miga-base 0.3.1.6 → 0.3.1.7

data/utils/enveomics/Scripts/JPlace.to_iToL.rb

@@ -1,306 +1,320 @@
 #!/usr/bin/env ruby
 
-#
 # @author: Luis M. Rodriguez-R
-# @update: Feb-28-2016
 # @license: artistic license 2.0
-#
 
 $:.push File.expand_path("../lib", __FILE__)
+require "enveomics_rb/enveomics"
 require "enveomics_rb/jplace"
 require "optparse"
 require "json"
 
-o = {:q=>false, :regex=>'^(?<dataset>.+?):.*', :area=>false, :norm=>:counts,
-   :olditol=>false}
-ARGV << '-h' if ARGV.size==0
+o = {q: false, regex: "^(?<dataset>.+?):.*", area: false, norm: :counts,
+  olditol: false}
+ARGV << "-h" if ARGV.size==0
 OptionParser.new do |opts|
-   opts.banner = "
+  opts.banner = "
 Generates iToL-compatible files from a .jplace file (produced by RAxML's EPA
 or pplacer), that can be used to draw pie-charts in the nodes of the reference
 tree.
 
 Usage: #{$0} [options]"
-   opts.separator ""
-   opts.separator "Mandatory"
-   opts.on("-i", "--in FILE", ".jplace input file containing the read placement."){ |v| o[:in]=v }
-   opts.on("-o", "--out FILE", "Base of the output files."){ |v| o[:out]=v }
-   opts.separator ""
-   opts.separator "Other Options"
-   opts.on("-u", "--unique STR", "Name of the dataset (if only one is used). Conflicts with -r and -s."){ |v| o[:unique]=v }
-   opts.on("-r", "--regex STR", "Regular expression capturing the sample ID (named dataset) in read names.",
-   	"By default: '#{o[:regex]}'. Conflicts with -s."){ |v| o[:regex]=v }
-   opts.on("-s", "--separator STR", "String separating the dataset name and the rest of the read name.",
-   	"It assumes that the read name starts by the dataset name. Conflicts with -r."){ |v| o[:regex]="^(?<dataset>.+?)#{v}" }
-   opts.on("-m", "--metadata FILE", "Datasets metadata in tab-delimited format with a header row.",
-   	"Valid headers: name (required), color (in Hex), size (# reads), norm (any float)."){ |v| o[:metadata]=v }
-   opts.on("-n", "--norm STR", %w[none counts size norm], "Normalization strategy. Must be one of:",
-   	"none: Direct read counts are reported without normalization.",
-	"count (default): The counts are normalized (divided) by the total counts per dataset.",
-	"size: The counts are normalized (divided) by the size column in metadata (must be integer).",
-	"norm: The counts are normalized (divided) by the norm column in metadata (can be any float)."){ |v| o[:norm]=v.to_sym }
-   opts.on("--old-itol",
-      "Generate output file using the old iToL format (pre v3.0)."
-      ){ |v| o[:olditol] = v }
-   opts.on("-c", "--collapse FILE", "Internal nodes to collapse (requires rootted tree)."){ |v| o[:collapse]=v }
-   opts.on("-a", "--area", "If set, the area of the pies is proportional to the placements. Otherwise, the radius is."){ o[:area]=true }
-   opts.on("-q", "--quiet", "Run quietly (no STDERR output)."){ o[:q] = true }
-   opts.on("-h", "--help", "Display this screen.") do
-      puts opts
-      exit
-   end
-   opts.separator ""
-   opts.separator "Quick how-to in 5 steps"
-   opts.separator "    1. Create the placement file using RAxML's EPA [1] or pplacer [2]. You can use any other software"
-   opts.separator "       producing a compliant .jplace file [3]. If you're using multiple datasets, include the name of"
-   opts.separator "       the dataset somewhere in the read names."
-   opts.separator "    2. If you have multiple datasets, it's convenient to create a metadata table. It's not necessary,"
-   opts.separator "       but it allows you to control the colors and the normalization method (see -m)."
-   opts.separator "    3. Execute this script passing the .jplace file created in step 1 (see -i).  If you have a single"
-   opts.separator "       dataset, use the option -u to give it a short name.  If you have multiple datasets, use the -s"
-   opts.separator "       or -r options to tell the script how to find the dataset name within the read name.  Note that"
-   opts.separator "       some programs (like CheckM) may produce nonstandard characters that won't be correctly parsed."
-   opts.separator "       To avoid this problem,  install iconv support (gem install iconv) before running this script"
-   opts.separator "       (currently "+(JPlace::Tree.has_iconv? ? "" : "NOT ")+"installed)."
-   opts.separator "    4. Upload the tree (.nwk file) to iToL  [4].  Make sure you check 'Keep internal node IDs' in the"
-   opts.separator "       advanced options.  In that same page, upload the dataset (.itol file), pick a name, and select"
-   opts.separator "       the data type 'Multi-value Bar Chart or Pie Chart'. If you used the -c option, upload the list"
-   opts.separator "       of nodes to collapse (.collapse file) in the 'Pre-collapsed clades' field (advanced options)."
-   opts.separator "    5. Open the tree. You can now see the names of the internal nodes. If you want to collapse nodes,"
-   opts.separator "       simply list the nodes to collapse and go back to step 3, this time using the -c option."
-   opts.separator ""
-   opts.separator "References"
-   opts.separator "    [1] SA Berger, D Krompass and A Stamatakis, 2011, Syst Biol 60(3):291-302."
-   opts.separator "        http://sysbio.oxfordjournals.org/content/60/3/291"
-   opts.separator "    [2] FA Matsen, RB Kodner and EV Armbrust, 2010, BMC Bioinf 11:538."
-   opts.separator "        http://www.biomedcentral.com/1471-2105/11/538/"
-   opts.separator "    [3] FA Matsen, NG Hoffman, A Gallagher and A Stamatakis, 2012, PLoS ONE 7(2):e31009."
-   opts.separator "        http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0031009"
-   opts.separator "    [4] I Letunic and P Bork, 2011, NAR 39(suppl 2):W475-W478."
-   opts.separator "        http://nar.oxfordjournals.org/content/39/suppl_2/W475.full"
-   opts.separator ""
+  opts.separator ""
+  opts.separator "Mandatory"
+  opts.on("-i", "--in FILE",
+    ".jplace input file containing the read placement."){ |v| o[:in]=v }
+  opts.on("-o", "--out FILE", "Base of the output files."){ |v| o[:out]=v }
+  opts.separator ""
+  opts.separator "Other Options"
+  opts.on("-u", "--unique STR",
+    "Name of the dataset (if only one is used). Conflicts with -r and -s."
+    ){ |v| o[:unique]=v }
+  opts.on("-r", "--regex STR",
+    "Regular expression capturing the sample ID (named dataset) in read names.",
+    "By default: '#{o[:regex]}'. Conflicts with -s."){ |v| o[:regex]=v }
+  opts.on("-s", "--separator STR",
+    "String separating the dataset name and the rest of the read name.",
+    "It assumes that the read name starts by the dataset name. Conflicts with -r."
+    ){ |v| o[:regex]="^(?<dataset>.+?)#{v}" }
+  opts.on("-m", "--metadata FILE",
+    "Datasets metadata in tab-delimited format with a header row.",
+    "Valid headers: name (required), color (in Hex), size (# reads), norm (any float)."
+    ){ |v| o[:metadata]=v }
+  opts.on("-n", "--norm STR", %w[none counts size norm],
+    "Normalization strategy. Must be one of:",
+    "none: Direct read counts are reported without normalization.",
+    "count (default): The counts are normalized (divided) by the total counts per dataset.",
+    "size: The counts are normalized (divided) by the size column in metadata (must be integer).",
+    "norm: The counts are normalized (divided) by the norm column in metadata (can be any float)."
+    ){ |v| o[:norm]=v.to_sym }
+  opts.on("--old-itol",
+    "Generate output file using the old iToL format (pre v3.0)."
+    ){ |v| o[:olditol] = v }
+  opts.on("-c", "--collapse FILE",
+    "Internal nodes to collapse (requires rootted tree)."){ |v| o[:collapse]=v }
+  opts.on("-a", "--area",
+    "If set, the area of the pies is proportional to the placements. Otherwise, the radius is."
+    ){ o[:area]=true }
+  opts.on("-q", "--quiet", "Run quietly (no STDERR output)."){ o[:q] = true }
+  opts.on("-h", "--help", "Display this screen.") do
+    puts opts
+    exit
+  end
+  opts.separator ""
+  opts.separator "Quick how-to in 5 steps"
+  opts.separator "    1. Create the placement file using RAxML's EPA [1] or pplacer [2]. You can use any other software"
+  opts.separator "       producing a compliant .jplace file [3]. If you're using multiple datasets, include the name of"
+  opts.separator "       the dataset somewhere in the read names."
+  opts.separator "    2. If you have multiple datasets, it's convenient to create a metadata table. It's not necessary,"
+  opts.separator "       but it allows you to control the colors and the normalization method (see -m)."
+  opts.separator "    3. Execute this script passing the .jplace file created in step 1 (see -i).  If you have a single"
+  opts.separator "       dataset, use the option -u to give it a short name.  If you have multiple datasets, use the -s"
+  opts.separator "       or -r options to tell the script how to find the dataset name within the read name.  Note that"
+  opts.separator "       some programs (like CheckM) may produce nonstandard characters that won't be correctly parsed."
+  opts.separator "       To avoid this problem,  install iconv support (gem install iconv) before running this script"
+  opts.separator "       (currently "+(JPlace::Tree.has_iconv? ? "" : "NOT ")+"installed)."
+  opts.separator "    4. Upload the tree (.nwk file) to iToL  [4].  Make sure you check 'Keep internal node IDs' in the"
+  opts.separator "       advanced options.  In that same page, upload the dataset (.itol file), pick a name, and select"
+  opts.separator "       the data type 'Multi-value Bar Chart or Pie Chart'. If you used the -c option, upload the list"
+  opts.separator "       of nodes to collapse (.collapse file) in the 'Pre-collapsed clades' field (advanced options)."
+  opts.separator "    5. Open the tree. You can now see the names of the internal nodes. If you want to collapse nodes,"
+  opts.separator "       simply list the nodes to collapse and go back to step 3, this time using the -c option."
+  opts.separator ""
+  opts.separator "References"
+  opts.separator "    [1] SA Berger, D Krompass and A Stamatakis, 2011, Syst Biol 60(3):291-302."
+  opts.separator "        http://sysbio.oxfordjournals.org/content/60/3/291"
+  opts.separator "    [2] FA Matsen, RB Kodner and EV Armbrust, 2010, BMC Bioinf 11:538."
+  opts.separator "        http://www.biomedcentral.com/1471-2105/11/538/"
+  opts.separator "    [3] FA Matsen, NG Hoffman, A Gallagher and A Stamatakis, 2012, PLoS ONE 7(2):e31009."
+  opts.separator "        http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0031009"
+  opts.separator "    [4] I Letunic and P Bork, 2011, NAR 39(suppl 2):W475-W478."
+  opts.separator "        http://nar.oxfordjournals.org/content/39/suppl_2/W475.full"
+  opts.separator ""
 end.parse!
 abort "-o is mandatory" if o[:out].nil?
 
 ##### CLASSES:
 
 class Dataset
-   attr_reader :name, :data
-   def initialize(name)
-      @name = name
-      @data = {:count=>0}
-   end
-   def count
-      self.datum :count
-   end
-   def add_count(n)
-      @data[:count] += n
-   end
-   def datum(k)
-      @data[k]
-   end
-   def add_datum(k, v)
-      @data[k] = v
-   end
-   def color
-      if @data[:color].nil?
-	 @data[:color] = '#' + (1 .. 3).map{ |i| sprintf("%02X", rand(255)) }.join('')
-      end
-      @data[:color].sub(/^#?/, '#')
-      self.datum :color
-   end
-   def size
-      self.datum :size
-   end
-   def norm
-      self.datum :norm
-   end
+  attr_reader :name, :data
+  def initialize(name)
+    @name = name
+    @data = {:count=>0}
+  end
+  def count
+    self.datum :count
+  end
+  def add_count(n)
+    @data[:count] += n
+  end
+  def datum(k)
+    @data[k]
+  end
+  def add_datum(k, v)
+    @data[k] = v
+  end
+  def color
+    if @data[:color].nil?
+      @data[:color] = "#" + (1 .. 3).map{ |i|
+        sprintf("%02X", rand(255)) }.join("")
+    end
+    @data[:color].sub(/^#?/, "#")
+    self.datum :color
+  end
+  def size
+    self.datum :size
+  end
+  def norm
+    self.datum :norm
+  end
 end
 
 class Metadata
-   attr_reader :datasets
-   def initialize
-      @datasets = {}
-   end
-   def load_table(file)
-      f = File.open(file, 'r')
-      h = f.gets.chomp.split(/\t/)
-      name_idx = h.find_index 'name'
-      color_idx = h.find_index 'color'
-      size_idx = h.find_index 'size'
-      norm_idx = h.find_index 'norm'
-      abort "The metadata table must contain a 'name' column." if name_idx.nil?
-      while ln = f.gets
-         vals = ln.chomp.split(/\t/)
-	 name = vals[name_idx]
-	 self[name] # Create sample, in case "name" is the only column
-	 self[name].add_datum(:color, vals[color_idx]) unless color_idx.nil?
-	 self[name].add_datum(:size, vals[size_idx].to_i) unless size_idx.nil?
-	 self[name].add_datum(:norm, vals[norm_idx].to_f) unless norm_idx.nil?
-      end
-      f.close
-   end
-   def [](name)
-      self << Dataset.new(name) unless @datasets.has_key?(name)
-      @datasets[name]
-   end
-   def <<(dataset)
-      @datasets[dataset.name] = dataset
-   end
-   def names
-      @datasets.keys
-   end
-   def colors
-      @datasets.values.map{ |d| d.color }
-   end
-   def data(k)
-      self.names.map{ |name| self[name].datum[k] }
-   end
-   def set_unique!(n)
-      u = self[n]
-      @datasets = {}
-      @datasets[n] = u
-   end
-   def size
-      self.datasets.length
-   end
+  attr_reader :datasets
+  def initialize
+    @datasets = {}
+  end
+  def load_table(file)
+    f = File.open(file, "r")
+    h = f.gets.chomp.split(/\t/)
+    name_idx = h.find_index "name"
+    color_idx = h.find_index "color"
+    size_idx = h.find_index "size"
+    norm_idx = h.find_index "norm"
+    abort "The metadata table must contain a 'name' column." if name_idx.nil?
+    while ln = f.gets
+      vals = ln.chomp.split(/\t/)
+      name = vals[name_idx]
+      self[name] # Create sample, in case "name" is the only column
+      self[name].add_datum(:color, vals[color_idx]) unless color_idx.nil?
+      self[name].add_datum(:size, vals[size_idx].to_i) unless size_idx.nil?
+      self[name].add_datum(:norm, vals[norm_idx].to_f) unless norm_idx.nil?
+    end
+    f.close
+  end
+  def [](name)
+    self << Dataset.new(name) unless @datasets.has_key?(name)
+    @datasets[name]
+  end
+  def <<(dataset)
+    @datasets[dataset.name] = dataset
+  end
+  def names
+    @datasets.keys
+  end
+  def colors
+    @datasets.values.map{ |d| d.color }
+  end
+  def data(k)
+    self.names.map{ |name| self[name].datum[k] }
+  end
+  def set_unique!(n)
+    u = self[n]
+    @datasets = {}
+    @datasets[n] = u
+  end
+  def size
+    self.datasets.length
+  end
 end
 
 ##### MAIN:
 begin
-   $stderr.puts "Parsing metadata." unless o[:q]
-   metadata = Metadata.new
-   metadata.load_table(o[:metadata]) unless o[:metadata].nil?
-   metadata.set_unique! o[:unique] unless o[:unique].nil?
+  $stderr.puts "Parsing metadata." unless o[:q]
+  metadata = Metadata.new
+  metadata.load_table(o[:metadata]) unless o[:metadata].nil?
+  metadata.set_unique! o[:unique] unless o[:unique].nil?
 
 
-   $stderr.puts "Loading jplace file." unless o[:q]
-   ifh = File.open(o[:in], 'r')
-   jplace = JSON.load(ifh)
-   ifh.close
-   
+  $stderr.puts "Loading jplace file." unless o[:q]
+  ifh = File.open(o[:in], 'r')
+  jplace = JSON.load(ifh)
+  ifh.close
 
-   $stderr.puts "Parsing tree." unless o[:q]
-   tree = JPlace::Tree.from_nwk(jplace["tree"])
-   
 
-   $stderr.puts "Parsing placements." unless o[:q]
-   JPlace::Placement.fields = jplace["fields"]
-   placements_n = 0
-   jplace["placements"].each do |placement|
-      JPlace::Node.link_placement(JPlace::Placement.new(placement))
-      placements_n += 1
-   end
-   $stderr.puts " #{placements_n} placements." unless o[:q]
-   tree.pre_order do |n|
-      n.placements.each do |p|
-	 p.nm.each do |r|
-	    m = (o[:unique].nil? ? (/#{o[:regex]}/.match(r[:n]) or
-	       abort "Cannot parse read name: #{r[:n]}, placed at edge #{n.index}") : {:dataset=>o[:unique]})
-	    metadata[ m[:dataset] ].add_count(r[:m])
-	 end
-      end
-   end
+  $stderr.puts "Parsing tree." unless o[:q]
+  tree = JPlace::Tree.from_nwk(jplace["tree"])
 
 
-   unless o[:collapse].nil?
-      $stderr.puts "Collapsing nodes." unless o[:q]
-      collapse = File.readlines(o[:collapse]).map do |ln|
-	 l = ln.chomp.split(/\t/)
-	 l[1] = l[0] if l[1].nil?
-	 l
-      end.inject({}) do |hash,ar|
-	 hash[ar[0]] = ar[1]
-	 hash
+  $stderr.puts "Parsing placements." unless o[:q]
+  JPlace::Placement.fields = jplace["fields"]
+  placements_n = 0
+  jplace["placements"].each do |placement|
+    JPlace::Node.link_placement(JPlace::Placement.new(placement))
+    placements_n += 1
+  end
+  $stderr.puts " #{placements_n} placements." unless o[:q]
+  tree.pre_order do |n|
+    n.placements.each do |p|
+      p.nm.each do |r|
+        m = (o[:unique].nil? ? (/#{o[:regex]}/.match(r[:n]) or
+          abort "Cannot parse read name: #{r[:n]}, placed at edge #{n.index}") :
+          {:dataset=>o[:unique]})
+        metadata[ m[:dataset] ].add_count(r[:m])
       end
-      f = File.open(o[:out] + ".collapse", 'w')
-      coll_n = 0
-      tree.pre_order do |n|
-	 if collapse.keys.include? n.cannonical_name
-	    n.collapse!
-	    n.name = collapse[n.cannonical_name]
-	    f.puts n.name
-	    coll_n += 1
-	 end
+    end
+  end
+
+
+  unless o[:collapse].nil?
+    $stderr.puts "Collapsing nodes." unless o[:q]
+    collapse = File.readlines(o[:collapse]).map do |ln|
+      l = ln.chomp.split(/\t/)
+      l[1] = l[0] if l[1].nil?
+      l
+    end.inject({}) do |hash,ar|
+      hash[ar[0]] = ar[1]
+      hash
+    end
+    f = File.open(o[:out] + ".collapse", "w")
+    coll_n = 0
+    tree.pre_order do |n|
+      if collapse.keys.include? n.cannonical_name
+        n.collapse!
+        n.name = collapse[n.cannonical_name]
+        f.puts n.name
+        coll_n += 1
       end
-      f.close
-      $stderr.puts " #{coll_n} nodes collapsed (#{collapse.length} requested)." unless o[:q]
-   end
+    end
+    f.close
+    $stderr.puts " #{coll_n} nodes collapsed (#{collapse.length} requested)." unless o[:q]
+  end
    
 
-   $stderr.puts "Estimating normalizing factors by #{o[:norm].to_s}." unless o[:q] or o[:norm]==:none
-   case o[:norm]
-   when :none
-      metadata.datasets.values.each{ |d| d.add_datum :norm, 1.0 }
-   when :counts
-      metadata.datasets.values.each{ |d| d.add_datum :norm, d.count.to_f }
-   when :size
-      abort "Column 'size' required in metadata." if metadata.datasets.values[0].size.nil?
-      metadata.datasets.values.each{ |d| d.add_datum :norm, d.size.to_f }
-   when :norm
-      abort "Column 'norm' required in metadata." if metadata.datasets.values[0].norm.nil?
-   end
-   max_norm = metadata.datasets.values.map{ |d| d.norm }.max
+  $stderr.puts "Estimating normalizing factors by #{o[:norm].to_s}." unless o[:q] or o[:norm]==:none
+  case o[:norm]
+  when :none
+    metadata.datasets.values.each{ |d| d.add_datum :norm, 1.0 }
+  when :counts
+    metadata.datasets.values.each{ |d| d.add_datum :norm, d.count.to_f }
+  when :size
+    abort "Column 'size' required in metadata." if metadata.datasets.values[0].size.nil?
+    metadata.datasets.values.each{ |d| d.add_datum :norm, d.size.to_f }
+  when :norm
+    abort "Column 'norm' required in metadata." if metadata.datasets.values[0].norm.nil?
+  end
+  max_norm = metadata.datasets.values.map{ |d| d.norm }.max
 
 
-   $stderr.puts "Generating iToL dataset." unless o[:q]
-   f = File.open(o[:out] + ".itol.txt", "w")
-   if o[:olditol]
-      f.puts "LABELS\t" + metadata.names.join("\t")
-      f.puts "COLORS\t" + metadata.colors.join("\t")
-   else
-      f.puts "DATASET_PIECHART"
-      f.puts "SEPARATOR TAB"
-      f.puts "DATASET_LABEL\tReadPlacement"
-      f.puts "COLOR\t#1f2122"
-      f.puts "FIELD_LABELS\t" + metadata.names.join("\t")
-      f.puts "FIELD_COLORS\t" + metadata.colors.join("\t")
-      f.puts "DATA"
-   end
-   max_norm_sum,min_norm_sum,max_norm_n,min_norm_n = 0.0,Float::INFINITY,"",""
-   tree.pre_order do |n|
-      ds_counts = Hash.new(0.0)
-      n.placements.each do |p|
-	 p.nm.each do |r|
-	    m = (o[:unique].nil? ? (/#{o[:regex]}/.match(r[:n]) or
-	       abort "Cannot parse read name: #{r[:n]}, placed at edge #{n.index}") : {:dataset=>o[:unique]})
-	    ds_counts[ m[:dataset] ] += r[:m] / metadata[ m[:dataset] ].norm
-	 end
+  $stderr.puts "Generating iToL dataset." unless o[:q]
+  f = File.open(o[:out] + ".itol.txt", "w")
+  if o[:olditol]
+    f.puts "LABELS\t" + metadata.names.join("\t")
+    f.puts "COLORS\t" + metadata.colors.join("\t")
+  else
+    f.puts "DATASET_PIECHART"
+    f.puts "SEPARATOR TAB"
+    f.puts "DATASET_LABEL\tReadPlacement"
+    f.puts "COLOR\t#1f2122"
+    f.puts "FIELD_LABELS\t" + metadata.names.join("\t")
+    f.puts "FIELD_COLORS\t" + metadata.colors.join("\t")
+    f.puts "DATA"
+  end
+  max_norm_sum,min_norm_sum,max_norm_n,min_norm_n = 0.0,Float::INFINITY,"",""
+  tree.pre_order do |n|
+    ds_counts = Hash.new(0.0)
+    n.placements.each do |p|
+      p.nm.each do |r|
+        m = (o[:unique].nil? ? (/#{o[:regex]}/.match(r[:n]) or
+          abort "Cannot parse read name: #{r[:n]}, placed at edge #{n.index}") :
+          {:dataset=>o[:unique]})
+        ds_counts[ m[:dataset] ] += r[:m] / metadata[ m[:dataset] ].norm
       end
-      counts_sum = ds_counts.values.reduce(:+)
-      unless counts_sum.nil?
-         # In the area option, the radius is "twice" to make the smallest > 1 (since counts_sum is >= 1)
-	 radius = (o[:area] ? 2*Math.sqrt(counts_sum/Math::PI) : counts_sum)*max_norm
-	 f.puts n.cannonical_name +
-	    "#{"\t0.5" unless o[:olditol]}\t#{"R" if o[:olditol]}" +
-	    radius.to_i.to_s + "\t" +
-	    metadata.names.map{ |n| ds_counts[n] }.join("\t")
-	 if counts_sum > max_norm_sum
-	    max_norm_n = n.cannonical_name
-	    max_norm_sum = counts_sum
-	 end
-	 if counts_sum < min_norm_sum
-	    min_norm_n = n.cannonical_name
-	    min_norm_sum = counts_sum
-	 end
+    end
+    counts_sum = ds_counts.values.reduce(:+)
+    unless counts_sum.nil?
+      # In the area option, the radius is "twice" to make the smallest > 1 (since counts_sum is >= 1)
+      radius = (o[:area] ? 2*Math.sqrt(counts_sum/Math::PI) : counts_sum)*max_norm
+      f.puts n.cannonical_name +
+        "#{"\t0.5" unless o[:olditol]}\t#{"R" if o[:olditol]}" +
+        radius.to_i.to_s + "\t" +
+        metadata.names.map{ |n| ds_counts[n] }.join("\t")
+      if counts_sum > max_norm_sum
+        max_norm_n = n.cannonical_name
+        max_norm_sum = counts_sum
       end
-   end
-   f.close
-   units = {:none=>'', :counts=>' per million placements', :size=>' per million reads', :norm=>' per normalizing unit'}
-   $stderr.puts " The pie #{o[:area] ? 'areas' : 'radii'} are proportional to the placements#{units[o[:norm]]}." unless o[:q]
-   $stderr.puts " The minimum radius (#{min_norm_n}) represents #{min_norm_sum*(([:none, :norm].include? o[:norm]) ? 1 : 1e6)} placements#{units[o[:norm]]}." unless o[:q]
-   $stderr.puts " The maximum radius (#{max_norm_n}) represents #{max_norm_sum*(([:none, :norm].include? o[:norm]) ? 1 : 1e6)} placements#{units[o[:norm]]}." unless o[:q]
-   
+      if counts_sum < min_norm_sum
+        min_norm_n = n.cannonical_name
+        min_norm_sum = counts_sum
+      end
+    end
+  end
+  f.close
+  units = {none: "", counts: " per million placements",
+    size: " per million reads", norm: " per normalizing unit"}
+  $stderr.puts " The pie #{o[:area] ? "areas" : "radii"} are proportional to the placements#{units[o[:norm]]}." unless o[:q]
+  $stderr.puts " The minimum radius (#{min_norm_n}) represents #{min_norm_sum*(([:none, :norm].include? o[:norm]) ? 1 : 1e6)} placements#{units[o[:norm]]}." unless o[:q]
+  $stderr.puts " The maximum radius (#{max_norm_n}) represents #{max_norm_sum*(([:none, :norm].include? o[:norm]) ? 1 : 1e6)} placements#{units[o[:norm]]}." unless o[:q]
 
-   $stderr.puts "Re-formatting tree for iToL." unless o[:q]
-   f = File.open(o[:out]+'.nwk', "w")
-   f.puts tree.to_s+';'
-   f.close
+
+  $stderr.puts "Re-formatting tree for iToL." unless o[:q]
+  f = File.open(o[:out] + ".nwk", "w")
+  f.puts tree.to_s + ";"
+  f.close
 
 rescue => err
-   $stderr.puts "Exception: #{err}\n\n"
-   err.backtrace.each { |l| $stderr.puts l + "\n" }
-   err
+  $stderr.puts "Exception: #{err}\n\n"
+  err.backtrace.each { |l| $stderr.puts l + "\n" }
+  err
 end
 
-