From affdf5d3f0497fd89569c43b5eee4ef9d1ad779f Mon Sep 17 00:00:00 2001
From: Torsten Seemann <torsten.seemann@gmail.com>
Date: Thu, 12 Nov 2015 15:23:47 +1100
Subject: [PATCH] Add some helper scripts (only useful to MDU-PHL really)

---
 scripts/mdu-make_nullarbor_input.pl | 139 ++++++++++++++++++++++++++++
 scripts/xlsx2tsv.pl                 |  53 +++++++++++
 2 files changed, 192 insertions(+)
 create mode 100755 scripts/mdu-make_nullarbor_input.pl
 create mode 100755 scripts/xlsx2tsv.pl

diff --git a/scripts/mdu-make_nullarbor_input.pl b/scripts/mdu-make_nullarbor_input.pl
new file mode 100755
index 0000000..a0f21a5
--- /dev/null
+++ b/scripts/mdu-make_nullarbor_input.pl
@@ -0,0 +1,139 @@
+#!/usr/bin/env perl
+use strict;
+use warnings;
+use Getopt::Long;
+use File::Spec;
+use Cwd qw(abs_path);
+use Spreadsheet::Read;
+use Data::Dumper;
+use Cwd;
+
+my $id_re = '\b(\d{4}-\d{5})\b';
+my $read_re = '_R?([12])(?:_\d+)?.f';
+my $in  = '';
+my $dir = '/mnt/seq/MDU/READS';
+my $out = 'samples.tab';
+my $longid = 0;
+my $verbose = 0;
+
+sub usage {
+  print "$0 [--verbose] [--longid] [--fastqdir $dir] [--out $out] --in jobdetails.xlsx\n";
+  exit;
+}
+
+@ARGV or usage();
+
+GetOptions(
+  "help"        => \&usage,
+  "verbose!"    => \&verbose,
+  "in=s"        => \$in,
+  "fastqdir=s"  => \$dir,
+  "out=s"       => \$out,
+  "longid!"    => \$longid,
+  "id_regexp=s"    => \$id_re,
+  "read_regexp=s"    => \$read_re,
+) 
+or usage();
+
+if (!$in and @ARGV > 0) {
+  $in = shift @ARGV;
+  print STDERR "Guessing --in $in\n";
+}
+
+$in or die "need ID file with --in job.xls";
+-r $in or die "can't read ID file '$in'";
+
+$dir or die "need top-level folder containing FASTQ files with --dir";
+-d $dir or die "--dir '$dir' is not a directory";
+
+$out or die "need --out file to save results to";
+
+# compile regexps
+$id_re = qr"$id_re";
+$read_re = qr"$read_re";
+
+print STDERR "Scanning '$in' for MDU sample IDs...\n";
+
+my $book = ReadData( $in, cells=>0, strip=>3, attr=>1 );
+#print Dumper($book); exit;
+my @row = Spreadsheet::Read::rows($book->[1]);
+my %id;
+for my $row (@row) {
+  my $line = join(' ', grep { defined $_ } @$row); 
+  if ($line =~ $id_re) {
+    my $ID = $1;
+    $line =~ s/\s+/--/g;
+    $line =~ s/[_-]+$//;
+    $line =~ s/^[_-]+//;
+    $id{$ID} = $line;
+  }
+}
+
+printf STDERR "Found %d sample IDs:\n", scalar(keys %id);
+
+if (0 == keys %id) {
+  print STDERR "ERROR: no IDs found in '$in'\n";
+  exit -1;
+}
+
+print STDERR map { "$_\n" } sort keys %id if $verbose;
+
+print STDERR "Scanning '$dir' for read files...\n";
+
+my %want_id = (map { ($_ => 1) } keys %id);
+my %sample;
+
+open DIR, "find $dir -type f -name '*.f*q.gz' |";
+while (my $file = <DIR>) {
+  chomp $file;
+  if ($file =~ $id_re and exists $want_id{$1}) {
+    my $id = $1;
+    my(undef, undef, $name) = File::Spec->splitpath($file);
+    if ($name =~ $read_re) {
+      my $read = $1;
+#      print STDERR "$id $read\n";
+      $sample{$id}{$read} = abs_path($file);
+      print STDERR "Found $id $read : $file\n" if $verbose;
+    }
+    else {
+      print STDERR "WARNING: found $id but not $read_re in $name\n";
+    }
+  }
+}
+
+print STDERR "Creating output file: $out\n";
+open OUT, '>', $out;
+
+#use Data::Dumper;
+#print STDERR Dumper(\%id);
+
+for my $id (sort keys %id) {
+  if (exists $sample{$id}{1} and exists $sample{$id}{2}) {
+    my $label = $longid ? $id{$id} : $id;
+    print STDERR "$id - both reads found, labelling as '$label'\n";
+    printf OUT join("\t", $label, $sample{$id}{1}, $sample{$id}{2})."\n";
+  }
+  elsif (!exists $sample{$id}{1} and !exists $sample{$id}{2}) {
+    print STDERR "$id - NO READ FOUNDS !!!\n";
+  }
+  elsif (!exists $sample{$id}{1}) {
+    print STDERR "$id - MISSING Read 1 FILE !!!\n";
+  }
+  elsif (!exists $sample{$id}{2}) {
+    print STDERR "$id - MISSING Read 2 FILE !!!\n";
+  }
+  else {
+    print STDERR "$id - THIS LINE SHOULD NEVER BE REACHED\n";
+  }
+}
+
+print STDERR "Result in '$out'\n";
+print STDERR "Done.\n";
+
+my $name = qx(basename `pwd`);
+chomp $name;
+my $cmd = "nullarbor.pl --name $name --outdir nullarbor --input samples.tab --cpus 4 --mlst FIXME --ref FIXME";
+
+print STDERR "Your next command is probably this:\n$cmd\n";
+
+
diff --git a/scripts/xlsx2tsv.pl b/scripts/xlsx2tsv.pl
new file mode 100755
index 0000000..a46d759
--- /dev/null
+++ b/scripts/xlsx2tsv.pl
@@ -0,0 +1,53 @@
+#!/usr/bin/env perl
+use warnings;
+use strict;
+use Spreadsheet::Read;
+use Data::Dumper;
+
+my(@Options, $verbose, $sep, $informat);
+setOptions();
+
+my $book = ReadData( $ARGV[0], cells=>0, strip=>1 );
+
+my @row = Spreadsheet::Read::rows($book->[1]);
+
+for my $row (@row) {
+  my @r = map { $_ || '' } @$row;
+  print join($sep, @r),"\n";
+}
+
+#----------------------------------------------------------------------
+# Option setting routines
+
+sub setOptions {
+  use Getopt::Long;
+
+  @Options = (
+    {OPT=>"help",    VAR=>\&usage,             DESC=>"This help"},
+    {OPT=>"verbose!",  VAR=>\$verbose, DEFAULT=>0, DESC=>"Verbose output"},
+#    {OPT=>"informat=s",  VAR=>\$informat, DEFAULT=>"xlsx", DESC=>"Input format: xls xlsx csv ods sxc"},
+    {OPT=>"sep=s",  VAR=>\$sep, DEFAULT=>"\t", DESC=>"Output separator"},
+  );
+
+  #(!@ARGV) && (usage());
+
+  &GetOptions(map {$_->{OPT}, $_->{VAR}} @Options) || usage();
+
+  # Now setup default values.
+  foreach (@Options) {
+    if (defined($_->{DEFAULT}) && !defined(${$_->{VAR}})) {
+      ${$_->{VAR}} = $_->{DEFAULT};
+    }
+  }
+}
+
+sub usage {
+  print "Usage: $0 [options] [<] file.xlsx > file.csv\n";
+  foreach (@Options) {
+    printf "  --%-13s %s%s.\n",$_->{OPT},$_->{DESC},
+           defined($_->{DEFAULT}) ? " (default '$_->{DEFAULT}')" : "";
+  }
+  exit(1);
+}
+ 
+#----------------------------------------------------------------------