mspire 0.1.5 → 0.1.7

data/bin/gi2annot.rb

@@ -1,13 +1,6 @@
 #!/usr/bin/ruby -w
 
-require 'open-uri'
-require 'rexml/document'
-require 'rexml/streamlistener'
-
-BATCH_SIZE = 500
-
-$LOG = nil
-$ANNOTS = []
+require 'gi'
 
 if ARGV.size < 1
   puts "usage: #{File.basename(__FILE__)} <gi> ..."
@@ -15,108 +8,7 @@ if ARGV.size < 1
 end
 
 
-# db=
-# retstart=
-# retmax=
-
-class Listener
-  include REXML
-  include StreamListener
-  def initialize
-    @get_title = false
-  end
-
-  def tag_start(name, attributes)
-    #puts "NAME" + name
-    #p attributes
-    if name == "Item" && attributes["Name"] == "Title"
-      @get_title = true 
-    end
-  end
-  def text(text)
-    #puts "TEXT: " + text + @get_title.to_s
-    if @get_title
-      #puts "GETTING TITLE!"
-      $ANNOTS.push text.chomp
-      @get_title = false
-    end
-  end
-end
-
-
-# Returns a list of Annotation strings
-def parse_etool_output(handle)
-  listener = Listener.new
-  parser = REXML::Parsers::StreamParser.new(handle, listener)
-  parser.parse 
-   
-  $ANNOTS
-end
-
-
-#$LOG = File.open("log.log", "w")
-
 gis = ARGV.to_a.dup
 
-while(true) do
-  batch = gis.slice!(0..BATCH_SIZE)
-  if batch.size == 0 then break end
-  string = batch.join(",")
-  url = "http://eutils.ncbi.nlm.nih.gov/entrez/eutils/esummary.fcgi?db=protein&retmode=xml&id=#{string}"
-  #puts url
-  annots = []
-  open(url) do |handle|
-    annots = parse_etool_output(handle)
-  end
-  puts annots.join("\n")
-end
-
-
-#$LOG.close
-
-
-
-
-=begin
-
-<?xml version="1.0" encoding="ISO-8859-1"?>
-<!DOCTYPE eSummaryResult PUBLIC "-//NLM//DTD eSummaryResult, 11 May 2002//EN" "http://www.ncbi.nlm.nih.gov/entrez/query/DTD/eSummary_041029.dtd">
-<eSummaryResult>
-
-<DocSum>
-<Id>24115498</Id>
-<Item Name="Caption" Type="String">NP_710008</Item>
-<Item Name="Title" Type="String">chaperonin GroEL [Shigella flexneri 2a str. 301]</Item>
-<Item Name="Extra" Type="String">gi|24115498|ref|NP_710008.1|[24115498]</Item>
-<Item Name="Gi" Type="Integer">24115498</Item>
-<Item Name="CreateDate" Type="String">2002/10/16</Item>
-
-<Item Name="UpdateDate" Type="String">2006/04/03</Item>
-<Item Name="Flags" Type="Integer">512</Item>
-<Item Name="TaxId" Type="Integer">198214</Item>
-<Item Name="Status" Type="String">live</Item>
-<Item Name="ReplacedBy" Type="String"></Item>
-<Item Name="Comment" Type="String"><![CDATA[  ]]></Item>
-</DocSum>
-
-
-<DocSum>
-<Id>434011</Id>
-<Item Name="Caption" Type="String">CAA24741</Item>
-
-<Item Name="Title" Type="String">unnamed protein product [Escherichia coli]</Item>
-<Item Name="Extra" Type="String">gi|434011|emb|CAA24741.1|[434011]</Item>
-<Item Name="Gi" Type="Integer">434011</Item>
-<Item Name="CreateDate" Type="String">1983/12/06</Item>
-<Item Name="UpdateDate" Type="String">2005/04/18</Item>
-<Item Name="Flags" Type="Integer">0</Item>
-<Item Name="TaxId" Type="Integer">562</Item>
-<Item Name="Status" Type="String">live</Item>
-<Item Name="ReplacedBy" Type="String"></Item>
-
-<Item Name="Comment" Type="String"><![CDATA[  ]]></Item>
-</DocSum>
-
-</eSummaryResult>
+puts( GI.gi2annot(gis).join("\n") )
 
-=end

data/bin/id_class_anal.rb

@@ -4,6 +4,7 @@ require 'spec_id'
 require 'generator'
 require 'optparse'
 require 'ostruct'
+require 'roc'
 
 def file_noext(file)
   file.sub(/#{Regexp.escape(File.extname(file))}$/, '')
@@ -21,7 +22,8 @@ jtplot_file = jtplot_base + '.toplot'
 OptionParser.new do |op|
   op.on("-p", "--prefix PREFIX", "prefix for false positive proteins") {|v| opt.p = v.split(',') }
   op.on("-j", "--jtplot", "output file '#{jtplot_file}' for jtp plotting program") {|v| opt.j = v }
-  op.on("-e", "--peptides", "runs a full analysis on peptides") {|v| opt.e = v }
+#  op.on("-e", "--peptides", "runs a full analysis on peptides") {|v| opt.e = v }
+  op.on("-a", "--area", "outputs area under the curve") {|v| opt.a = v }
 end.parse!
 
 if ARGV.size < 1
@@ -32,55 +34,59 @@ if ARGV.size < 1
   probabilities) or protein_prophet-prot.xml file which has been run with
   decoy proteins.  
   
-  Outputs tp's, precision, and the false positive rate [as calculated by Gygi
-  2*(#mod/(#norm+#mod))].  Each of these will be in a column with a label at
-  the top.  Outputs columns (delimited by '\\t') to STDOUT.
-  To capture to file: #{File.basename(__FILE__)} protein_file.xml > out.csv
+  Outputs tp's and precision.  
+  [The false positive predictive rate (FPPR) is 1 - precision]  
+  The two columns will be labeled at the top. 
+  (delimited by '\\t') to STDOUT.  To capture to file:
+  #{File.basename(__FILE__)} protein_file.xml > out.csv
 
-  Also takes gzipped (extension: xml.gz) files. 
-  
   OPTIONS:
   <s> = string
   -p  --prefix <s[,s...]>  Prefix(s) by which to determine decoy proteins (default #{def_pre})
   -j  --jtplot        outputs #{jtplot_file} for plotting by plot.rb
                       [% plot.rb -w lp --yrange n0.1:1.1 --noenhanced <file> ]
+  -a  --area          outputs area under the curve instead of tps/precision
   
   NOTE: protein prophet files not yet functional!!!
   ABBR:
     TP = True Positives
     FP = False Positives
     Prec = Precision = TP/(TP+FP)
-    FPR = False Positive Rate (as defined by Gygi) 2*[FP/(TP+FP)]
   "
   exit
 end
 
+###########################################################
+# I DON"T think option -e is functional yet...
+###########################################################
+
 files = ARGV.to_a
 
 out = nil
 if opt.j
   out = File.open(jtplot_file, "w")
-  lines = ['XYData', jtplot_base, "Classification Analysis", "Num TPs", "(Prec|FPR)"]
+  lines = ['XYData', jtplot_base, "Classification Analysis", "Num Hits", "Precision"]
   lines.each {|l| out.puts l}
 end
 
 headings = files.collect do |file|
-  %w(TP Prec FPR).collect {|v| v + " (#{file_noext(file)})" }
+  %w(TP Precision).collect {|v| v + " (#{file_noext(file)})" }
 end
-#headings = ["# True Positives", "Precision (TP/(TP+FP))", "FP Rate 2*(FP/(TP+FP))"]
-puts headings.flatten.join(delimiter)
 
 all_arrs = []
 files.each_with_index do |file,i|
   sp = SpecID.new(file)
-  #puts sp.prots.first.respond_to? 
-  if opt.e
-    headers = ["#{file_noext(file)} Precision [TP/(TP+FP)]", ]
-    arrs = sp.tps_and_precision_and_fpr2_times2_for_prob(opt.p[i])
-  else
-    headers = ["#{file_noext(file)} Precision [TP/(TP+FP)]", "#{file_noext(file)} FPR [FP/(FP+TP)]"]
-    arrs = sp.tps_and_precision_and_fpr2_times2_for_prob(opt.p[i])
+  headers = [file_noext(file)]
+  arrs = sp.num_hits_and_ppv_for_prob(opt.p[i])
+  
+  if opt.a
+    (num_hits, prec) = arrs
+    roc = ROC.new
+    prec_area = roc.area_under_curve(num_hits, prec)
+    puts "#{file} (area under curve [num_hits, precision])"
+    puts "Prec [#TPPrec = TP/(TP+FP)]:\t#{prec_area}"
   end
+
   all_arrs.push(*arrs)
 
   lns = []
@@ -95,8 +101,12 @@ files.each_with_index do |file,i|
   end
 end
 
-SyncEnumerator.new(*all_arrs).each do |row|
-  puts row.join(delimiter)
+
+unless opt.a
+  puts headings.flatten.join(delimiter)
+  SyncEnumerator.new(*all_arrs).each do |row|
+    puts row.join(delimiter)
+  end
 end
 
 out.close if opt.j

data/bin/id_precision.rb

@@ -13,7 +13,7 @@ opts = OptionParser.new do |op|
   op.banner = "usage: #{File.basename(__FILE__)} prefix bioworks.xml"
   op.separator ""
   op.separator "takes Bioworks 3.2 xml output files (with probabilities)"
-  op.separator "rank orders the probabilities and outputs tp's and sensitivity"
+  op.separator "rank orders the probabilities and outputs num hits and precision"
   op.separator "Also takes gzipped (xml.gz) files labeled as such"
   op.separator ""
   op.separator "Outputs a comma separated value to STDOUT (.csv)"
@@ -50,10 +50,11 @@ tp_obj.peps = tp
 two_lists = [tp_obj, fp_obj].map do |obj|
   list = []
   list.push( obj.pep_probs_by_pep_prots )
-  list.push( obj.pep_probs_by_seq_charge )
+
+  list.push( obj.pep_probs_by_bn_seq_charge )
   # These each have a by_min and a by_top10
-  list.push(*( obj.pep_probs_by_scan ) )
-  list.push(*( obj.pep_probs_by_scan_charge ) )
+  list.push(*( obj.pep_probs_by_bn_scan ) )
+  list.push(*( obj.pep_probs_by_bn_scan_charge ) )
   list
 end
 
@@ -61,19 +62,22 @@ end
 headings = ["PepProts", "SeqCharge", "Scan(TopHit)", "Scan(Top10)", "ScanCharge(TopHit)", "ScanCharge(Top10)"]
 csv_headings = []
 headings.each do |head|
-  csv_headings << head + ": TP"
+  csv_headings << head + ": NH"
   csv_headings << head + ": PR"
 end
 
 pairs = two_lists[0].zip two_lists[1]
 
-roc = ROC.new
+roc = DecoyROC.new
 x_y= []
 area_under_curve = []
 #start_x = []
 #end_x = []
 pairs.each do |pair|
-  x,y = roc.tps_and_precision(pair[0], pair[1])
+  #x,y = roc.pred_and_tps_and_ppv(pair[0], pair[1])
+  (num_hits, tps, ppv) = roc.pred_and_tps_and_ppv(pair[0], pair[1])
+  x = num_hits
+  y = ppv
   if $AREAS_ONLY
     x.unshift 0
     y.unshift 1.0
@@ -99,7 +103,7 @@ end
 # X axis is the number of peptides id# (i.e., # of peps in TP db)
 # Y axis is the precision = TP/(TP+FP)
 
-## Make some legend comments at the top of the file:
+puts "#  NH = number of hits"
 puts "#  TP = true positives"
 puts "#  FP = false positives"
 puts "#  PR = precision = TP/(TP+FP)"

data/bin/{false_positive_rate.rb → precision.rb}

@@ -2,4 +2,4 @@
 
 require 'spec_id'
 
-SpecID.false_positive_rate(ARGV)
+SpecID.precision(ARGV)

data/bin/protein_summary.rb

@@ -8,6 +8,7 @@ require 'spec_id'
 
 #############################################################
 # GLOBALS:
+PRECISION_PROGRAM_BASE = 'precision'
 DEF_PREFIX = "INV_"
 DEF_PERCENT_FP = "5.0"
 #############################################################
@@ -62,7 +63,7 @@ class Runner
             background-color: #FF0000;
             color: #FFFFFF
         }
-        div.file_info, div.software, div.fpr, div.num_proteins{
+        div.file_info, div.software, div.fppr, div.num_proteins{
             margin-left: 20px;
             margin-top: 20px;
         }
@@ -173,38 +174,38 @@ class Runner
   end
 
   # assumes that these are sorted on probability
-  # desired_fpr is a float
-  # returns [number_of_prots, actual_fpr]
-  def num_prots_above_fpr(prots, desired_fpr)
-    current_fpr_rate_percent = 0.0
-    previous_fpr_rate_percent = 0.0
+  # desired_fppr is a float
+  # returns [number_of_prots, actual_fppr]
+  def num_prots_above_fppr(prots, desired_fppr)
+    current_fppr_rate_percent = 0.0
+    previous_fppr_rate_percent = 0.0
     current_sum_one_minus_prob = 0.0
-    proteins_within_fpr = 0
-    actual_fpr = nil
+    proteins_within_fppr = 0
+    actual_fppr = nil
     already_found = false
     prot_cnt = 0
     prots.each do |prot|
       prot_cnt += 1
       # SUM(1-probX)/#prots
       current_sum_one_minus_prob += 1.0 - prot._probability.to_f
-      current_fpr_rate_percent = (current_sum_one_minus_prob / prot_cnt) * 100
+      current_fppr_rate_percent = (current_sum_one_minus_prob / prot_cnt) * 100
 
-      if current_fpr_rate_percent > desired_fpr && !already_found
-        actual_fpr = previous_fpr_rate_percent
-        proteins_within_fpr = prot_cnt
+      if current_fppr_rate_percent > desired_fppr && !already_found
+        actual_fppr = previous_fppr_rate_percent
+        proteins_within_fppr = prot_cnt
         already_found = true
       end
-      previous_fpr_rate_percent = current_fpr_rate_percent
+      previous_fppr_rate_percent = current_fppr_rate_percent
     end
-    [proteins_within_fpr, actual_fpr]
+    [proteins_within_fppr, actual_fppr]
   end
 
-    ####    #readable_previous_fpr_rate_percent = sprintf("%.2f", previous_fpr_rate_percent)
+    ####    #readable_previous_fppr_rate_percent = sprintf("%.2f", previous_fppr_rate_percent)
 
   # returns a string of the table rows
   # false_positive_rate (give as a %) is the cutoff mark
-  # returns the number of proteins at the desired_fpr (if given)
-  def table_rows(uniq_prots, prefix, false_positive_rate_percent, num_cols, desired_fpr, actual_percent_fp, peptide_count_filename=nil)
+  # returns the number of proteins at the desired_fppr (if given)
+  def table_rows(uniq_prots, prefix, false_positive_rate_percent, num_cols, desired_fppr, actual_percent_fp, peptide_count_filename=nil)
     prot_cnt = 0
     uniq_prots.map do |prot| 
       tr do
@@ -267,18 +268,20 @@ class Runner
     "<div class=\"software\"><h3>Software Information</h3>#{yield}<br/>Ruby package: #{mspire_version}<br/>Command: #{[File.basename(__FILE__), *@orig_argv].join(" ")}</div>"
   end
 
-  def proph_output(file, outfn, opt, fpr_output_as_html)
+  def proph_output(file, outfn, opt, fppr_output_as_html)
     header_anchors = [at('#', 'number'), at('prob','protein probability (for Prophet, higher is better)'), at('ref', 'gi number if available (or complete reference)'), at('annotation', 'annotation from the fasta file'), at('%cov', 'percent of protein sequence covered by corresponding peptides'), at('peps', 'unique peptides identified (includes non-contributing peptides). Click number to show/hide'), at('#peps', 'total number of corresponding peptides that contributed to protein probability'),  at('%ids', 'fraction of correct dataset peptide identifications corresponding to protein')]
     num_cols = header_anchors.size
     theaders = ths(header_anchors)
 
     root = AXML.parse_file(file)
     prots = []
-    ## find the min_prob at a fpr of XX
-    min_prob_redline = 1.01  # if no fpr is less than what they give, then all are redlined!
+    ## find the min_prob at a fppr of XX
+    min_prob_redline = 1.01  # if no fppr is less than what they give, then all are redlined!
 
-    if opt.c
+    if opt.c 
       actual_percent_fp = opt.c.to_f
+    elsif opt.cut_at
+      actual_percent_fp = opt.cut_at.to_f
     else
       actual_percent_fp = nil
     end
@@ -289,20 +292,24 @@ class Runner
     end
     uniq_prots = prots.hash_by(:_protein_name).map{|name,prot_arr| prot_arr.first }
     filtered_sorted_prots = filter_and_sort(uniq_prots, opt.f)
-    output_peptide_counts_file(filtered_sorted_prots, opt.peptide_count) if opt.peptide_count
 
     ## num proteins above cutoff (if opt.c)
     num_prots_html = ''
-    if opt.c
-      (num_prots, actual_fpr) = num_prots_above_fpr(filtered_sorted_prots, opt.c.to_f)
-      num_prots_html = num_prots_to_html(opt.c.to_f, actual_fpr, num_prots)
+    if opt.c || opt.cut_at
+      (num_prots, actual_fppr) = num_prots_above_fppr(filtered_sorted_prots, actual_percent_fp)
+      num_prots_html = num_prots_to_html(actual_percent_fp, actual_fppr, num_prots)
+    end
+    if opt.cut_at
+      filtered_sorted_prots = filtered_sorted_prots[0,num_prots]
     end
 
+    output_peptide_counts_file(filtered_sorted_prots, opt.peptide_count) if opt.peptide_count
+
     table_string = table do 
       tr{theaders} + table_rows(filtered_sorted_prots, opt.f, actual_percent_fp, num_cols, opt.c.to_f, actual_percent_fp, opt.peptide_count)
     end
-    er_info = opt.fpr ? error_info(file) : ""
-    html_pieces = [outfn, header, fpr_output_as_html, er_info, file_info(file), protproph_script_info, num_prots_html, table_string, trailer]
+    er_info = opt.precision ? error_info(file) : ""
+    html_pieces = [outfn, header, fppr_output_as_html, er_info, file_info(file), protproph_script_info, num_prots_html, table_string, trailer]
     print_html_pieces(*html_pieces)
   end # proph_output
 
@@ -311,7 +318,7 @@ class Runner
     "<a href=\"#prot#{prot_num}\" onclick=\"toggle_vis('#{prot_num}');\">#{peptide_sequences.size}</a><div id=\"#{prot_num}\" style=\"display:none;\">#{peptide_sequences.join(', ')}</div>"
   end
 
-  def bioworks_output(file, outfn, opt, fpr_output_as_html)
+  def bioworks_output(file, outfn, opt, fppr_output_as_html)
     header_anchors = [at('#', 'number'), at('prob','protein probability (for Bioworks, lower is better)'), at('ref', 'gi number if available (or complete reference)'), at('annotation', 'annotation from the fasta file'), at('%cov', 'percent of protein sequence covered by corresponding peptides'), at('peps', 'unique peptides identified (at any confidence) Click number to show/hide.'), at('#peps', 'total number of peptides seen (not unique)')]
     num_cols = header_anchors.size
     theaders = ths(header_anchors)
@@ -339,7 +346,7 @@ class Runner
     table_string = table do 
       tr{theaders} + rows
     end
-    print_html_pieces(outfn, header, fpr_output_as_html, file_info(file), bioworks_script_info(bio_obj), table_string, trailer)
+    print_html_pieces(outfn, header, fppr_output_as_html, file_info(file), bioworks_script_info(bio_obj), table_string, trailer)
   end # bioworks_output
 
   def num_prots_to_html(desired_cutoff, actual_cutoff, num_proteins)
@@ -358,7 +365,7 @@ class Runner
     #puts lines ?? is this supposed to be commented out?
     lines = lines.reject do |obj| obj =~ /\*{10}/ end
     lines.map! do |line| "#{line}<br/>" end
-    "<div class=\"fpr\">
+    "<div class=\"fppr\">
     <h3>Classification Analysis</h3>
     #{lines.join("\n")} 
     </div>"
@@ -366,7 +373,7 @@ class Runner
 
   # transforms the output string of file_as_decoy into html
   def prefix_as_decoy_to_html(string)  
-    "<div class=\"fpr\">
+    "<div class=\"fppr\">
     <h3>Classification Analysis</h3>
     </div>" +
     string
@@ -384,21 +391,18 @@ class Runner
       op.separator "    outputs: <file>.summary.html"
       op.separator ""
       op.on("-f", "--false <prefix>", "ignore proteins with prefix (def: #{DEF_PREFIX})") {|v| opt.f = v }
-      op.separator("             if --fpr then -f is used to specify a file or prefix")
-      op.separator("             to indicate false positives.")
+      op.on("-p", "--precision", "include the output from precision.rb") {|v| opt.p = v }
+      op.separator("             if --precision then -f is used to specify a file or prefix")
+      op.separator("             that indicates the false positives.")
       op.on("--peptide_count <filename>", "outputs text file with # peptides per protein") {|v| opt.peptide_count = v}
       op.separator ""
-      op.separator "Options for False Positive Rate:"
-      op.on("--fpr", "include output of false_positive_rate.rb,") {|v| opt.fpr = v}
-      op.separator("                                     type 'false_positive_rate.rb' for details")
-      op.separator("  These options are passed on:")
-      op.on("-g", "--gygi", "also show Gygi's estimate of FPR (2*FPR)") {|v| opt.g = v}
-      op.on("-p", "--prec", "also show precision (TP/(TP+FP))") {|v| opt.p = v}
-      op.on("-n", "--nofpr", "don't show FPR") {|v| opt.n = v}
-
+      op.separator "Options for #{PRECISION_PROGRAM_BASE}.rb :"
+      op.on("--#{PRECISION_PROGRAM_BASE}", "include output of #{PRECISION_PROGRAM_BASE}.rb,") {|v| opt.precision = v}
+      op.separator("                                     type '#{PRECISION_PROGRAM_BASE}.rb' for details")
       op.separator ""
       op.separator "Specific to ProteinProphet (with no concatenated DB):"
-      op.on("-c", "--cutoff percent", "displays red line at given % fpr") {|v| opt.c = v }
+      op.on("-c", "--cutoff percent", "includes FPR summary at given cutoff") {|v| opt.c = v }
+      op.on("--cut_at percent", "only reports proteins within FPR percent") {|v| opt.cut_at = v }
     end
 
     opts.parse!
@@ -408,31 +412,23 @@ class Runner
       exit
     end
 
-    fpr_output_as_html = ''
+    fppr_output_as_html = ''
     files = argv.to_a
     files.each do |file| 
       outfn = file.gsub(/\.xml$/, '.summary.html')
       ## False Positive Rate Calculation:
-      if opt.fpr
+      if opt.precision
         opt.o = outfn # won't actually be written over, but used
-        to_use_argv = create_false_positive_rate_argv(file, opt)
-        (out_string, opt, file_as_decoy) = SpecID::FalsePositiveRate.new.false_positive_rate(to_use_argv)
-        if file_as_decoy  ## need to wrap this guy up in some html
-          ## DISABLE the opt.f (it's a filename) so it doesn't interfere with
-          ## filtering:
-          opt.f = nil 
-
-          fpr_output_as_html = file_as_decoy_to_html(out_string)
-        else
-          fpr_output_as_html = prefix_as_decoy_to_html(out_string)
-        end
+        to_use_argv = create_precision_argv(file, opt)
+        (out_string, opt) = SpecID::Precision.new.precision(to_use_argv)
+        fppr_output_as_html = prefix_as_decoy_to_html(out_string)
       end
 
       case SpecID.file_type(file)
       when "protproph"
-        proph_output(file, outfn, opt, fpr_output_as_html)
+        proph_output(file, outfn, opt, fppr_output_as_html)
       when "bioworks"
-        bioworks_output(file, outfn, opt, fpr_output_as_html)
+        bioworks_output(file, outfn, opt, fppr_output_as_html)
       else 
         abort "filetype for #{file} not recognized!"
       end
@@ -440,15 +436,12 @@ class Runner
 
   end # method go
 
-  def create_false_positive_rate_argv(file, opt)
+  def create_precision_argv(file, opt)
     # include only those options specific
     new_argv = [file]
     if opt.f ; new_argv << '-f' << opt.f end
-    if opt.g ; new_argv << '-g' end
-    if opt.p ; new_argv << '-p' end
-    if opt.n ; new_argv << '-n' end
     if opt.o ; new_argv << '-o' << opt.o end
-    new_argv 
+    new_argv
   end
 
 end   # Runner

data/changelog.txt

@@ -0,0 +1,34 @@
+
+1. A couple of scripts and subroutines were hashing peptides but not on the file
+basename.  This would result in slightly incorrect results (any time there
+were overlapping scan numbers in multiple datasets, only the top one would be
+chosen).  The results would be correct for single runs.
+
+Output files that could be affected:
+*.top_per_scan.txt
+*.all_peps_per_scan.txt
+
+Scripts that could be affected:
+script/top_hit_per_scan.rb
+bin/filter_spec_id.rb
+script/filter-peps.rb
+bin/id_precision.rb
+
+Subroutines that were affected:
+spec_id.rb (pep_probs_by_* )
+spec_id.rb (top_peps_prefilter!)
+proph.rb uniq_by_seqcharge
+align.rb called uniq_by_seqcharge
+
+
+2. false_positive_rate.rb and protein_summary.rb (by extension) were using
+number of true positives on the x axis while in reality I was plotting the
+number of hits.  I've updated x axis labels to reflect this change.  In
+addition, since the term 'false positive rate' has such a distinct definition
+in classical ROC plots and binary statistics, I've decided to work primarily
+in terms of precision (TP/(TP+FP)).  I've purged the terms 'False Positive
+Rate' and 'FPR' from the package. It's been suggested that FP/(TP+FP) be
+called the False Positive Predictive Rate (FPPR).  I will probably implement
+this in a future release.
+
+

data/lib/align.rb

@@ -1,6 +1,5 @@
 
 require 'spec/mzxml/parser'
-require 'hash_by'
 require 'spec/msrun'
 require 'spec_id/proph'
 require 'vec'