miga-base 0.3.0.0 → 0.3.0.1

data/utils/enveomics/Scripts/lib/enveomics_rb/remote_data.rb

@@ -0,0 +1,74 @@
+
+#
+# @author: Luis M. Rodriguez-R
+# @license: artistic license 2.0
+#
+
+require "enveomics_rb/enveomics"
+use "restclient"
+use "json"
+
+class RemoteData
+  # Class-level variables
+  @@EUTILS = "https://eutils.ncbi.nlm.nih.gov/entrez/eutils"
+  @@EBIREST = "http://www.ebi.ac.uk/Tools"
+   
+  # Class-level methods
+  def self.eutils(script, params={}, outfile=nil)
+    response = nil
+    10.times do
+      begin
+        response = RestClient.get "#{@@EUTILS}/#{script}", {:params=>params}
+      rescue => err
+        warn "Request failed #{response.nil? ? "without error code" :
+          "with error code #{response.code}"}."
+        next
+      end
+      break if response.code == 200
+    end
+    abort "Unable to reach NCBI EUtils, error code #{response.code}." unless
+      response.code == 200
+    unless outfile.nil?
+      ohf = File.open(outfile, "w")
+      ohf.print response.to_s
+      ohf.close
+    end
+    response.to_s
+  end
+  def self.efetch(*etc)
+    eutils "efetch.fcgi", *etc
+  end
+  def self.elink(*etc)
+    eutils "elink.fcgi", *etc
+  end
+  def self.esummary(*etc)
+    eutils "esummary.fcgi", *etc
+  end
+  def self.update_gi(db, old_gi)
+    summ = JSON.parse RemoteData.esummary({:db=>db, :id=>old_gi,
+      :retmode=>"json"})
+    return nil,nil if summ["result"].nil? or summ["result"][old_gi.to_s].nil?
+    new_acc = summ["result"][old_gi.to_s]["replacedby"]
+    new_gi = (new_acc.nil? ? nil :
+      RemoteData.efetch({:db=>db, :id=>new_acc, :rettype=>"gi"}))
+    return new_gi,summ["result"][old_gi.to_s]["status"]
+  end
+  def self.ebiFetch(db, id, format, outfile=nil)
+    url = "#{@@EBIREST}/dbfetch/dbfetch/#{db}/#{id}/#{format}"
+    response = RestClient::Request.execute(:method=>:get,
+      :url=>url, :timeout=>600)
+    raise "Unable to reach EBI REST client, error code " +
+      response.code.to_s + "." unless response.code == 200
+    response.to_s
+  end
+  def self.ebiseq2taxid(id,db)
+    doc = RemoteData.ebiFetch(db, id, "annot").split(/[\n\r]/)
+    ln = doc.grep(/^FT\s+\/db_xref="taxon:/).first
+    ln = doc.grep(/^OX\s+NCBI_TaxID=/).first if ln.nil?
+    return nil if ln.nil?
+    ln.sub!(/.*(?:"taxon:|NCBI_TaxID=)(\d+)["; ].*/, "\\1")
+    return nil unless ln =~ /^\d+$/
+    ln
+  end
+end
+

data/utils/enveomics/Scripts/lib/enveomics_rb/seq_range.rb

@@ -0,0 +1,237 @@
+
+##### CLASSES:
+# SeqRange.parse(str): Initializes a new SeqRange from a string. A SeqRange is a
+#    representation of any collection of coordinates in a given sequence.
+#    Coordinates here are 1-based and base-located. Admitedly, the
+#    0-based/interbase-located system is much more convenient for range
+#    operations, but GenBank (together with most common Software) is built on
+#    the 1-based/base-located system.
+# str: A string describing the sequence range as in GenBank records.
+#    Note that "ID:location" notation is NOT supported by this implementation,
+#    althought it is permitted by GenBank. Some examples of valid `str`:
+#       "<1..123"
+#       "complement(3..6)"
+#       "join(complement(join(13..43,complement(45..46),complement(1..12),
+#           <1..12)),12..15,13..22)"
+#    The last one is valid, but once parsed it's internally simplified as:
+#       "join(complement(<1..12),1..12,45..46,complement(13..43),12..15,13..22)"
+#    Which is exactly equivalent. The common (but non-GenBank-compliant)
+#    practice of inverting coordinates instead of using the `complement()`
+#    operator is also supported. For example:
+#       "123..3"
+#    Is interpreted as:
+#       "complement(3..123)"
+# See also http://www.insdc.org/files/feature_table.html
+# 
+# SeqRange.new(c): Initializes a new SeqRange from an object.
+# c: Any object supported by the `<<` operator, or `nil` to create an empty
+#    SeqRange.
+# 
+# See also ContigSeqRange.parse.
+class SeqRange
+   # Class-level
+   def self.parse(str)
+      str.gsub!(/[^A-Za-z0-9\.\(\)<>,]/,"")
+      sr = nil
+      if str =~ /^join\((.+)\)$/i
+	 str1 = $1
+	 str2 = ""
+	 sr = SeqRange.new
+	 parens = 0
+	 str1.each_char do |chr|
+	    if chr=="," and parens==0
+	       sr += SeqRange.parse(str2)
+	       str2 = ""
+	       next
+	    elsif chr=="("
+	       parens += 1
+	    elsif chr==")"
+	       parens -= 1
+	       raise "Unbalanced parenthesis in '#{str1}'." if parens < 0
+	    end
+	    str2 += chr
+	 end
+	 sr += SeqRange.parse(str2) unless str2.empty?
+	 sr
+      elsif str =~ /^complement\((.+)\)$/i
+	 sr = SeqRange.parse($1)
+	 sr.reverse!
+	 sr
+      else
+	 sr = SeqRange.new(ContigSeqRange.parse(str))
+      end
+      sr
+   end
+   # Instance-level
+   attr_reader :contig
+   def initialize(c=nil)
+      @contig = []
+      self << c unless c.nil?
+   end
+   def leftmost; contig.map{ |c| c.left }.min; end
+   def rightmost; contig.map{ |c| c.right }.max; end
+   def size; contig.map{ |c| c.size }.inject(0,:+); end
+   def +(sr)
+      return(self + SeqRange.new(sr)) if sr.is_a? ContigSeqRange
+      raise "Unsupported operation '+' with class #{sr.class.to_s}." unless
+	 sr.is_a? SeqRange
+      out = SeqRange.new(self)
+      out << sr
+      out
+   end
+   def /(sr)
+      if sr.is_a? SeqRange
+	 sr2 = sr.sort.compact
+	 raise "Denominator is not a contiguous domain." unless sr2.size==1
+	 return(self/sr2.contig.first)
+      end
+      raise "Unsupported operation '/' with class #{sr.class.to_s}" unless
+	 sr.is_a? ContigSeqRange
+      raise "Denominator doesn't span the whole domain of numerator." unless
+	 sr.left <= leftmost and sr.right >= rightmost
+      i = ContigSeqRange.IGNORE_STRAND
+      ContigSeqRange.IGNORE_STRAND = false
+      range = self.sort.compact.size
+      ContigSeqRange.IGNORE_STRAND = i
+      range.to_f / sr.size
+   end
+   def <<(c)
+      if c.is_a? ContigSeqRange
+	 @contig << c
+      elsif c.is_a? SeqRange
+	 @contig += c.contig
+      elsif c.is_a? Array
+	 raise "Array must contain only objects of class ContigSeqRange." unless
+	    c.map{ |cc| cc.is_a? ContigSeqRange }.all?
+	 @contig += c
+      else
+	 raise "Unsupported operation '<<' with class #{c.class.to_s}."
+      end
+   end
+   def reverse ; SeqRange.new(self).reverse! ; end
+   def sort ; SeqRange.new(self).sort! ; end
+   def compact ; SeqRange.new(self).compact! ; end
+   def reverse!
+      @contig.each{ |c| c.reverse! }
+      @contig.reverse!
+      self
+   end
+   def sort!
+      @contig.sort!{ |x,y| x.left <=> y.left }
+      self
+   end
+   def compact!
+      return self if contig.size < 2
+      clean = false
+      while not clean
+	 clean = true
+	 (2 .. contig.size).each do |i|
+	    next unless contig[i-2].reverse? == contig[i-1].reverse?
+	    next unless contig[i-2].contig? contig[i-1]
+	    contig[i-2] += contig[i-1]
+	    contig[i-1] = nil
+	    clean = false
+	    break
+	 end
+	 @contig.compact!
+      end
+      self
+   end
+   def to_s
+      o = contig.map{ |c| c.to_s }.join(",")
+      o = "join(#{o})" if contig.size > 1
+      o
+   end
+end
+
+
+# ContigSeqRange.parse(str): Initializes a new ContigSeqRange from a string. A 
+#    ContigSeqRange is a primitive of `SeqRange` that doesn't support the
+#    `join()` operator. Other than that, syntax is identical to `SeqRange`.
+# str: A string describing the sequence range as in GenBank records (except
+#    `join()`).
+#
+# ContigSeqRange.new(a,b): Initializes a new ContigSeqRange from the
+#    coordinates as integers.
+# a: Start of the range.
+# b: End of the range. If a>b, the `complement()` operator is assumed.
+#
+# ContigSeqRange.IGNORE_STRAND = true: Use this pragma to ignore strandness.
+#    If set, it globally affects the behavior of of the class. Note that
+#    `SeqRange` instances contain a collection of `ContigSeqRange` objects, so
+#    that class is also affected.
+class ContigSeqRange
+   # Class-level
+   @@IGNORE_STRAND = false
+   def self.IGNORE_STRAND=(v); @@IGNORE_STRAND = !!v ; end
+   def self.IGNORE_STRAND; @@IGNORE_STRAND ; end
+   def self.parse(str)
+      str.downcase!
+      m = %r{^
+	 (?<c>complement\()?	# Reverse
+	 (?<lt><?)		# Open-ended to the left
+	 (?<left>\d+)		# Left coordinate
+	 (
+	    \.\.\.?		# 2 or 3 dots
+	    (?<gt1>>?)		# Open-ended to the right
+	    (?<right>\d+)	# Right coordinate
+	 )?
+	 (?<gt2>>?)		# Open-ended to the right
+	 \)?			# If reverse
+      $}x.match(str)
+      raise "Cannot parse range: #{str}." if m.nil?
+      c = ContigSeqRange.new(m[:left].to_i, m[:right].to_i)
+      c.open_left = true if m[:lt]=="<"
+      c.open_right = true if m[:gt1]==">" or m[:gt2]==">"
+      c.reverse! if m[:c]=="complement("
+      c
+   end
+   # Instance-level
+   attr_accessor :open_left, :open_right
+   attr_reader :coords
+   def initialize(a,b)
+      @coords = [[a,b].min, [a,b].max]
+      @open_left = false
+      @open_right = false
+      @reverse = (a > b)
+   end
+   def from; coords[ reverse ? 1 : 0 ] ; end
+   def to; coords[ reverse ? 0 : 1 ] ; end
+   def left; coords[0] ; end
+   def right; coords[1] ; end
+   def size; right-left+1 ; end
+   def reverse?; @reverse ; end
+   def reverse!
+      @reverse = ! reverse? unless @@IGNORE_STRAND
+      self
+   end
+   def overlap?(sr) !(right < sr.left or left > sr.right) ; end
+   def contig?(sr) !(right+1 < sr.left or left-1 > sr.right) ; end
+   def +(sr)
+      raise "Unsupported operation '+' with class #{sr.class.to_s}" unless
+	 sr.is_a? ContigSeqRange
+      raise "Non-contiguous ranges cannot be added." unless contig? sr
+      raise "Ranges in different strands cannot be added." unless
+	 reverse? == sr.reverse?
+      out = ContigSeqRange.new([left,sr.left].min, [right,sr.right].max)
+      out.reverse! if reverse?
+      out.open_left=true if (left < sr.left ? self : sr).open_left
+      out.open_right=true if (right > sr.right ? self : sr).open_right
+      out
+   end
+   def to_s
+      o = ""
+      o += "<" if open_left
+      o += left.to_s
+      if left == right
+	 o += ">" if open_right
+      else
+	 o += ".."
+	 o += ">" if open_right
+	 o += right.to_s
+      end
+      o = "complement(#{o})" if reverse?
+      o
+   end
+end
+

data/utils/enveomics/Scripts/lib/enveomics_rb/stat.rb

@@ -0,0 +1,30 @@
+
+module Enve
+  class Stat
+    # Generates a random number from the +dist+ distribution with +params+
+    # parameters. This is simply a wrapper to the r_* functions below.
+    def self.rand(dist=:unif, *params)
+      send("r_#{dist}", *params)
+    end
+
+    # Generates a random number from the uniform distribution between +min+ and
+    # +max+. By default generates random numbers between 0.0 and 1.0.
+    def self.r_unif(min=0.0, max=1.0)
+      min + (max-min)*Random::rand
+    end
+    
+    # Generates a random number from the geometric distribution with support
+    # {0, 1, 2, ...} and probability of success +p+.
+    def self.r_geom(p)
+      (Math::log(1.0 - rand)/Math::log(1.0-p) - 1.0).ceil
+    end
+
+    # Generates a random number from the shifted geometric distribution with
+    # support {1, 2, 3, ...} and probability of success +p+.
+    def self.r_sgeom(p)
+      (Math::log(1.0 - rand)/Math::log(1.0-p)).ceil
+    end
+    
+  end
+end
+

data/utils/enveomics/Scripts/lib/enveomics_rb/vcf.rb

@@ -0,0 +1,135 @@
+# @author  Luis M. Rodriguez-R
+# @license Artistic-2.0
+
+##
+# Enveomics representation of a Variant Call Format (VCF) file.
+class VCF
+  
+  ##
+  # File-handler, a File object.
+  attr_reader :fh
+  def initialize(file)
+    @fh = (file.is_a?(String) ? File.open(file, "r") : file )
+  end
+  
+  ##
+  # Iterate through each variant (i.e., each non-comment line), passing a
+  # VCF::Variant object to +blk+.
+  def each_variant(&blk)
+    fh.rewind
+    fh.each_line do |ln|
+      next if ln =~ /^#/
+      blk.call VCF::Variant.new(ln)
+    end
+  end
+
+  ##
+  # Iterate through each header (i.e., each comment line), passing a String to
+  # +blk+.
+  def each_header(&blk)
+    fh.rewind
+    fh.each_line do |ln|
+      next unless ln =~ /^#/
+      blk.call ln
+    end
+  end
+end
+
+class VCF::Variant
+  
+  ##
+  # Column definitions in VCF.
+  @@COLUMNS = [:chrom,:pos,:id,:ref,:alt,:qual,:filter,:info,:format,:bam]
+  
+  ##
+  # An Array of String, containing each of the VCF entrie's columns.
+  attr_reader :data
+
+  ##
+  # Initialize VCF::Variant from String +line+, a non-comment line in the VCF.
+  def initialize(line)
+    @data = line.chomp.split("\t")
+    # Qual as float
+    @data[5] = data[5].to_f
+    # Split info
+    info = data[7].split(";").map{ |i| i=~/=/ ? i.split("=", 2) : [i, true] }
+    @data[7] = Hash[*info.map{ |i| [i[0].to_sym, i[1]] }.flatten]
+    # Read formatted data
+    unless data[9].nil? or data[9].empty?
+      f = format.split(":")
+      b = bam.split(":")
+      f.each_index{ |i| @data[7][f[i].to_sym] = b[i] }
+    end
+    @data[7][:INDEL] = true if ref.size != alt.split(",").first.size
+  end
+
+  ##
+  # Named functions for each column.
+  @@COLUMNS.each_index do |i|
+    define_method(@@COLUMNS[i]) { @@COLUMNS[i]==:pos ? data[i].to_i : data[i] }
+  end
+
+  ##
+  # Sequencing depth.
+  def dp
+    return nil if info[:DP].nil?
+    info[:DP].to_i
+  end
+
+  ##
+  # Sequencing depth of FWD-REF, REV-REF, FWD-ALT, and REV-ALT.
+  def dp4
+    return nil if info[:DP4].nil?
+    @dp4 ||= info[:DP4].split(",").map{ |i| i.to_i }
+    @dp4
+  end
+
+  ##
+  ## Sequencing depth of REF and ALT.
+  def ad
+    return nil if info[:AD].nil?
+    @ad ||= info[:AD].split(",").map{ |i| i.to_i }
+    @ad
+  end
+  
+  ##
+  # Sequencing depth of the REF allele.
+  def ref_dp
+    return dp4[0] + dp4[1] unless dp4.nil?
+    return ad[0] unless ad.nil?
+    nil
+  end
+
+  ##
+  # Sequencing depth of the ALT allele.
+  def alt_dp
+    return dp4[2] + dp4[3] unless dp4.nil?
+    return ad[1] unless ad.nil?
+    nil
+  end
+
+  ##
+  # Information content of the variant in bits (from 0 to 1).
+  def shannon
+    return @shannon unless @shannon.nil?
+    a = ref_dp
+    b = alt_dp
+    ap = a.to_f/(a+b)
+    bp = b.to_f/(a+b)
+    @shannon = -(ap*Math.log(ap,2) + bp*Math.log(bp,2))
+    @shannon
+  end
+
+  ##
+  # Is it an indel?
+  def indel? ; !info[:INDEL].nil? and info[:INDEL] ; end
+
+  ##
+  # Return as String.
+  def to_s ; (data[0..6] + [info_to_s] + data[8..-1].to_a).join("\t") + "\n" ; end
+
+  ##
+  # Returns the INFO entry as String.
+  def info_to_s ; data[7].to_a.map{ |i| i.join("=") }.join(";") ; end
+  
+end