miga-base 0.7.26.0 → 1.0.0.1

data/utils/enveomics/Scripts/anir.rb

@@ -0,0 +1,137 @@
+#!/usr/bin/env ruby
+
+# frozen_string_literal: true
+
+$:.push File.expand_path('../lib', __FILE__)
+require 'enveomics_rb/enveomics'
+require 'enveomics_rb/anir'
+$VERSION = 1.0
+
+o = {
+  q: false, threads: 2,
+  r_format: :fastq, g_format: :fasta, m_format: :sam, r_type: :single,
+  identity: 95.0, algorithm: :auto, bimodality: 0.5, bin_size: 1.0,
+  coefficient: :sarle
+}
+
+OptionParser.new do |opt|
+  cmd = File.basename($0)
+  opt.banner = <<~BANNER
+
+    [Enveomics Collection: #{cmd} v#{$VERSION}]
+
+    Estimates ANIr: the Average Nucleotide Identity of reads against a genome
+
+    Usage
+        # [ Input/output modes ]
+        # Run mapping and (optionally) save it as SAM
+        # Requires bowtie2
+        #{cmd} -r reads.fastq -g genome.fasta -m out_map.sam [options]
+
+        # Read mapping from BAM file
+        # Requires samtools
+        #{cmd} -m map.bam --m-format bam [options]
+
+        # Read mapping from other formats: SAM or Tabular BLAST
+        #{cmd} -m map.blast --m-format tab [options]
+
+        # Read a list of identities as percentage (contig filtering off)
+        #{cmd} -m identities.txt --m-format list [options]
+
+        # [ Identity threshold modes ]
+        #{cmd} -i 95 -a fix [options] # Set fixed identity threshold
+        #{cmd} -a gmm [options]       # Find valley by EM of GMM
+        #{cmd} -a auto [options]      # Pick method by bimodality (default)"
+
+  BANNER
+
+  opt.separator 'Input/Output'
+  opt.on('-r', '--reads PATH', 'Metagenomic reads') { |v| o[:r] = v }
+  opt.on('-g', '--genome PATH', 'Genome assembly') { |v| o[:g] = v }
+  opt.on('-m', '--mapping PATH', 'Mapping file') { |v| o[:m] = v }
+  opt.on('-L', '--list PATH', 'Output file with identities') { |v| o[:L] = v }
+  opt.on('-H', '--hist PATH', 'Output file with histogram') { |v| o[:H] = v }
+  opt.on(
+    '-T', '--tab PATH', 'Output file with results in tabular format'
+  ) { |v| o[:T] = v }
+  opt.separator ''
+
+  opt.separator 'Formats'
+  opt.on(
+    '--r-format STRING',
+    'Metagenomic reads format: fastq (default) or fasta',
+    'Both options support compression with .gz file extension'
+  ) { |v| o[:r_format] = v.downcase.to_sym }
+  opt.on(
+    '--r-type STRING', 'Type of metagenomic reads:',
+    '~ single (default): Single reads',
+    '~ coupled: Coupled reads in separate files (-m must be comma-delimited)',
+    '~ interleaved: Coupled reads in a single interposed file'
+  ) { |v| o[:r_type] = v.downcase.to_sym }
+  opt.on(
+    '--g-format STRING',
+    'Genome assembly format: fasta (default) or list',
+    'Both options support compression with .gz file extension',
+    'If passed in mapping-read mode, filters only matches to these contigs'
+  ) { |v| o[:g_format] = v.downcase.to_sym }
+  opt.on(
+    '--m-format STRING',
+    'Mapping file format: sam (default), bam, tab, or list',
+    'sam, tab, and list options support compression with .gz file extension'
+  ) { |v| o[:m_format] = v.downcase.to_sym }
+  opt.separator ''
+
+  opt.separator 'Identity threshold'
+  opt.on(
+    '-i', '--identity FLOAT', Float,
+    "Set a fixed threshold of percent identity (default: #{o[:identity]})"
+  ) { |v| o[:identity] = v }
+  opt.on(
+    '-a', '--algorithm STRING',
+    'Set an algorithm to automatically detect identity threshold:',
+    '~ gmm: Valley detection by E-M of Gaussian Mixture Model',
+    '~ fix: Fixed threshold, see -i',
+    '~ auto (default): Pick gmm or fix depending on bimodality, see -b'
+  ) { |v| o[:algorithm] = v.downcase.to_sym }
+  opt.on(
+    '-b', '--bimodality FLOAT', Float,
+    'Threshold of bimodality below which the algorithm is set to fix',
+    'The coefficient used is the de Michele & Accantino (2014) B index',
+    "By default: #{o[:bimodality]}"
+  ) { |v| o[:bimodality] = v }
+  opt.on(
+    '--coefficient STRING',
+    'Coefficient of bimodality for -a auto:',
+    '~ sarle (default): Sarle\'s bimodality coefficient b',
+    '~ dma: de Michele and Accatino (2014 PLoS ONE) B index, use with -b 0.1'
+  ) { |v| o[:coefficient] = v.downcase.to_sym }
+  opt.on(
+    '--bin-size FLOAT', Float,
+    "Width of histogram bins (in percent identity). By default: #{o[:bin_size]}"
+  ) { |v| o[:bin_size] = v }
+  opt.separator ''
+
+  opt.separator 'General'
+  opt.on(
+    '-t', '--threads INT', Integer, 'Threads to use'
+  ) { |v| o[:threads] = v }
+  opt.on('-l', '--log PATH', 'Log file to save output') { |v| o[:log] = v }
+  opt.on('-q', '--quiet', 'Run quietly') { |v| o[:q] = v }
+  opt.on('-h', '--help', 'Display this screen') do
+    puts opt
+    exit
+  end
+  opt.separator ''
+end.parse!
+
+anir = Enveomics::ANIr.new(o)
+anir.go!
+if o[:T]
+  File.open(o[:T], 'w') do |fh|
+    fh.puts "anir\tsd\treads\tid_threshold"
+    fh.puts [
+      anir.sample.mean, anir.sample.sd, anir.sample.n, anir.opts[:identity]
+    ].join("\t")
+  end
+end
+

data/utils/enveomics/Scripts/clust.rand.rb

@@ -0,0 +1,102 @@
+#!/usr/bin/env ruby
+
+#
+# @author: Luis M. Rodriguez-R
+# @license: artistic license 2.0
+#
+
+require "optparse"
+
+o = { q:false, prec:6 }
+ARGV << "-h" if ARGV.empty?
+OptionParser.new do |opts|
+   opts.banner = "
+Calculates the Rand Index and the Adjusted Rand Index between two clusterings.
+
+The clustering format is a raw text file with one cluster per line, each
+defined as comma-delimited members, and a header line (ignored). Note that this
+is equivalent to the OGs format for 1 genome.
+
+Usage: #{$0} [options]"
+  opts.separator ""
+  opts.separator "Mandatory"
+  opts.on("-1", "--clust1 FILE", "First input file."){ |v| o[:clust1]=v }
+  opts.on("-2", "--clust2 FILE", "Second input file."){ |v| o[:clust2]=v }
+  opts.separator ""
+  opts.separator "Other options"
+  opts.on("-p", "--prec INT",
+    "Precision to report. By default: #{o[:prec]}"){ |v| o[:prec]=v.to_i }
+  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 ""
+end.parse!
+abort "-1 is mandatory" if o[:clust1].nil?
+abort "-2 is mandatory" if o[:clust2].nil?
+
+def load_clust(file, q)
+  $stderr.puts "Reading clusters in '#{file}'." unless q
+  out = []
+  File.open(file, "r") do |fh|
+    fh.each_line do |ln|
+      next if $.==1
+      out[$.-2] = ln.chomp.split(",")
+    end
+  end
+  $stderr.puts " Loaded clusters: #{out.size}." unless q
+  out
+end
+
+def choose_2(n)
+  return 0 if n<2
+  n*(n-1)/2
+end
+
+##### MAIN:
+begin
+  # Read the pre-computed OGs
+  clust1 = load_clust(o[:clust1], o[:q])
+  clust2 = load_clust(o[:clust2], o[:q])
+  
+  # Contingency table
+  $stderr.puts "Estimating the contingency table." unless o[:q]
+  cont = []
+  b_sums = []
+  clust1.each_with_index do |x_i, i|
+    cont[i] = []
+    clust2.each_with_index do |y_j, j|
+      cont[i][j] = (x_i & y_j).size
+      b_sums[j]||= 0
+      b_sums[j] += cont[i][j]
+    end
+  end
+  a_sums = cont.map{ |i| i.inject(:+) }
+
+  # Calculate variables
+  # - see http://i11www.iti.kit.edu/extra/publications/ww-cco-06.pdf
+  $stderr.puts "Estimating indexes." unless o[:q]
+  n = clust1.map{ |i| i.size }.inject(:+)
+  pairs = choose_2(n)
+  n11 = clust1.each_index.map do |i|
+    clust2.each_index.map do |j|
+      choose_2(cont[i][j])
+    end.inject(:+)
+  end.inject(:+).to_f
+  t1 = a_sums.map{ |a_i| choose_2(a_i) }.inject(:+).to_f
+  t2 = b_sums.map{ |b_j| choose_2(b_j) }.inject(:+).to_f
+  t3 = 2*t1*t2/(n*(n-1))
+  n00 = pairs + n11 - t1 - t2
+  r_index = (n11 + n00)/pairs
+  r_adjusted = (n11 - t3)/((t1+t2)/2 - t3)
+  
+  # Report
+  puts "Rand Index = %.#{o[:prec]}f" % r_index
+  puts "Adjusted Rand Index = %.#{o[:prec]}f" % r_adjusted
+rescue => err
+  $stderr.puts "Exception: #{err}\n\n"
+  err.backtrace.each { |l| $stderr.puts l + "\n" }
+  err
+end
+

data/utils/enveomics/Scripts/gi2tax.rb

@@ -0,0 +1,103 @@
+#!/usr/bin/env ruby
+
+#
+# @author  Luis M. Rodriguez-R <lmrodriguezr at gmail dot com>
+# @license artistic license 2.0
+#
+
+$:.push File.expand_path(File.dirname(__FILE__) + "/lib")
+require "enveomics_rb/remote_data"
+use "nokogiri"
+
+#================================[ Options parsing ]
+o = {
+  :q=>false, :gis=>[], :dbfrom=>"nuccore", :header=>true,
+  :exact_gi=>false, :no_nil=>false, :ret=>"ScientificName",
+  :ranks=>%w(superkingdom phylum class order family genus species)}
+OptionParser.new do |opt|
+  opt.banner = "
+  Maps a list of NCBI GIs to their corresponding taxonomy using the NCBI
+  EUtilities. Avoid using this script on millions of entries at a time, since
+  each entry elicits two requests to NCBI's servers.
+
+  *IMPORTANT NOTE*: NCBI is phasing out support for GIs. Please use acc.ver
+  instead with NCBIacc2tax.rb.
+
+  Usage: #{$0} [options]".gsub(/^ +/,"")
+  opt.separator ""
+  opt.on("-g", "--gis GI1,GI2,...", Array,
+    "Comma-separated list of GIs. Required unless -i is passed."
+    ){ |v| o[:gis]=v }
+  opt.on("-i", "--infile FILE",
+    "Raw text file containing the list of GIs, one per line.",
+    "Required unless -g is passed."){ |v| o[:infile]=v }
+  opt.on("-p", "--protein",
+    "Use if the GIs are proteins. Otherwise, GIs are assumed to be from " +
+    "the Nuccore Database."){ o[:dbfrom]="protein" }
+  opt.on("-r", "--ranks RANK1,RANK2,...", Array,
+    "Taxonomic ranks to report. By default: #{o[:ranks].join(",")}."
+    ){ |v| o[:ranks]=v }
+  opt.on("-n", "--noheader",
+    "Do not include a header in the output."){ o[:header]=false }
+  opt.on("-t", "--taxids",
+    "Return Taxonomy IDs instead of scientific names."){ o[:ret]="TaxId" }
+  opt.on("--exact-gi",
+    "Returns only taxonomy associated with the exact GI passed.",
+    "By default, it attempts to update accession versions if possible."
+    ){ |v| o[:exact_gi]=v }
+  opt.on("--ignore-missing",
+    "Does not report missing GIs in the output file.",
+    "By default, it reports GI and empty values for all other columns."
+    ){ |v| o[:no_nil]=v }
+  opt.on("-q", "--quiet", "Run quietly."){ |v| o[:q]=true }
+  opt.on("-h", "--help","Display this screen") do
+    puts opt
+    exit
+  end
+  opt.separator ""
+end.parse!
+
+#================================[ Functions ]
+def gi2taxid(db, gi)
+  doc = Nokogiri::XML( RemoteData.elink({:dbfrom=>db,
+    :db=>"taxonomy", :id=>gi}) )
+  doc.at_xpath("/eLinkResult/LinkSet/LinkSetDb/Link/Id")
+end
+#================================[ Main ]
+begin
+  o[:gis] += File.readlines(o[:infile]).map{ |l| l.chomp } unless
+    o[:infile].nil?
+  o[:ranks].map!{ |r| r.downcase }
+  puts (["GI", "TaxId"] + o[:ranks].map{ |r| r.capitalize }).join("\t") if
+    o[:header]
+  o[:gis].each do |gi|
+    taxid = gi2taxid(o[:dbfrom], gi)
+    status = ""
+    if taxid.nil? and not o[:exact_gi]
+      new_gi, status = RemoteData.update_gi(o[:dbfrom], gi)
+      taxid = gi2taxid(o[:dbfrom], new_gi) unless new_gi.nil?
+    end
+    if taxid.nil?
+      warn "Cannot find link to taxonomy: #{gi} #{status}"
+      puts ([gi, ""] + o[:ranks].map{ |i| "" }).join("\t") unless o[:no_nil]
+      next
+    end
+    taxonomy = {}
+    unless taxid.nil?
+      doc = Nokogiri::XML( RemoteData.efetch({:db=>"taxonomy",
+        :id=>taxid.content}) )
+      taxonomy[ doc.at_xpath("/TaxaSet/Taxon/Rank").content ] =
+        doc.at_xpath("/TaxaSet/Taxon/#{o[:ret]}").content
+      doc.xpath("/TaxaSet/Taxon/LineageEx/Taxon").each do |taxon|
+        taxonomy[ taxon.at_xpath("./Rank").content ] =
+          taxon.at_xpath("./#{o[:ret]}").content
+      end
+    end
+    puts ([gi, taxid.content] +
+      o[:ranks].map{ |rank| taxonomy[ rank ] ||= "" }).join("\t")
+  end
+rescue => err
+  $stderr.puts "Exception: #{err}\n\n"
+  err.backtrace.each { |l| $stderr.puts l + "\n" }
+  err
+end

data/utils/enveomics/Scripts/in_silico_GA_GI.pl

@@ -0,0 +1,96 @@
+# usage perl in_silico_GA.pl [options]
+
+use Getopt::Long;
+use Math::Random qw(:all);
+
+$argu=GetOptions('in=s'=>\$infile, # input fasta chr file
+                 'out=s'=>\$outfile, # output file name
+                 'coverage=s'=>\$cov, # desired output
+                 'seq_error=s'=>\$seq_error, # sequencing error
+                 'read_len=s'=>\$read_len, # simulated read length
+                 'ins_len=s'=>\$ins_len,  # insertion length
+                 'ins_var=s'=>\$ins_var);
+
+$chr='';
+open(IN,$infile);
+open(OUT,">$outfile");
+%code=();
+$code{'0'}='C';
+$code{'1'}='A';
+$code{'2'}='T';
+$code{'3'}='G';
+
+while(<IN>){
+  chomp;
+  if(!/^\>/){
+	$chr.=$_;
+  }
+  else{
+    $gi=$_;
+    if($gi= ~/^\>gi\|(\S+)\|\S+\|\S+/){
+      $gi=$1;}
+
+}
+}
+close(IN);
+  
+$chr_size=length $chr;
+print "chromosome size: $chr_size\n";
+$seg_size=2*$read_len+$ins_len;
+$reads_number=int($cov*$chr_size/($read_len*2));
+print "generated reads $reads_number x 2\n";
+
+for(1..$reads_number){
+  $index=$_;
+  $l=length $index;
+  $k=8-$l;
+  $kk='0' x $k;
+  $id= 'read'.$kk.$index.'_'.$gi;
+
+#make start site;
+  $start_site=int(rand($chr_size));
+#make short seg length;
+  $seg_length=int(random_normal(1,$seg_size,$ins_var));
+
+#extract the segment
+  $seg=substr($chr,$start_site,$seg_length);
+  $s_len=length $seg;
+  $gap=$seg_length-$s_len;
+  if($gap!=0){
+	$makeup=substr($chr,0,$gap);
+	$seg.=$makeup;
+  }
+
+  $id.='.start'.$start_site.'.seg_len'.$seg_length;
+
+#get the reads
+  $seq1=substr($seg,0,$read_len);
+  #$seg=~tr/ATCG/TAGC/ this line can change the orientation of the second read;
+  $seq2=substr($seg,-$read_len);
+# sequencing error introducing
+  @seq1=split(//,$seq1);
+  @seq2=split(//,$seq2);
+  @mut1=random_binomial($read_len,1,$seq_error);
+  @mut2=random_binomial($read_len,1,$seq_error);
+
+  for(0..$#mut1){
+	$i=$_;
+	if($mut1[$i]==1){
+	  $r=int(rand(4));
+	  $seq1[$i]=$code{$r};
+	}
+	if($mut2[$i]==1){
+	  $r=int(rand(4));
+	  $seq2[$i]=$code{$r};
+	}
+  }
+  $seq1=join('',@seq1);
+  $seq2=join('',@seq2);
+
+  $id1=$id.'#0/1';
+  $id2=$id.'#0/2';
+  
+  print OUT ">$id1\n$seq1\n>$id2\n$seq2\n";
+}
+	
+

data/utils/enveomics/Scripts/lib/enveomics.R

@@ -0,0 +1 @@
+../../enveomics.R