miga-base 0.3.1.7 → 0.3.2.0

data/utils/enveomics/Makefile

@@ -7,7 +7,7 @@ include globals.mk
 
 TEST=Tests
 enveomics_r=enveomics.R
-enveomics_r_v=enveomics.R_1.1.6
+enveomics_r_v=enveomics.R_1.3
 .PHONY: test install install-scripts install-r uninstall install-deps
 
 test: $(enveomics_r_v).tar.gz

data/utils/enveomics/Manifest/Tasks/blasttab.json

@@ -623,6 +623,7 @@
         { "r_package": "optparse" },
         { "r_package": "enveomics.R" }
       ],
+      "see_also": [ "RecPlot2.compareIdentities.R" ],
       "options": [
         {
           "opt": "--prefix",
@@ -644,6 +645,12 @@
           "default": 300,
           "description": ["Breaks in the identity histogram."]
         },
+        {
+          "opt": "--id-free-range",
+          "description": ["Indicates that the range should be freely set from",
+            "the observed values. Otherwise, 70-100% is included in the",
+            "identity histogram (default)."]
+        },
         {
           "opt": "--id-metric",
           "arg": "select",
@@ -684,6 +691,13 @@
           "description": ["Color of peaks, mandatory for peak-finding (e.g.,",
             "darkred)."]
         },
+        {
+          "opt": "--peaks-method",
+          "arg": "select",
+          "values": ["emauto","em","mower"],
+          "default": "emauto",
+          "description": "Method to detect peaks."
+        },
         {
           "name": "R Object Output",
           "arg": "out_file",
@@ -707,6 +721,58 @@
           "description": "Height of the plot in inches (7 by default)."
         }
       ]
+    },
+    {
+      "task": "RecPlot2.compareIdentities.R",
+      "description": ["Calculates the difference between identity",
+        "distributions of two recruitment plots."],
+      "help_arg": "--help",
+      "requires": [
+        { "r_package": "optparse" },
+        { "r_package": "enveomics.R" }
+      ],
+      "see_also": [ "BlastTab.recplot2.R" ],
+      "options": [
+        {
+          "opt": "--method",
+          "arg": "string",
+          "default": "hellinger",
+          "description": ["Distance method to use. This should be (an",
+            "unambiguous abbreviation of) one of:",
+            "'hellinger' (Hellinger, 1090, doi:10.1515/crll.1909.136.210),",
+            "'bhattacharyya' (Bhattacharyya, 1943, Bull. Calcutta Math. Soc.",
+            "35), 'kl' or 'kullback–leibler' (Kullback & Leibler, 1951,",
+            "doi:10.1214/aoms/1177729694), or 'euclidean'."]
+        },
+        {
+          "opt": "--pseudocounts",
+          "arg": "float",
+          "default": 0.0,
+          "description": ["Smoothing parameter for Laplace smoothing. Use 0",
+            "for no smoothing, or 1 for add-one smoothing."]
+        },
+        {
+          "opt": "--max-deviation",
+          "arg": "float",
+          "description": ["Maximum mean deviation between identity breaks",
+            "tolerated (as percent identity). Difference in number of",
+            "id.breaks is never tolerated."]
+        },
+        {
+          "arg": "in_file",
+          "name": "RecPlot A",
+          "mandatory": true,
+          "description": ["File containing the R object `rp` for the first",
+            "recruitment plot."]
+        },
+        {
+          "arg": "in_file",
+          "name": "RecPlot B",
+          "mandatory": true,
+          "description": ["File containing the R object `rp` for the second",
+            "recruitment plot."]
+        }
+      ]
     }
   ]
 }

data/utils/enveomics/Manifest/Tasks/fasta.json

@@ -158,12 +158,19 @@
       "warn": ["Please note that this script will check for the consistency of",
         "the names (assuming a pair of related reads contains the same name",
         "varying only in a trailing slash (/) followed by a digit. If you want",
-        "to turn this feature off just set the $eval_T variable to zero. If",
+        "to turn this feature off just set the checking period to zero. If",
         "you want to decrease the sampling period (to speed the script up) or",
-        "increase it (to make it more sensitive to errors) just change $eval_T",
-        "accordingly."],
+        "increase it (to make it more sensitive to errors) just change the ",
+        "checking period accordingly."],
       "help_arg": "",
       "options": [
+        {
+          "name": "Checking period",
+          "opt": "-T",
+          "arg": "integer",
+          "default": 1000,
+          "description": "Sampling period for names evaluation."
+        },
         {
           "arg": "out_file",
           "mandatory": true,

data/utils/enveomics/Manifest/Tasks/fastq.json

@@ -45,10 +45,10 @@
       "warn": ["Note that this script will check for the consistency of the",
         "names (assuming a pair of related reads contains the same name",
         "varying only in a trailing slash (/) followed by a digit. If you want",
-        "to turn this feature off just set the -T option to zero. If you want",
-        "to decrease the sampling period (to speed the script up) or increase",
-        "it (to make it more sensitive to errors) just change -T option",
-        "accordingly."],
+        "to turn this feature off just set the checking period to zero. If you",
+        "want to decrease the sampling period (to speed the script up) or",
+        "increase it (to make it more sensitive to errors) just change the",
+        "checking period accordingly."],
       "see_also": ["FastQ.split.pl","FastA.interpose.pl"],
       "help_arg": "",
       "options": [

data/utils/enveomics/Manifest/Tasks/mapping.json

@@ -8,7 +8,8 @@
         "missing from the file. If you produce your BedGraph file with",
         "bedtools genomecov and want to consider zero-coverage position, be",
         "sure to use -bga (not -bg)."],
-      "see_also": ["BlastTab.seqdepth.pl", "BlastTab.seqdepth_ZIP.pl"],
+      "see_also": ["BedGraph.window.rb",
+        "BlastTab.seqdepth.pl", "BlastTab.seqdepth_ZIP.pl"],
       "help_arg": "--help",
       "options": [
         {
@@ -23,6 +24,42 @@
           "default": 0.5,
           "description": ["Central range to consider, between 0 and 1. By",
             "default: inter-quartile range (0.5)."]
+        },
+        {
+          "opt": "--per-seq",
+          "description": ["Calculate averages per reference sequence, not",
+            "total. Assumes a sorted BedGraph file."]
+        },
+        {
+          "opt": "--length",
+          "description": "Add sequence length to the output."
+        }
+      ]
+    },
+    {
+      "task": "BedGraph.window.rb",
+      "description": ["Estimates the sequencing depth per windows from a",
+        "BedGraph file."],
+      "warn": ["This script doesn't consider zero-coverage positions if",
+        "missing from the file. If you produce your BedGraph file with",
+        "bedtools genomecov and want to consider zero-coverage position, be",
+        "sure to use -bga (not -bg)."],
+      "see_also": ["BedGraph.tad.rb",
+        "BlastTab.seqdepth.pl", "BlastTab.seqdepth_ZIP.pl"],
+      "help_arg": "--help",
+      "options": [
+        {
+          "opt": "--input",
+          "arg": "in_file",
+          "mandatory": true,
+          "description": "Input BedGraph file."
+        },
+        {
+          "name": "Window size",
+          "opt": "--win",
+          "arg": "float",
+          "default": 1000,
+          "description": "Window size, in base pairs."
         }
       ]
     }

data/utils/enveomics/Manifest/categories.json

@@ -3,6 +3,7 @@
     "Sequence similarity search": {
       "Statistics": [
         "BedGraph.tad.rb",
+        "BedGraph.window.rb",
         "BlastPairwise.AAsubs.pl",
         "BlastTab.advance.bash",
         "BlastTab.recplot2.R",
@@ -10,7 +11,8 @@
         "BlastTab.seqdepth_nomedian.pl",
         "BlastTab.seqdepth_ZIP.pl",
         "BlastTab.sumPerHit.pl",
-        "FastQ.test-error.rb"
+        "FastQ.test-error.rb",
+        "RecPlot2.compareIdentities.R"
       ],
       "Manipulation": [
         "BlastTab.addlen.rb",
@@ -134,6 +136,14 @@
       "Clustering": [
         "ogs.mcl.rb",
         "clust.rand.rb"
+      ],
+      "Read recruitments": [
+        "BedGraph.tad.rb",
+        "BedGraph.window.rb",
+        "BlastTab.catsbj.pl",
+        "BlastTab.pairedHits.rb",
+        "BlastTab.recplot2.R",
+        "RecPlot2.compareIdentities.R"
       ]
     }
   }

data/utils/enveomics/Manifest/examples.json

@@ -56,8 +56,8 @@
       "task": "BlastTab.recplot2.R",
       "description": ["Generates recruitment plots for a comparison",
         "between a virome containing HIV and the HIV-1 genome."],
-      "values": ["hiv_mix-hiv1.blast.tsv",50,100,null,null,null,null,null,"NA",
-        "hiv_mix-hiv1.Rdata","hiv_mix-hiv1.pdf",null,null]
+      "values": ["hiv_mix-hiv1.blast.tsv",50,100,null,null,null,null,null,null,
+        null,null,"hiv_mix-hiv1.Rdata","hiv_mix-hiv1.pdf",null,null]
     },
     {
       "_": "== Examples of functional annotations ==",

data/utils/enveomics/README.md

@@ -29,6 +29,8 @@ http://www.perlfoundation.org/artistic_license_2_0.
 Most scripts in this repository are self-documented.  However,
 more extensive documentation (and some discussion) can be found at the
 [documentation website](http://enve-omics.ce.gatech.edu/enveomics/docs).
+Additional documentation for recruitment plots can be found
+[here](Docs/recplot2.md).
 
 ## Citation
 

data/utils/enveomics/Scripts/Aln.cat.rb

@@ -129,6 +129,7 @@ begin
       puts ">#{key}", a[key].join("").gsub(/(.{1,60})/, "\\1\n")
       a.delete(key)
    end
+   $stderr.puts "  #{lengths.inject(:+)} columns." unless o[:q]
    unless o[:coords].nil?
       $stderr.puts "Generating coordinates." unless o[:q]
       coords = File.open(o[:coords],"w")

data/utils/enveomics/Scripts/BedGraph.tad.rb

@@ -1,9 +1,9 @@
 #!/usr/bin/env ruby
 
-require "optparse"
+require 'optparse'
 
-o = {range:0.5}
-ARGV << "-h" if ARGV.empty?
+o = {range: 0.5, perseq: false, length: false}
+ARGV << '-h' if ARGV.empty?
 OptionParser.new do |opt|
   opt.banner = "
   Estimates the truncated average sequencing depth (TAD) from a BedGraph file.
@@ -13,20 +13,26 @@ OptionParser.new do |opt|
   want to consider zero-coverage position, be sure to use -bga (not -bg).
 
   Usage: #{$0} [options]"
-  opt.separator ""
-  opt.on("-i", "--input PATH",
-    "Input BedGraph file (mandatory)."){ |v| o[:i]=v }
-  opt.on("-r", "--range FLOAT",
-    "Central range to consider, between 0 and 1.",
+  opt.separator ''
+  opt.on('-i', '--input PATH',
+    'Input BedGraph file (mandatory).'){ |v| o[:i]=v }
+  opt.on('-r', '--range FLOAT',
+    'Central range to consider, between 0 and 1.',
     "By default: #{o[:range]} (inter-quartile range)."
     ){ |v| o[:range]=v.to_f }
-  opt.on("-h", "--help", "Display this screen.") do
+  opt.on('-s', '--per-seq',
+    'Calculate averages per reference sequence, not total.',
+    'Assumes a sorted BedGraph file.'
+    ){ |v| o[:perseq] = v }
+  opt.on('-l', '--length',
+    'Add sequence length to the output.'){ |v| o[:length] = v }
+  opt.on('-h', '--help', 'Display this screen.') do
     puts opt
     exit
   end
-  opt.separator ""
+  opt.separator ''
 end.parse!
-abort "-i is mandatory." if o[:i].nil?
+abort '-i is mandatory.' if o[:i].nil?
 
 def pad(d, idx, r)
   idx.each do |i|
@@ -39,33 +45,49 @@ def pad(d, idx, r)
   d
 end
 
+def report(sq, d, ln, o)
+  # Estimate padding ranges
+  pad = (1.0-o[:range])/2.0
+  r = (pad*ln).round
+
+  # Pad
+  d = pad(d, d.each_index.to_a, r+0)
+  d = pad(d, d.each_index.to_a.reverse, r+0)
+
+  # Average
+  y = [0.0]
+  unless d.compact.empty?
+    s = d.each_with_index.to_a.map{ |v,i| v.nil? ? 0 : i*v }.inject(0,:+)
+    y[0] = s.to_f/d.compact.inject(:+)
+  end
+
+  # Report
+  y.unshift(sq) if o[:perseq]
+  y << ln if o[:length]
+  puts y.join("\t")
+end
+
 # Read BedGraph
-d = []
+d  = []
 ln = 0
+pre_sq = nil
 File.open(o[:i], "r") do |ifh|
   ifh.each_line do |i|
     next if i =~ /^#/
-    r = i.chomp.split("\t")[1 .. -1].map{ |j| j.to_i }
+    r  = i.chomp.split("\t")
+    sq = r.shift
+    if o[:perseq] and !pre_sq.nil? and pre_sq!=sq
+      report(pre_sq, d, ln, o)
+      d  = []
+      ln = 0
+    end
+    r.map! { |j| j.to_i }
     l = r[1]-r[0]
     d[ r[2] ] ||= 0
-    d[ r[2] ] += l
+    d[ r[2] ]  += l
     ln += l
+    pre_sq = sq
   end
 end
-
-# Estimate padding ranges
-pad = (1.0-o[:range])/2.0
-r = (pad*ln).round
-
-# Pad
-d = pad(d, d.each_index.to_a, r+0)
-d = pad(d, d.each_index.to_a.reverse, r+0)
-
-# Average
-if d.compact.empty?
-  p 0.0
-else
-  s = d.each_with_index.to_a.map{ |v,i| v.nil? ? 0 : i*v }.inject(0,:+)
-  p s.to_f/d.compact.inject(:+)
-end
+report(pre_sq, d, ln, o)
 

data/utils/enveomics/Scripts/BedGraph.window.rb

@@ -0,0 +1,71 @@
+#!/usr/bin/env ruby
+
+require 'optparse'
+
+o = {win: 1000}
+ARGV << '-h' if ARGV.empty?
+OptionParser.new do |opt|
+  opt.banner = "
+  Estimates the sequencing depth per windows from a BedGraph file.
+
+  IMPORTANT: This script doesn't consider zero-coverage positions if missing
+  from the file. If you produce your BedGraph file with bedtools genomecov and
+  want to consider zero-coverage position, be sure to use -bga (not -bg).
+
+  Usage: #{$0} [options]"
+  opt.separator ''
+  opt.on('-i', '--input PATH',
+    'Input BedGraph file (mandatory).'){ |v| o[:i]=v }
+  opt.on('-w', '--win INT',
+    'Window size, in base pairs.', "By default: #{o[:win]}."
+    ){ |v| o[:win]=v.to_i }
+  opt.on('-h', '--help', 'Display this screen.') do
+    puts opt
+    exit
+  end
+  opt.separator ''
+end.parse!
+abort '-i is mandatory.' if o[:i].nil?
+
+def report(d, a, b, seqs)
+  # Average
+  y = 0.0
+  unless d.compact.empty?
+    s = d.each_with_index.to_a.map{ |v,i| v.nil? ? 0 : i*v }.inject(0,:+)
+    y = s.to_f/d.compact.inject(:+)
+  end
+
+  # Report
+  puts [a, b, y, seqs.keys.join(",")].join("\t")
+end
+
+# Read BedGraph
+d  = []
+ln = 0
+a = 1
+seqs = {}
+b = o[:win]
+File.open(o[:i], "r") do |ifh|
+  ifh.each_line do |i|
+    next if i =~ /^#/
+    r  = i.chomp.split("\t")
+    sq = r.shift
+    seqs[sq] = 1
+    r.map!{ |j| j.to_i }
+    l = r[1]-r[0]
+    d[ r[2] ] ||= 0
+    d[ r[2] ]  += l
+    ln += l
+    while ln >= b
+      d[ r[2] ] -= (ln-b)
+      report(d, a, b, seqs)
+      seqs = {}
+      seqs[ sq ] = 1 if ln > b
+      d = []
+      d[ r[2] ] = (ln-b)
+      a = b + 1
+      b = a + o[:win] - 1
+    end
+  end
+end
+

data/utils/enveomics/Scripts/BlastTab.recplot2.R

@@ -19,11 +19,12 @@ opt <- enve.cliopts(enve.recplot2,
   o_desc=list(pos.breaks="Breaks in the positions histogram.",
     id.breaks="Breaks in the identity histogram.",
     id.summary="Function summarizing the identity bins. By default: sum.",
-    peaks.col="Color of peaks, mandatory for peak-finding (e.g., darkred)."),
+    peaks.col="Color of peaks, mandatory for peak-finding (e.g., darkred).",
+    peaks.method="Method to detect peaks; one of emauto, em, or mower."),
   p_desc=paste("","Produce recruitment plot objects provided that",
     "BlastTab.catsbj.pl has been previously executed.", sep="\n\t"),
   ignore=c("plot"),
-  defaults=c(id.metric="identity", peaks.col=NA))
+  defaults=c(id.metric="identity", peaks.col=NA, peaks.method="emauto"))
 
 #= Run it!
 if(length(opt$args)>1){
@@ -35,6 +36,10 @@ if(length(opt$args)>1){
 }
 pc <- opt$options[["peaks.col"]]
 if(!is.na(pc) && pc=="NA") opt$options[["peaks.col"]] <- NA
+if(!is.null(opt$options[["peaks.method"]])){
+  opt$options[["peaks.opts"]] <- list(method=opt$options[["peaks.method"]])
+  opt$options[["peaks.method"]] <- NULL
+}
 rp <- do.call("enve.recplot2", opt$options)
 save(rp, file=opt$args[1])
 if(length(opt$args)>1) dev.off()