($project, $topicCounter);
1726                            $loadDrugTopic->Add("topic");
1727                        }
1728                        # Now we know the topic ID exists in the hash and the topic will
1729                        # appear in the database, so we get this topic's ID.
1730                        my $topicID = $topics{$topicData};
1731                        # If the feature in this line is new, we need to save its conservation
1732                        # number.
1733                        if (! exists $features{$peg}) {
1734                            $loadFeatureConservation->Put($peg, $conservation);
1735                            $features{$peg} = 1;
1736                        }
1737                        # Now we have two PDBs to deal with-- a bound PDB and a free PDB.
1738                        # The free PDB will have data about docking points; the bound PDB
1739                        # will have data about docking. We store both types as PDBs, and
1740                        # the special data comes from relationships. First we process the
1741                        # bound PDB.
1742                        if ($pdbBound) {
1743                            $loadPDB->Add("bound line");
1744                            # Insure this PDB is in the database.
1745                            $self->CreatePDB($pdbBound, lc "$pdbFreeTitle (bound)", "bound", \%pdbs, $loadPDB);
1746                            # Connect it to this topic.
1747                            $loadIncludesBound->Put($topicID, $pdbBound);
1748                            # Check for CLIBE data.
1749                            if ($clibeInfo) {
1750                                $loadLigand->Add("clibes");
1751                                # We have CLIBE data, so we create a ligand and relate it to the PDB.
1752                                if (! exists $ligands{$clibeInfo}) {
1753                                    # This is a new ligand, so create its record.
1754                                    $loadLigand->Put($clibeInfo);
1755                                    $loadLigand->Add("ligand");
1756                                    # Make sure we know this ligand already exists.
1757                                    $ligands{$clibeInfo} = 1;
1758                                }
1759                                # Now connect the PDB to the ligand using the CLIBE data.
1760                                $loadBindsWith->Put($pdbBound, $clibeInfo, $clibeURL, $clibeHBonds, $clibeEI,
1761                                                    $clibeSolvationE, $clibeVanderwaals);
1762                            }
1763                            # Connect this PDB to the feature.
1764                            $loadDescribesProteinForFeature->Put($pdbBound, $peg, $protDistInfo, $pdbBoundEval);
1765                        }
1766                        # Next is the free PDB.
1767                        if ($pdbFree) {
1768                            $loadPDB->Add("free line");
1769                            # Insure this PDB is in the database.
1770                            $self->CreatePDB($pdbFree, lc $pdbFreeTitle, "free", \%pdbs, $loadPDB);
1771                            # Connect it to this topic.
1772                            $loadContainsAnalysisOf->Put($topicID, $pdbFree, $passAspInfo,
1773                                                         $passWeightFile, $passWeightInfo, $passAspFile);
1774                            # Connect this PDB to the feature.
1775                            $loadDescribesProteinForFeature->Put($pdbFree, $peg, $protDistInfo, $pdbFreeEval);
1776                        }
1777                        # If we have both PDBs, we may need to link them.
1778                        if ($pdbFree && $pdbBound) {
1779                            $loadIsBoundIn->Add("connection");
1780                            # Insure we only link them once.
1781                            my $bindingKey =  "$pdbFree\t$pdbBound";
1782                            if (! exists $bindings{$bindingKey}) {
1783                                $loadIsBoundIn->Add("newConnection");
1784                                $loadIsBoundIn->Put($pdbFree, $pdbBound);
1785                                $bindings{$bindingKey} = 1;
1786                            }
1787                        }
1788                    }
1789                    # Close off this project.
1790                    close PROJECT;
1791                }
1792          }          }
         close TMP;  
1793      }      }
1794      # Finish the load.      # Finish the load.
1795      my $retVal = $self->_FinishAll();      my $retVal = $self->_FinishAll();
1796      return $retVal;      return $retVal;
1797  }  }
1798    
1799    
1800  =head2 Internal Utility Methods  =head2 Internal Utility Methods
1801    
1802    =head3 SpecialAttribute
1803    
1804    C<< my $count = SproutLoad::SpecialAttribute($id, \@attributes, $idxMatch, $idxValue, $pattern, $loader); >>
1805    
1806    Look for special attributes of a given type. A special attribute is found by comparing one of
1807    the columns of the incoming attribute list to a search pattern. If a match is found, then
1808    another column is put into an output table connected to the specified ID.
1809    
1810    For example, when processing features, the attribute list we look at has three columns: attribute
1811    name, attribute value, and attribute value HTML. The IEDB attribute exists if the attribute name
1812    begins with C<iedb_>. The call signature is therefore
1813    
1814        my $found = SpecialAttribute($fid, \@attributeList, 0, 2, '^iedb_', $loadFeatureIEDB);
1815    
1816    The pattern is matched against column 0, and if we have a match, then column 2's value is put
1817    to the output along with the specified feature ID.
1818    
1819    =over 4
1820    
1821    =item id
1822    
1823    ID of the object whose special attributes are being loaded. This forms the first column of the
1824    output.
1825    
1826    =item attributes
1827    
1828    Reference to a list of tuples.
1829    
1830    =item idxMatch
1831    
1832    Index in each tuple of the column to be matched against the pattern. If the match is
1833    successful, an output record will be generated.
1834    
1835    =item idxValue
1836    
1837    Index in each tuple of the column to be put as the second column of the output.
1838    
1839    =item pattern
1840    
1841    Pattern to be matched against the specified column. The match will be case-insensitive.
1842    
1843    =item loader
1844    
1845    An object to which each output record will be put. Usually this is an B<ERDBLoad> object,
1846    but technically it could be anything with a C<Put> method.
1847    
1848    =item RETURN
1849    
1850    Returns a count of the matches found.
1851    
1852    =item
1853    
1854    =back
1855    
1856    =cut
1857    
1858    sub SpecialAttribute {
1859        # Get the parameters.
1860        my ($id, $attributes, $idxMatch, $idxValue, $pattern, $loader) = @_;
1861        # Declare the return variable.
1862        my $retVal = 0;
1863        # Loop through the attribute rows.
1864        for my $row (@{$attributes}) {
1865            # Check for a match.
1866            if ($row->[$idxMatch] =~ m/$pattern/i) {
1867                # We have a match, so output a row.
1868                $loader->Put($id, $row->[$idxValue]);
1869                $retVal++;
1870            }
1871        }
1872        Trace("$retVal special attributes found for $id and loader " . $loader->RelName() . ".") if T(4) && $retVal;
1873        # Return the number of matches.
1874        return $retVal;
1875    }
1876    
1877    =head3 CreatePDB
1878    
1879    C<< $loader->CreatePDB($pdbID, $title, $type, \%pdbHash); >>
1880    
1881    Insure that a PDB record exists for the identified PDB. If one does not exist, it will be
1882    created.
1883    
1884    =over 4
1885    
1886    =item pdbID
1887    
1888    ID string (usually an unqualified file name) for the desired PDB.
1889    
1890    =item title
1891    
1892    Title to use if the PDB must be created.
1893    
1894    =item type
1895    
1896    Type of PDB: C<free> or C<bound>
1897    
1898    =item pdbHash
1899    
1900    Hash containing the IDs of PDBs that have already been created.
1901    
1902    =item pdbLoader
1903    
1904    Load object for the PDB table.
1905    
1906    =back
1907    
1908    =cut
1909    
1910    sub CreatePDB {
1911        # Get the parameters.
1912        my ($self, $pdbID, $title, $type, $pdbHash, $pdbLoader) = @_;
1913        $pdbLoader->Add("PDB check");
1914        # Check to see if this is a new PDB.
1915        if (! exists $pdbHash->{$pdbID}) {
1916            # It is, so we create it.
1917            $pdbLoader->Put($pdbID, $title, $type);
1918            $pdbHash->{$pdbID} = 1;
1919            # Count it.
1920            $pdbLoader->Add("PDB-$type");
1921        }
1922    }
1923    
1924  =head3 TableLoader  =head3 TableLoader
1925    
1926  Create an ERDBLoad object for the specified table. The object is also added to  Create an ERDBLoad object for the specified table. The object is also added to
# Line 1172  Line 1935 
1935    
1936  Name of the table (relation) being loaded.  Name of the table (relation) being loaded.
1937    
1938  =item rowCount (optional)  =item ignore
1939    
1940  Estimated maximum number of rows in the table.  TRUE if the table should be ignored entirely, else FALSE.
1941    
1942  =item RETURN  =item RETURN
1943    
# Line 1186  Line 1949 
1949    
1950  sub _TableLoader {  sub _TableLoader {
1951      # Get the parameters.      # Get the parameters.
1952      my ($self, $tableName, $rowCount) = @_;      my ($self, $tableName, $ignore) = @_;
1953      # Create the load object.      # Create the load object.
1954      my $retVal = ERDBLoad->new($self->{erdb}, $tableName, $self->{loadDirectory}, $rowCount);      my $retVal = ERDBLoad->new($self->{erdb}, $tableName, $self->{loadDirectory}, $self->LoadOnly,
1955                                   $ignore);
1956      # Cache it in the loader list.      # Cache it in the loader list.
1957      push @{$self->{loaders}}, $retVal;      push @{$self->{loaders}}, $retVal;
1958      # Return it to the caller.      # Return it to the caller.
# Line 1222  Line 1986 
1986      my $retVal = Stats->new();      my $retVal = Stats->new();
1987      # Get the loader list.      # Get the loader list.
1988      my $loadList = $self->{loaders};      my $loadList = $self->{loaders};
1989        # Create a hash to hold the statistics objects, keyed on relation name.
1990        my %loaderHash = ();
1991      # 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
1992      # 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.
1993      while (my $loader = pop @{$loadList}) {      while (my $loader = pop @{$loadList}) {
1994            # Get the relation name.
1995            my $relName = $loader->RelName;
1996            # Check the ignore flag.
1997            if ($loader->Ignore) {
1998                Trace("Relation $relName not loaded.") if T(2);
1999            } else {
2000                # Here we really need to finish.
2001                Trace("Finishing $relName.") if T(2);
2002          my $stats = $loader->Finish();          my $stats = $loader->Finish();
2003                $loaderHash{$relName} = $stats;
2004            }
2005        }
2006        # Now we loop through again, actually loading the tables. We want to finish before
2007        # loading so that if something goes wrong at this point, all the load files are usable
2008        # and we don't have to redo all that work.
2009        for my $relName (sort keys %loaderHash) {
2010            # Get the statistics for this relation.
2011            my $stats = $loaderHash{$relName};
2012            # Check for a database load.
2013            if ($self->{options}->{dbLoad}) {
2014                # Here we want to use the load file just created to load the database.
2015                Trace("Loading relation $relName.") if T(2);
2016                my $newStats = $self->{sprout}->LoadUpdate(1, [$relName]);
2017                # Accumulate the statistics from the DB load.
2018                $stats->Accumulate($newStats);
2019            }
2020          $retVal->Accumulate($stats);          $retVal->Accumulate($stats);
         my $relName = $loader->RelName;  
2021          Trace("Statistics for $relName:\n" . $stats->Show()) if T(2);          Trace("Statistics for $relName:\n" . $stats->Show()) if T(2);
2022      }      }
2023      # Return the load statistics.      # Return the load statistics.

Legend:
Removed from v.1.7  
changed lines
  Added in v.1.76

MCS Webmaster
ViewVC Help
Powered by ViewVC 1.0.3