0.40-release

README.md

@@ -19,6 +19,7 @@ dadaist2  [options] -i INPUT_DIR -o OUTPUT_DIR
 Other options:
 * `-1`, `--for-tag` STRING, string identifying a file as being _forward_ (default: \_R1)
 * `-2`, `--rev-tag` STRING, string identifying a file as being _reverse_ (default: \_R2)
+* `-m`, `--metadata` FILE, metadata file in TSV format with first column as sample ID. If not supplied will be autogenerated using _makeSampleSheet_.
 * `-d`, `--database` FILE, database in gzipped FASTA format for taxonomy assignment (skipped if not provided)
 * `-q`, `--min-qual` FLOAT, minimum average quality at a position to set truncation start in DADA2 (default: 28)
 * `-r`, `--save-rds` will save a R data structure of the feature table
@@ -35,6 +36,8 @@ Other options:
 
 A directory containing the FASTQ files, that are usually gzipped. Paired-end are expected, but this will change in a future release. See the `data` directory for an example.
 
+An optional _metadata file_ can be used for secondary analyses, if not supplied will be autogenerated.
+
 ## Output
 
 The output directory will contain:

bin/dadaist2

@@ -40,6 +40,7 @@ my $opt_threads      = 2;
 my $opt_tax_db       = "skip";
 my $opt_save_rds     = 0;
 my $save_rds         = "no";
+my $opt_metadata_separator = "\t";
 
 # FASTP Preprocess
 # To be autocalculated:
@@ -151,6 +152,19 @@ if ($opt_tax_db ne 'skip') {
 	}
 }
 
+# Metadata?
+my $meta_samples;
+if ($opt_metadata) {
+  $meta_samples = check_metadata($opt_metadata);
+  copy($opt_metadata,
+		 File::Spec->catfile($opt_output_directory,'metadata.tsv')) || die " ERROR:\n Unable to copy metadù file to $opt_output_directory.\n";
+
+} else {
+  $opt_metadata = File::Spec->catfile($opt_output_directory, 'metadata.tsv');
+  my $meta_cmd = qq($RealBin/makeSampleSheet -i "$opt_input_directory" -o "$opt_metadata");
+  $S->run($meta_cmd, {candie => 1});
+  $meta_samples = check_metadata($opt_metadata);
+}
 $temp_dir = prepare_temporary_directories($temp_dir);
 
 $S->verbose("Threads: $opt_threads");
@@ -434,6 +448,7 @@ if ( $msa_exec->{'exit'} != 0 or $tree_exec->{'exit'} != 0) {
 }
 
 
+# Delete, or not, temp dir
 if ($opt_debug) {
   $S->verbose("Temporary directory _not_ deleted: $temp_dir");
 } else {
@@ -442,16 +457,13 @@ if ($opt_debug) {
 
 }
 
-my $output_message = "DADA2 finished, output files saved:\n";
+# Print relevant output files produced
+my $output_message = "Dadaist finished, output files saved:\n";
 for my $f (keys %output_files) {
   $output_message .= " * $f: " . $output_files{$f} . "\n";
 }
 $S->verbose($output_message);
 
-#copy(File::Spec->catfile($dada2_temp, 'stats.tsv'),
-#	 File::Spec->catfile($opt_output_directory, 'dada2_stats.tsv')) || die " ERROR:\n Unable to copy stats.tsv file from $dada2_temp to $opt_output_directory.\n";
-
-#dada2.rds  dada2.tsv  stats.tsv
 
 
 sub version {
@@ -462,7 +474,9 @@ sub version {
 sub usage {
 	say STDERR<<END;
 USAGE:
- dadaist -i INPUT_DIR -o OUTPUT_DIR [-t TEMP_DIR]
+ dadaist2 -i INPUT_DIR -o OUTPUT_DIR [-t TEMP_DIR]
+
+ dadaist2 --help for full manual
 END
 
  if ($_[0]) {
@@ -471,6 +485,24 @@ END
  }
 }
 
+sub check_metadata {
+  my $file = shift @_;
+  my $I;
+  if (not open ($I, '<', $file)) {
+    $S->verbose("Unable to load metadata from $file");
+    exit 1;
+  }
+  my $count_lines = 0;
+  my %samples = ();
+  while (my $line = readline($I)) {
+    $count_lines++;
+    next if ($line =~/^#/);
+    my @fields = split /$opt_metadata_separator/, $line;
+    $samples{ $fields[0] }++;
+  }
+  return \%samples;
+}
+
 sub get_qualified_positions {
 	# Scan a list of qualityes and return the boundaries of qualified quality
 	my $len = @_;
@@ -624,6 +656,11 @@ Directory containing the paired end files in FASTQ format, gzipped or not.
 
 Output directory (will be created).
 
+=item I<-m>, I<--metadata> FILE
+
+Metadata file in TSV format, first column must match sample IDs. If not supplied
+a template will be autogenerated using C<makeSampleSheet>.
+
 =item I<-d>, I<--database> DATABASE
 
 Reference database in gzipped FASTA format, specify 'skip' not to assign