Diff of /Sprout/SimLoad.pl

-revision 1.3, Fri Jan  6 20:35:01 2006 UTC
+revision 1.7, Wed Feb  1 03:18:41 2006 UTC
 Line 4
  use CGI;
  use Tracer;
  use Genome;
+ use DBKernel;
  use SimBlocks;
  use File::Path;
  use BasicLocation;
-Line 18
+Line 19
  database (B<load>).
  The script takes a single parameter-- a directory name.
- The default directory name is taken from the config file
+ The default directory name is C<"$FIG_Config::data/CloseStrainSets">.
- parameter C<$fig_config::SimBlastData>. The output files
+ The input files should be in subdirectories called
- will be produced in the similarity block data directory
+ C<Block> under the subdirectories of the input directory.
- C<$fig_config::SimBlockData>, which will be created if
+ The subdirectory names themselves are considered the name
- it does not exist. The input directory and all its
+ of the close-strain set. So, for example
+     Data/CloseStrainSets/Vibrio/Block
+ would be presumed to contain the Vibrio strains.
+ The output files will be produced in the similarity block
+ data directory C<$fig_config::SimBlockData>, which will be
+ created if it does not exist. The input directory and all its
  subdirectories will be processed for input files.
  In addition to the directory name, the following
-Line 33
+Line 42
  =item trace
  Trace level for output messages. A higher number means more
- messages. The default is C<2>. Trace messages are sent to
+ messages. The default is C<3>. Trace messages are sent to
  the file C<trace.log> in the B<$FIG_Config::tmp>
  directory.
+ =item sql
+ If specified, SQL activity will be traced at the specified
+ trace level.
  =item load
  C<yes> to load the data into the database, else C<no>.
-Line 47
+Line 61
  C<yes> to generate output files from input files, else
  C<no>. The default is C<yes>.
+ =item createDB
+ If specified, the database will be dropped and
+ re-created before loading.
  =back
  For example, the following command line will process the
- input files in the C</Users/fig/BlastData> directory and
+ input files in the C</Users/fig/BlastData> directory tree
- run at a trace level of 3.
+ and run at a trace level of 3.
  C<< SimLoad -trace=3 /Users/fig/BlastData >>
-Line 63
+Line 82
  =head2 Input Directory
- The following files must exist in each input directory.
+ The following files must exist in each C<Block> directory
+ under the input directory.
  =over 4
-Line 280
+Line 300
  my $TRAILER = 999999999;
  # Parse the command line.
- my ($options, @arguments) = Tracer::ParseCommand({ trace => 1,
+ my ($options, @arguments) = StandardSetup(['SimBlocks'],
-                                                   load => 'yes',
+                                                 { load => ['yes', "'no' to suppress loading the database"],
-                                                   generate => 'yes'},
+                                                   generate => ['yes', "'no' to suppress generating the load files"],
+                                                   createDB => [0, 'drop and create the database before loading']
+                                                 },
+                                                 "directoryName",
                                                   @ARGV);
  # Extract the directory name from the argument array.
- my $inDirectoryTree = $FIG_Config::simBlastData;
+ my $inDirectoryTree = ($arguments[0] ? Cwd::abs_path($arguments[0]) : "$FIG_Config::data/CloseStrainSets");
- if ($arguments[0]) {
+ # Check to see if we need to create the database.
-     $inDirectoryTree = Cwd::abs_path($arguments[0]);
+ if ($options->{createDB}) {
- }
+     Trace("Creating database.") if T(2);
- # Set up tracing.
+     DBKernel::CreateDB($FIG_Config::simBlocksDB);
- my $traceLevel = $options->{trace};
+ }
- TSetup("$traceLevel Tracer ERDB SimBlocks DBKernel SQL", "+>$FIG_Config::temp/trace.log");
  # Get the output directory.
  my $outDirectory = $FIG_Config::simBlocksData;
  # Insure that it exists.
  if (! -d $outDirectory) {
      Trace("Creating output directory $outDirectory.") if T(2);
      mkpath($outDirectory);
- } else {
+ } elsif ($options->{generate} eq 'yes') {
-     # Here we have an output directory already. Clear any
+     # Here we have an output directory already and are going to generate new
-     # leftover data from previous runs.
+     # load files. Clear any leftover data from previous runs.
      my @files = grep { $_ =~ /.dtx$/ } Tracer::OpenDir($outDirectory);
      my $numFiles = @files;
      if ($numFiles > 0) {
-Line 322
+Line 344
          Confess("Input directory \"$inDirectoryTree\" not found.");
      }
      # Loop through the subdirectories.
-     for my $inputDirectory (Tracer::OpenDir($inDirectoryTree, 0)) {
+     for my $inputDirectory (Tracer::OpenDir($inDirectoryTree, 1)) {
-         # Verify that this is a directory. If it's ".", we use
+         # Verify that this is a directory.
-         # the top-level directory.
+         my $inDirectory = "$inDirectoryTree/$inputDirectory/Blocks";
-         my $inDirectory = ($inputDirectory eq "." ? $inDirectoryTree :
+         if (-d $inDirectory) {
-                                                     "$inDirectoryTree/$inputDirectory");
-         if (($inputDirectory !~ /\../) && -d $inDirectory) {
              # Here we have a directory to process. Check for a genome
              # file.
              my $genomeFileName = "$inDirectory/Genome.tbl";
-Line 336
+Line 356
              } else {
                  # Now we can process the directory and accumulate the error
                  # count.
-                 $errorCount += ProcessDirectory($inDirectory, $outDirectory);
+                 $errorCount += ProcessDirectory($inDirectory, $outDirectory, $inputDirectory);
                  $dirCount++;
              }
          }
-Line 361
+Line 381
  # Process a single input directory.
  sub ProcessDirectory {
-     my ($inDirectory, $outDirectory) = @_;
+     my ($inDirectory, $outDirectory, $groupName) = @_;
      Trace("Processing directory $inDirectory.") if T(2);
      # Our first task is to copy the genome data to the output directory
-     # and add them to the genome list.
+     # and add the genomes to the genome list.
      my %genomes = ();
      Open(\*GENOMESIN, "<$inDirectory/Genome.tbl");
      Open(\*GENOMESOUT, ">>$outDirectory/Genome.dtx");
-Line 374
+Line 394
      # Loop through the input.
      while (my $genomeData = <GENOMESIN>) {
          # Echo the genome record to the output.
-         print GENOMESOUT $genomeData;
+         my @fields = Tracer::ParseRecord($genomeData);
+         print GENOMESOUT join("\t", @fields, $groupName). "\n";
          # Extract the genome ID.
-         my ($genomeID) = Tracer::ParseRecord($genomeData);
+         my $genomeID = $fields[0];
          # Store it in the genomes hash. We start with a value of 0. If
          # contig information for the genome is found, we change the value
          # to 1. When we're all done with the regions, we can check the
-Line 500
+Line 521
                  Trace("No regions found for block $blockID at $blocksCount in block input file.") if T(0);
                  $errorCount++;
              } else {
-                 Trace("$regionCounter regions found in block $blockID.") if T(2);
+                 Trace("$regionCounter regions found in block $blockID.") if T(4);
-                 # Write the block record.
+                 # Write the block record. Note that the block ID is prefixed by the group name to
+                 # make it unique.
                  my $variance = $blockData{"snip-count"} / $blockData{len};
-                 print BLOCKSOUT join("\t", $blockID, $blockData{description}, $blockData{len},
+                 print BLOCKSOUT join("\t", "$groupName:$blockID", $blockData{description}, $blockData{len},
                                       $blockData{"snip-count"}, $variance, $blockData{pattern}) . "\n";
                  # Find all the variance points in the block pattern. We'll use them to create
                  # the content strings for each region.
-Line 527
+Line 549
                                            $regionPeg{$region}, $location->Left, $content) . "\n";
                      print CONTAINSOUT join("\t", $location->Contig, $region,
                                             $location->Length, $location->Left) . "\n";
-                     print INCLUDESOUT join("\t", $blockID, $region);
+                     print INCLUDESOUT join("\t", "$groupName:$blockID", $region) . "\n";
                  }
                  # Finally, we need to connect this block to the genomes in which it occurs.
                  for my $genomeID (keys %genomesFound) {
-                     print INSTANCESOUT join("\t", $genomeID, $blockID) . "\n";
+                     print INSTANCESOUT join("\t", $genomeID, "$groupName:$blockID") . "\n";
                  }
                  # Count this block's regions.
                  $regionsCount += $regionCounter;
-Line 558
+Line 580
          $contigsCount++;
      }
      Trace("$contigsCount contigs found.") if T(2);
+     # Close the output files.
+     close CONTIGSOUT;
+     close CONSISTSOUT;
      # Now warn the user about all the genomes that didn't have blocks.
      for my $genomeID (keys %genomes) {
          if (! $genomes{$genomeID}) {

 Legend:



Removed from v.1.3
 


changed lines


 
Added in v.1.7
 Legend:



Removed from v.1.3
 


changed lines


 
Added in v.1.7
-Removed from v.1.3
+Added in v.1.7

MCS Webmaster	ViewVC Help
Powered by ViewVC 1.0.3