Diff of /Sprout/SproutSubsys.pm

-revision 1.20, Wed Oct 15 11:46:57 2008 UTC
+revision 1.21, Mon Jan 19 21:44:01 2009 UTC
 Line 74
  Map of roles to EC numbers for the Hope reactions. This object is not loaded
  until it is needed.
+ =item rows
+ Map of spreadsheet rows, keyed by genome ID. Each row is a list of cells. Each
+ cell is a list of feature ID.
+ =item featureData
+ Hash mapping feature IDs to assigned functions.
  =back
  =cut
-Line 188
+Line 197
                      reactionHash => undef,
                      # Version number.
                      version => $version,
+                     # Row hash, initially undefined.
+                     rows => undef,
+                     # Map of feature IDs to functional assignments
+                     featureData => {},
                  };
          # Bless and return it.
          bless $retVal, $class;
-Line 256
+Line 269
      my ($self, $rowIndex) = @_;
      # Get the genome ID for the specified row's genome.
      my $genomeID = $self->{genomes}->[$rowIndex]->[0];
-     # Read the row from the database. We won't get exactly what we want. Instead, we'll
+     # Get the row hash.
-     # get a list of triplets, each consisting of a role name, a feature ID, and a cluster
+     my $rowHash = $self->_get_spreadsheet();
-     # number. We need to convert this into a list of lists and stash the clustering information
+     # Declare the return variable.
-     # in the color hash.
+     my @retVal;
-     my @rowData = $self->{sprout}->GetAll([qw(Subsystem HasSSCell IsGenomeOf IsRoleOf ContainsFeature)],
+     # If this genome does not exist for the subsystem, all the cells are empty.
-                                           "Subsystem(id) = ? AND IsGenomeOf(from-link) = ?",
+     if (! exists $rowHash->{$genomeID}) {
-                                           [$self->{name}, $genomeID],
+         @retVal = map { [] } @{$self->{roles}};
-                                           [qw(IsRoleOf(from-link) ContainsFeature(to-link)
+     } else {
-                                               ContainsFeature(cluster-number))]);
+         # Here we just return the row.
-     # Now we do the conversion. We must first create an array of empty lists, one per
+         push @retVal, @{$rowHash->{$genomeID}};
-     # row index.
-     my @retVal = map { [] } @{$self->{roles}};
-     # Get the hash for converting role IDs to role indexes.
-     my $roleHash = $self->{roleHash};
-     # Now we stash all the feature IDs in the appropriate columns of the row list.
-     for my $rowDatum (@rowData) {
-         # Get the role ID, the peg ID, and the cluster number.
-         my ($role, $peg, $cluster) = @{$rowDatum};
-         # Put the peg in the role's peg list.
-         push @{$retVal[$roleHash->{$role}]}, $peg;
-         # Put the cluster number in the color hash.
-         $self->{colorHash}->{$peg} = $cluster;
      }
      # Return the result.
      return \@retVal;
-Line 307
+Line 308
  sub get_roles_for_genome {
      # Get the parameters.
      my ($self, $genome_id) = @_;
-     # This next statement gets all of the nonempty cells for the genome's row and memorizes
+     # Get the subsystem's spreadsheet.
-     # the roles by rolling them into a hash. The query connects four relationship tables on
+     my $rowHash = $self->_get_spreadsheet();
-     # a single common key-- the spreadsheet cell ID. The IsGenomeOf table insures the cell is for the
+     # Declare the return variable.
-     # correct genome. The HasSSCell table insures that it belongs to the correct subsystem. The
+     my @retVal;
-     # ContainsFeature table insures that it contains at least one feature. Finally, IsRoleOf tells
+     # Only proceed if this genome exists for this subsyste,
-     # us the cell's role. If a cell has more than one feature, the result list from the query will return
+     if (exists $rowHash->{$genome_id}) {
-     # one instance of the role for every distinct feature. The hash collapses the duplicates automatically.
+         # Get the role list.
-     my %retVal = map { $_ => 1 } $self->{sprout}->GetFlat([qw(ContainsFeature HasSSCell IsGenomeOf IsRoleOf)],
+         my $roles = $self->{roles};
-                                                           "HasSSCell(from-link) = ? AND IsGenomeOf(from-link) = ?",
+         # Get the row's cell list.
-                                                           [$self->{name}, $genome_id], 'IsRoleOf(from-link)');
+         my $row = $rowHash->{$genome_id};
+         # Loop through the cells. We'll save the role name for each
+         # nonempty cell.
+         my $cols = scalar @$roles;
+         for (my $i = 0; $i < $cols; $i++) {
+             my $cell = $row->[$i];
+             if (scalar @$cell) {
+                 push @retVal, $roles->[$i];
+             }
+         }
+     }
      # Return the result.
-     return keys %retVal;
+     return @retVal;
  }
  =head3 get_abbr_for_role
-Line 883
+Line 894
  Return the cluster number for the specified PEG, or C<-1> if the
  cluster number for the PEG is unknown or it is not clustered.
- The cluster number is read into the color hash by the
- L</get_pegs_from_cell> method. If the incoming PEG IDs do not
- come from the most recent cell retrievals, the information returned
- will be invalid. This is a serious design flaw which needs to be
- fixed soon.
  =over 4
  =item pegID
-Line 908
+Line 913
      my ($self, $pegID) = @_;
      # Declare the return variable.
      my $retVal = -1;
+     # Insure we have a color hash.
+     $self->_get_spreadsheet();
      # Check for a cluster number in the color hash.
      if (exists $self->{colorHash}->{$pegID}) {
          $retVal = $self->{colorHash}->{$pegID};
-Line 916
+Line 923
      return $retVal;
  }
  =head3 get_pegs_from_cell
      my @pegs = $sub->get_pegs_from_cell($rowstr, $colstr);
-Line 967
+Line 975
              $genomeID = $genomeList->[$rowstr]->[0];
          }
      }
-     # Construct the spreadsheet cell ID from the information we have.
+     # Get the spreadsheet.
-     my $cellID = $sprout->DigestKey($self->{name} . ":$genomeID:$colIdx");
+     my $rowHash = $self->_get_spreadsheet();
-     # Get the list of PEG IDs and cluster numbers for the indicated cell.
+     # Delcare the return variable.
-     my @pegList = $sprout->GetAll(['ContainsFeature'], 'ContainsFeature(from-link) = ?',
+     my @retVal;
-                                   [$cellID], ['ContainsFeature(to-link)',
+     # Only proceed if this genome is in this subsystem.
-                                               'ContainsFeature(cluster-number)']);
+     if (exists $rowHash->{$genomeID}) {
-     # Copy the pegs into the return list, and save the cluster numbers in the color hash.
+         # Push the cell's contents into the return list.
-     my @retVal = ();
+         push @retVal, @{$rowHash->{$genomeID}->[$colIdx]};
-     for my $pegEntry (@pegList) {
-         my ($peg, $cluster) = @{$pegEntry};
-         $self->{colorHash}->{$peg} = $cluster;
-         push @retVal, $peg;
      }
      # Return the list. If the spreadsheet cell was empty or non-existent, we'll end
      # up returning an empty list.
-Line 1279
+Line 1283
      return ($type, $fh);
  }
+ =head3 get_hope_scenario_names
+     my @names = $sub->get_hope_scenario_names();
+ Return a list of the names for the scenarios associated with this
+ subsystem.
+ =cut
+ sub get_hope_scenario_names {
+     # Get the parameters.
+     my ($self) = @_;
+     # Get the names from the database.
+     my $sprout = $self->{sprout};
+     my @retVal = $sprout->GetFlat("HasScenario",
+                                   "HasScenario(from-link) = ? ORDER BY HasScenario(to-link)",
+                                   [$self->{name}], 'to-link');
+     # Return the result.
+     return @retVal;
+ }
+ =head3 get_hope_input_compounds
+     my @compounds = $sub->get_hope_input_compounds($name);
+ Return a list of the input compounds for the named hope scenario.
+ =over 4
+ =item name
+ Name of a Hope scenario attached to this subsystem.
+ =item RETURN
+ Returns a list of compound IDs.
+ =back
+ =cut
+ sub get_hope_input_compounds {
+     # Get the parameters.
+     my ($self, $name) = @_;
+     # Ask for the compounds.
+     my @retVal = $self->{sprout}->GetFlat("IsInputFor", "IsInputFor(to-link) = ?",
+                                           [$name], "IsInputFor(from-link)");
+     # Return the result.
+     return @retVal;
+ }
+ =head3 get_hope_output_compounds
+     my ($main, $aux) = $sub->get_hope_output_compounds($name);
+ Return a list of the output compounds for the named hope scenario.
+ =over 4
+ =item name
+ Name of the relevant scenario.
+ =item RETURN
+ Returns two lists of compound IDs: one for the main outputs and one for the
+ auxiliary outputs.
+ =back
+ =cut
+ sub get_hope_output_compounds {
+     # Get the parameters.
+     my ($self, $name) = @_;
+     # Ask for the compounds.
+     my $sprout = $self->{sprout};
+     my @pairs = $sprout->GetAll("IsOutputOf", "IsOutputOf(to-link) = ?",
+                                 [$name], "from-link auxiliary");
+     # We now have a list of pairs in the form [name, aux-flag]. We put each
+     # name in the list indicated by its aux-flag.
+     my @retVal = ([], []);
+     for my $pair (@pairs) {
+         push @{$retVal[$pair->[1]]}, $pair->[0];
+     }
+     # Return the result.
+     return @retVal;
+ }
+ =head3 get_hope_map_ids
+     my @mapIDs = $sub->get_hope_map_ids($name);
+ Return a list of the ID numbers for the diagrams associated with the named
+ scenario.
+ =over 4
+ =item name
+ Name of the relevant scenario.
+ =item RETURN
+ Returns a list of the ID numbers for the KEGG diagrams associated with this
+ scenario. These are different from the diagram IDs, all of which begin with
+ the string "map". This recognizes a design incompatability between SEED and
+ Sprout.
+ =back
+ =cut
+ sub get_hope_map_ids {
+     # Get the parameters.
+     my ($self, $name) = @_;
+     # Get the map IDs.
+     my @diagrams = $self->{sprout}->GetFlat('IsOnDiagram', "IsOnDiagram(from-link) = ?",
+                                             [$name], 'to-link');
+     # Modify and return the result.
+     my @retVal = map { /(\d+)/ } @diagrams;
+     return @retVal;
+ }
+ =head3 all_functions
+     my $pegRoles = $sub->all_functions();
+ Return a hash of all the features in the subsystem. The hash maps each
+ feature ID to its functional assignment.
+ =cut
+ sub all_functions {
+     # Get the parameters.
+     my ($self) = @_;
+     # Insure we have a spreadsheet.
+     $self->_get_spreadsheet();
+     # Return the feature hash.
+     return $self->{featureData};
+ }
+ =head2 Internal Utility Methods
+ =head3 _get_spreadsheet
+     my $hash = $sub->_get_spreadsheet();
+ Return a reference to a hash mapping each of the subsystem's genomes to
+ their spreadsheet rows. Each row is a list of cells, and each cell is a
+ list of feature IDs. This method also creates the color hash that maps PEGs
+ to cluster numbers.
+ =cut
+ sub _get_spreadsheet {
+     # Get the parameters.
+     my ($self) = @_;
+     # Do we already have a spreadsheet?
+     my $retVal = $self->{rows};
+     if (! defined $retVal) {
+         # We don't, so we have to create one. Start with an empty hash.
+         $retVal = {};
+         # Ask for all the subsystem's cells and their features.
+         my $query = $self->{sprout}->Get("HasSSCell SSCell ContainsFeature Feature",
+                                          "HasSSCell(from-link) = ?",
+                                          [$self->{name}]);
+         # Loop through the features.
+         while (my $feature = $query->Fetch()) {
+             # Get the column number, the feature ID, and the cluster number.
+             my $featureID = $feature->PrimaryValue('ContainsFeature(to-link)');
+             my $cluster = $feature->PrimaryValue('ContainsFeature(cluster-number)');
+             my $column = $feature->PrimaryValue('SSCell(column-number)');
+             my $role = $feature->PrimaryValue('Feature(assignment)');
+             # Compute the genome.
+             my $genomeID = FIG::genome_of($featureID);
+             # If we don't have this genome in the hash, create it.
+             if (! exists $retVal->{$genomeID}) {
+                 # The initial value is a list of empty lists. Features
+                 # are then pushed into each individual list.
+                 my @row = map { [] } @{$self->{roles}};
+                 # Put this list of null lists in the hash.
+                 $retVal->{$genomeID} = \@row;
+             }
+             # Get this row. We know now that it exists.
+             my $row = $retVal->{$genomeID};
+             # Add this feature to the appropriate cell in the row.
+             push @{$row->[$column]}, $featureID;
+             # Put it in the color hash and the feature data hash.
+             $self->{colorHash}->{$featureID} = $cluster;
+             $self->{featureData}->{$featureID} = $role;
+         }
+         # Save the row hash.
+         $self->{rows} = $retVal;
+     }
+     # Return the result.
+     return $retVal;
+ }
+ =head3 get_col
+     my $cellArray = $sub->get_col($idx);
+ Return an array of the cells in the specified column of the subsystem
+ spreadsheet. Each cell is a reference to a list of the features for the
+ corresponding row in the specified column.
+ =over 4
+ =item idx
+ Index of the desired column.
+ =item RETURN
+ Returns a reference to a list containing the spreadsheet column's cells, in
+ row order.
+ =back
+ =cut
+ sub get_col {
+     # Get the parameters.
+     my ($self, $idx) = @_;
+     # Declare the return variable.
+     my @retVal;
+     # Get the subsystem spreadsheet.
+     my $sheet = $self->_get_spreadsheet();
+     # Loop through the row list.
+     for my $rowPair (@{$self->{genomes}}) {
+         # Get the genome for this row. Each row pair is [genomeID, variantCode].
+         my ($genomeID) = @$rowPair;
+         # Get the genome's row in the spreadsheet.
+         my $rowList = $sheet->{$genomeID};
+         # Push this column's cell into the output list.
+         push @retVal, $rowList->[$idx];
+     }
+     # Return the result.
+     return \@retVal;
+ }
 ;

 Legend:



Removed from v.1.20
 


changed lines


 
Added in v.1.21
 Legend:



Removed from v.1.20
 


changed lines


 
Added in v.1.21
-Removed from v.1.20
+Added in v.1.21

MCS Webmaster	ViewVC Help
Powered by ViewVC 1.0.3