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

Diff of /Sprout/Sprout.pm

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

revision 1.117, Tue Sep 16 18:57:59 2008 UTC revision 1.119, Tue Sep 30 15:23:55 2008 UTC
# Line 57  Line 57 
57    
58  =item dbName  =item dbName
59    
60  Name of the database.  Name of the database. If omitted, the default Sprout database name is used.
61    
62  =item options  =item options
63    
# Line 96  Line 96 
96  sub new {  sub new {
97      # Get the parameters.      # Get the parameters.
98      my ($class, $dbName, $options) = @_;      my ($class, $dbName, $options) = @_;
99        # Default the database name if it is missing.
100        if (! defined $dbName) {
101            $dbName = $FIG_Config::sproutDB;
102        } elsif (ref $dbName eq 'HASH') {
103            $options = $dbName;
104            $dbName = $FIG_Config::sproutDB;
105        }
106      # Compute the DBD directory.      # Compute the DBD directory.
107      my $dbd_dir = (defined($FIG_Config::dbd_dir) ? $FIG_Config::dbd_dir :      my $dbd_dir = (defined($FIG_Config::dbd_dir) ? $FIG_Config::dbd_dir :
108                                                    $FIG_Config::fig );                                                    $FIG_Config::fig );
# Line 140  Line 147 
147      $retVal->{groupHash} = undef;      $retVal->{groupHash} = undef;
148      # Set up space for the genome hash. We use this to identify NMPDR genomes.      # Set up space for the genome hash. We use this to identify NMPDR genomes.
149      $retVal->{genomeHash} = undef;      $retVal->{genomeHash} = undef;
150        # Remember the data directory name.
151        $retVal->{dataDir} = $dataDir;
152      # Connect to the attributes.      # Connect to the attributes.
153      if ($FIG_Config::attrURL) {      if ($FIG_Config::attrURL) {
154          Trace("Remote attribute server $FIG_Config::attrURL chosen.") if T(3);          Trace("Remote attribute server $FIG_Config::attrURL chosen.") if T(3);
# Line 714  Line 723 
723          # We don't have one pre-built, so we build and save it now.          # We don't have one pre-built, so we build and save it now.
724          $stemmer = BioWords->new(exceptions => "$FIG_Config::sproutData/Exceptions.txt",          $stemmer = BioWords->new(exceptions => "$FIG_Config::sproutData/Exceptions.txt",
725                                   stops => "$FIG_Config::sproutData/StopWords.txt",                                   stops => "$FIG_Config::sproutData/StopWords.txt",
726                                   cache => 1);                                   cache => 0);
727          $self->{stemmer} = $stemmer;          $self->{stemmer} = $stemmer;
728      }      }
729      # Try to stem the word.      # Try to stem the word.
# Line 3122  Line 3131 
3131      # Declare the return variable.      # Declare the return variable.
3132      my %retVal = ();      my %retVal = ();
3133      # Get a list of the genome features that participate in subsystems. For each      # Get a list of the genome features that participate in subsystems. For each
3134      # feature we get its spreadsheet cells and the corresponding roles.      # feature we get its subsystem ID and the corresponding roles.
3135      my @roleData = $self->GetAll(['HasFeature', 'ContainsFeature', 'IsRoleOf'],      my @roleData = $self->GetAll(['HasFeature', 'ContainsFeature', 'IsRoleOf', 'HasSSCell'],
3136                               "HasFeature(from-link) = ?", [$genomeID],                               "HasFeature(from-link) = ?", [$genomeID],
3137                               ['HasFeature(to-link)', 'IsRoleOf(to-link)', 'IsRoleOf(from-link)']);                                   ['HasFeature(to-link)', 'IsRoleOf(from-link)',  'HasSSCell(from-link)']);
3138      # Now we get a list of the spreadsheet cells and their associated subsystems. Subsystems      # Now we get a list of valid subsystems. These are the subsystems connected to the genome with
3139      # with an unknown variant code (-1) are skipped. Note the genome ID is at both ends of the      # a non-negative variant code.
3140      # list. We use it at the beginning to get all the spreadsheet cells for the genome and      my %subs = map { $_ => 1 } $self->GetFlat(['ParticipatesIn'],
3141      # again at the end to filter out participation in subsystems with a negative variant code.                                                  "ParticipatesIn(from-link) = ? AND ParticipatesIn(variant-code) >= 0",
3142      my @cellData = $self->GetAll(['IsGenomeOf', 'HasSSCell', 'ParticipatesIn'],                                                  [$genomeID], 'ParticipatesIn(to-link)');
3143                                   "IsGenomeOf(from-link) = ? AND ParticipatesIn(variant-code) >= 0 AND ParticipatesIn(from-link) = ?",      # We loop through @roleData to build the hash.
                                  [$genomeID, $genomeID], ['HasSSCell(to-link)', 'HasSSCell(from-link)']);  
     # Now "@roleData" lists the spreadsheet cell and role for each of the genome's features.  
     # "@cellData" lists the subsystem name for each of the genome's spreadsheet cells. We  
     # link these two lists together to create the result. First, we want a hash mapping  
     # spreadsheet cells to subsystem names.  
     my %subHash = map { $_->[0] => $_->[1] } @cellData;  
     # We loop through @cellData to build the hash.  
3144      for my $roleEntry (@roleData) {      for my $roleEntry (@roleData) {
3145          # Get the data for this feature and cell.          # Get the data for this feature and cell.
3146          my ($fid, $cellID, $role) = @{$roleEntry};          my ($fid, $role, $subsys) = @{$roleEntry};
3147          # Check for a subsystem name.          Trace("Subsystem for $fid is $subsys.") if T(4);
3148          my $subsys = $subHash{$cellID};          # Check the subsystem;
3149          if ($subsys) {          if ($subs{$subsys}) {
3150                Trace("Subsystem found.") if T(4);
3151              # Insure this feature has an entry in the return hash.              # Insure this feature has an entry in the return hash.
3152              if (! exists $retVal{$fid}) { $retVal{$fid} = []; }              if (! exists $retVal{$fid}) { $retVal{$fid} = []; }
3153              # Merge in this new data.              # Merge in this new data.
# Line 3984  Line 3987 
3987  sub CleanKeywords {  sub CleanKeywords {
3988      # Get the parameters.      # Get the parameters.
3989      my ($self, $searchExpression) = @_;      my ($self, $searchExpression) = @_;
3990      # Perform the standard cleanup.      # Get the stemmer.
3991      my $words = $self->ERDB::CleanKeywords($searchExpression);      my $stemmer = $self->GetStemmer();
3992      # Fix the periods in EC and TC numbers.      # Convert the search expression using the stemmer.
3993      $words =~ s/(\d+|\-)\.(\d+|-)\.(\d+|-)\.(\d+|-)/$1_$2_$3_$4/g;      my $retVal = $stemmer->PrepareSearchExpression($searchExpression);
     # Fix non-trailing periods.  
     $words =~ s/\.(\w)/_$1/g;  
     # Fix non-leading minus signs.  
     $words =~ s/(\w)[\-]/$1_/g;  
     # Fix the vertical bars and colons  
     $words =~ s/(\w)[|:](\w)/$1'$2/g;  
     # Now split up the list so that each keyword is in its own string. We keep the delimiters  
     # because they may contain boolean expression data.  
     my @words = split /([^A-Za-z'0-9_]+)/, $words;  
     # We'll convert the stemmable words into stems and re-assemble the result.  
     my $retVal = "";  
     for my $word (@words) {  
         my $stem = $self->Stem($word);  
         if (defined $stem) {  
             $retVal .= $stem;  
         } else {  
             $retVal .= $word;  
         }  
     }  
3994      Trace("Cleaned keyword list for \"$searchExpression\" is \"$retVal\".") if T(3);      Trace("Cleaned keyword list for \"$searchExpression\" is \"$retVal\".") if T(3);
3995      # Return the result.      # Return the result.
3996      return $retVal;      return $retVal;
3997  }  }
3998    
3999    =head3 GetSourceObject
4000    
4001        my $source = $erdb->GetSourceObject();
4002    
4003    Return the object to be used in creating load files for this database.
4004    
4005    =cut
4006    
4007    sub GetSourceObject {
4008        # Get access to the FIG code.
4009        require FIG;
4010        # Return a FIG object.
4011        return FIG->new();
4012    }
4013    
4014    =head3 SectionList
4015    
4016        my @sections = $erdb->SectionList($source);
4017    
4018    Return a list of the names for the different data sections used when loading this database.
4019    The default is an empty string, in which case there is only one section representing the
4020    entire database.
4021    
4022    =over 4
4023    
4024    =item source
4025    
4026    Source object used to access the data from which the database is loaded. This is the
4027    same object returned by L</GetSourceObject>; however, we ask the caller to pass it
4028    in as a parameter so that we don't end up creating multiple copies of a potentially
4029    expensive data structure.
4030    
4031    =item RETURN
4032    
4033    Returns a list of section names.
4034    
4035    =back
4036    
4037    =cut
4038    
4039    sub SectionList {
4040        # Get the parameters.
4041        my ($self, $source) = @_;
4042        # Ask the BaseSproutLoader for a section list.
4043        require BaseSproutLoader;
4044        my @retVal = BaseSproutLoader::GetSectionList($self, $source);
4045        # Return the list.
4046        return @retVal;
4047    }
4048    
4049    =head3 Loader
4050    
4051        my $groupLoader = $erdb->Loader($groupName, $source, $options);
4052    
4053    Return an [[ERDBLoadGroupPm]] object for the specified load group. This method is used
4054    by [[ERDBGeneratorPl]] to create the load group objects. If you are not using
4055    [[ERDBGeneratorPl]], you don't need to override this method.
4056    
4057    =over 4
4058    
4059    =item groupName
4060    
4061    Name of the load group whose object is to be returned. The group name is
4062    guaranteed to be a single word with only the first letter capitalized.
4063    
4064    =item source
4065    
4066    The source object used to access the data from which the load file is derived. This
4067    is the same object returned by L</GetSourceObject>; however, we ask the caller to pass
4068    it in as a parameter so that we don't end up creating multiple copies of a potentially
4069    expensive data structure.
4070    
4071    =item options
4072    
4073    Reference to a hash of command-line options.
4074    
4075    =item RETURN
4076    
4077    Returns an [[ERDBLoadGroupPm]] object that can be used to process the specified load group
4078    for this database.
4079    
4080    =back
4081    
4082    =cut
4083    
4084    sub Loader {
4085        # Get the parameters.
4086        my ($self, $groupName, $source, $options) = @_;
4087        # Compute the loader name.
4088        my $loaderClass = "${groupName}SproutLoader";
4089        # Pull in its definition.
4090        require "$loaderClass.pm";
4091        # Create an object for it.
4092        my $retVal = eval("$loaderClass->new(\$self, \$source, \$options)");
4093        # Insure it worked.
4094        Confess("Could not create $loaderClass object: $@") if $@;
4095        # Return it to the caller.
4096        return $retVal;
4097    }
4098    
4099    =head3 LoadGroupList
4100    
4101        my @groups = $erdb->LoadGroupList();
4102    
4103    Returns a list of the names for this database's load groups. This method is used
4104    by [[ERDBGeneratorPl]] when the user wishes to load all table groups. The default
4105    is a single group called 'All' that loads everything.
4106    
4107    =cut
4108    
4109    sub LoadGroupList {
4110        # Return the list.
4111        return qw(Genome Subsystem Feature Annotation Property Source Reaction Synonym Drug);
4112    }
4113    
4114    =head3 LoadDirectory
4115    
4116        my $dirName = $erdb->LoadDirectory();
4117    
4118    Return the name of the directory in which load files are kept. The default is
4119    the FIG temporary directory, which is a really bad choice, but it's always there.
4120    
4121    =cut
4122    
4123    sub LoadDirectory {
4124        # Get the parameters.
4125        my ($self) = @_;
4126        # Return the directory name.
4127        return $self->{dataDir};
4128    }
4129    
4130  =head2 Internal Utility Methods  =head2 Internal Utility Methods
4131    
4132    =head3 GetStemmer
4133    
4134        my $stermmer = $sprout->GetStemmer();
4135    
4136    Return the stemmer object for this database.
4137    
4138    =cut
4139    
4140    sub GetStemmer {
4141        # Get the parameters.
4142        my ($self) = @_;
4143        # Declare the return variable.
4144        my $retVal = $self->{stemmer};
4145        if (! defined $retVal) {
4146            # We don't have one pre-built, so we build and save it now.
4147            $retVal = BioWords->new(exceptions => "$FIG_Config::sproutData/Exceptions.txt",
4148                                     stops => "$FIG_Config::sproutData/StopWords.txt",
4149                                     cache => 1);
4150            $self->{stemmer} = $retVal;
4151        }
4152        # Return the result.
4153        return $retVal;
4154    }
4155    
4156  =head3 ParseAssignment  =head3 ParseAssignment
4157    
4158  Parse annotation text to determine whether or not it is a functional assignment. If it is,  Parse annotation text to determine whether or not it is a functional assignment. If it is,

Legend:
Removed from v.1.117  
changed lines
  Added in v.1.119

MCS Webmaster
ViewVC Help
Powered by ViewVC 1.0.3