miga-base 0.2.6.5 → 0.3.0.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: a69a12511e98d2cc6751efa31592661b664e9d33
-  data.tar.gz: 807b0a13efce0367c88369a2f389fffff34ee58d
+  metadata.gz: 141b39aed7636b3f2389837e5fd13348a8e11252
+  data.tar.gz: b4bcec9bc8a8fd8c15c670b6f0fae0925d36138e
 SHA512:
-  metadata.gz: bdc1db4adec179da57c45ab29a41ec36ef2895f34d8159792420560bd0bc15910052fe6e2949ee0a992500e0001a7db4d2bbbf4f4d5e1daefb4c1c0fa3bd4ebb
-  data.tar.gz: 40d5a05ec2c154b50eb8952b87212106461080740e8986f982f44fa79cea3100d31beaff1c647afa83892b3e2cde37464ddd11568ce658addcb60992b6395de3
+  metadata.gz: c92ae5fb577d1e945d2a8be0f3d3dbe05aefd61094ee386f0bfb198237e83e487660419fa3cafb1cdc13aa25929ddd2022fab03530cac3ab4e927cb35481e82e
+  data.tar.gz: 6566cad346ff51d1e6b55c2e7bbaa3b5d766ae8292db0e829430237e2254dbb09a67d1ea63cdb9ea42365e47236b01b2af41e6e722aa153523bf39438d5e7ccb

data/actions/test_taxonomy.rb

@@ -0,0 +1,66 @@
+#!/usr/bin/env ruby
+
+# @package MiGA
+# @license Artistic-2.0
+
+require "miga/tax_dist"
+
+o = {q:true, test:"both"}
+OptionParser.new do |opt|
+  opt_banner(opt)
+  opt_object(opt, o, [:project, :dataset])
+  opt.on("-t", "--test STRING",
+    "Test to perform. Supported values: intax, novel, both."
+    ){ |v| o[:test]=v.downcase }
+  opt_common(opt, o)
+end.parse!
+
+##=> Main <=
+opt_require(o, project:"-P", dataset:"-D")
+
+$stderr.puts "Loading project." unless o[:q]
+p = MiGA::Project.load(o[:project])
+raise "Impossible to load project: #{o[:project]}" if p.nil?
+
+$stderr.puts "Loading dataset." unless o[:q]
+ds = p.dataset(o[:dataset])
+
+$stderr.puts "Finding closest relative." unless o[:q]
+cr = ds.closest_relatives(1)
+
+unless cr.empty?
+  $stderr.puts "Querying probability distributions." unless o[:q]
+  cr = cr[0]
+  puts "Closest relative: #{cr[0]} with AAI: #{cr[1]}."
+  tax = p.dataset(cr[0]).metadata[:tax]
+  tax ||= {}
+  
+  if %w[intax both].include? o[:test]
+    # Intax
+    r = MiGA::TaxDist.aai_pvalues(cr[1], :intax).map do |k,v|
+      sig = ""
+      [0.5,0.1,0.05,0.01].each{ |i| sig << "*" if v<i }
+      [MiGA::Taxonomy.LONG_RANKS[k], (tax[k] || "?"), v, sig]
+    end
+    puts ""
+    puts "Taxonomic classification"
+    puts MiGA::MiGA.tabulate(%w[Rank Taxonomy P-value Signif.], r)
+  end
+  
+  if %w[novel both].include? o[:test]
+    # Novel
+    r = MiGA::TaxDist.aai_pvalues(cr[1], :novel).map do |k,v|
+      sig = ""
+      [0.5,0.1,0.05,0.01].each{ |i| sig << "*" if v<i }
+      [MiGA::Taxonomy.LONG_RANKS[k], v, sig]
+    end
+    puts ""
+    puts "Taxonomic novelty"
+    puts MiGA::MiGA.tabulate(%w[Rank P-value Signif.], r)
+  end
+  
+  puts ""
+  puts "Significance at p-value below: *0.5, **0.1, ***0.05, ****0.01."
+end
+
+$stderr.puts "Done." unless o[:q]

data/bin/miga

@@ -34,6 +34,7 @@ $task_desc = {
   date: "Returns the current date in standard MiGA format.",
   # Taxonomy
   add_taxonomy: "Registers taxonomic information for datasets.",
+  test_taxonomy: "Returns test of taxonomic distributions for query datasets.",
   index_taxonomy: "Creates a taxonomy-indexed list of the datasets.",
   tax_distributions: "Estimates distributions of distance by taxonomy.",
 }

data/lib/miga/dataset.rb

@@ -4,6 +4,7 @@
 require "miga/metadata"
 require "miga/result"
 require "miga/dataset_result"
+require "sqlite3"
 
 ##
 # Dataset representation in MiGA.
@@ -135,28 +136,32 @@ class MiGA::Dataset < MiGA::MiGA
   # Get standard metadata values for the dataset as Array.
   def info
     MiGA::Dataset.INFO_FIELDS.map do |k|
-      (k=="name") ? self.name : self.metadata[k.to_sym]
+      (k=="name") ? self.name : metadata[k.to_sym]
     end
   end
   
   ##
   # Is this dataset a reference?
-  def is_ref? ; !!self.metadata[:ref] ; end
+  def is_ref? ; !!metadata[:ref] ; end
+
+  ##
+  # Is this dataset a query (non-reference)?
+  def is_query? ; !metadata[:ref] ; end
   
   ##
   # Is this dataset known to be multi-organism?
   def is_multi?
-    return false if self.metadata[:type].nil? or
-      @@KNOWN_TYPES[self.metadata[:type]].nil?
-    @@KNOWN_TYPES[self.metadata[:type]][:multi]
+    return false if metadata[:type].nil? or
+      @@KNOWN_TYPES[type].nil?
+    @@KNOWN_TYPES[type][:multi]
   end
   
   ##
   # Is this dataset known to be single-organism?
   def is_nonmulti?
-    return false if self.metadata[:type].nil? or
-      @@KNOWN_TYPES[self.metadata[:type]].nil?
-    !@@KNOWN_TYPES[self.metadata[:type]][:multi]
+    return false if metadata[:type].nil? or
+      @@KNOWN_TYPES[type].nil?
+    !@@KNOWN_TYPES[type][:multi]
   end
   
   ##
@@ -264,5 +269,20 @@ class MiGA::Dataset < MiGA::MiGA
     end
     adv
   end
+  
+  ##
+  # Returns an Array of duples (Arrays) sorted by AAI:
+  # - +0+: A String with the name(s) of the reference dataset.
+  # - +1+: A Float with the AAI.
+  # This function is currently only supported for query datasets. It returns
+  # +nil+ if this analysis is not supported.
+  def closest_relatives(how_many=1)
+    return nil if is_ref? or project.is_multi?
+    r = result :distances
+    return nil if r.nil?
+    db = SQLite3::Database.new(r.file_path :aai_db)
+    db.execute("SELECT seq2, aai FROM aai WHERE seq2 != ? " +
+      "GROUP BY seq2 ORDER BY aai DESC LIMIT ?", [name, how_many])
+  end
 
 end # class MiGA::Dataset

data/lib/miga/project.rb

@@ -162,6 +162,10 @@ class MiGA::Project < MiGA::MiGA
   ##
   # Is this a clade project?
   def is_clade? ; type==:clade ; end
+
+  ##
+  # Is this a project for multi-organism datasets?
+  def is_multi? ; @@KNOWN_TYPES[type][:multi] ; end
   
   ##
   # Returns Array of MiGA::Dataset.

data/lib/miga/project_result.rb

@@ -55,6 +55,8 @@ module MiGA::ProjectResult
       r.add_file(:ogs, "miga-project.ogs")
       r.add_file(:stats, "miga-project.stats")
       r.add_file(:rbm, "miga-project.rbm")
+      r.add_file(:core_pan, "miga-project.core-pan.tsv")
+      r.add_file(:core_pan_plot, "miga-project.core-pan.pdf")
       r
     end
 

data/lib/miga/version.rb

@@ -10,15 +10,15 @@ module MiGA
   # - Float representing the major.minor version.
   # - Integer representing gem releases of the current version.
   # - Integer representing minor changes that require new version number.
-  VERSION = [0.2, 6, 5]
+  VERSION = [0.3, 0, 0]
   
   ##
   # Nickname for the current major.minor version.
-  VERSION_NAME = "pochoir"
+  VERSION_NAME = "tinge"
   
   ##
   # Date of the current gem release.
-  VERSION_DATE = Date.new(2017, 5, 22)
+  VERSION_DATE = Date.new(2017, 5, 29)
   
   ##
   # Reference of MiGA.

data/scripts/ogs.bash

@@ -11,30 +11,34 @@ cd "$PROJECT/data/10.clades/03.ogs"
 # Initialize
 miga date > "miga-project.start"
 
-echo -n "" > miga-project.log
 DS=$(miga list_datasets -P "$PROJECT" --ref --no-multi)
-
-# Extract RBMs
-[[ -d miga-project.rbm ]] || mkdir miga-project.rbm
-for i in $DS ; do
-  for j in $DS ; do
-    file="miga-project.rbm/$i-$j.rbm"
-    [[ -s $file ]] && continue
-    echo "SELECT id1,id2,id,0,0,0,0,0,0,0,evalue,bitscore from rbm" \
-      "where seq1='$i' and seq2='$j' ;" \
-      | sqlite3 "../../09.distances/02.aai/$i.db" | tr "\\|" "\\t" \
-      > "$file"
-    [[ -s "$file" ]] || rm "$file"
+if [[ ! -s miga-project.ogs ]] ; then
+  # Extract RBMs
+  [[ -d miga-project.rbm ]] || mkdir miga-project.rbm
+  echo -n "" > miga-project.log
+  for i in $DS ; do
+    for j in $DS ; do
+      file="miga-project.rbm/$i-$j.rbm"
+      [[ -s $file ]] && continue
+      echo "SELECT id1,id2,id,0,0,0,0,0,0,0,evalue,bitscore from rbm" \
+        "where seq1='$i' and seq2='$j' ;" \
+        | sqlite3 "../../09.distances/02.aai/$i.db" | tr "\\|" "\\t" \
+        > "$file"
+      [[ -s "$file" ]] || rm "$file"
+    done
+    echo "$i" >> miga-project.log
   done
-  echo "$i" >> miga-project.log
-done
 
-# Estimate OGs
-ogs.mcl.rb -o miga-project.ogs -d miga-project.rbm -t "$CORES"
-ogs.stats.rb -o miga-project.ogs -j miga-project.stats
+  # Estimate OGs and Clean RBMs
+  ogs.mcl.rb -o miga-project.ogs -d miga-project.rbm -t "$CORES"
+  rm -rf miga-project.rbm
+fi
 
-# Clean RBMs
-rm -rf miga-project.rbm
+# Calculate Statistics
+ogs.stats.rb -o miga-project.ogs -j miga-project.stats
+ogs.core-pan.rb -o miga-project.ogs -s miga-project.core-pan.tsv -t "$CORES"
+Rscript $MIGA/utils/core-pan-plot.R \
+  miga-project.core-pan.tsv miga-project.core-pan.pdf
 
 # Finalize
 miga date > "miga-project.done"

data/utils/core-pan-plot.R

@@ -0,0 +1,29 @@
+#!/usr/bin/env Rscript
+
+argv <- commandArgs(trailingOnly=T)
+
+plot_core_pan <- function(core_pan, pdf){
+  a <- read.table(core_pan, sep="\t", header=TRUE)
+  pdf(pdf, 7, 5)
+  plot(1, type="n", xlim=c(0, max(a$genomes)*1.05), xaxs="i", yaxs="i",
+    ylim=c(0, max(a$pan_q3)*1.05), xlab="Genomes", ylab="Orthologous Groups")
+  # Core
+  polygon(c(a$genomes, rev(a$genomes)), c(a$core_q1, rev(a$core_q3)),
+    border=NA, col=rgb(0, 121, 166, 128/2, max=255))
+  lines(a$genomes, a$core_avg, col=rgb(0,121,166,max=255), lty=2)
+  lines(a$genomes, a$core_q2, col=rgb(0,121,166,max=255), lty=1)
+  # Pan
+  polygon(c(a$genomes, rev(a$genomes)), c(a$pan_q1, rev(a$pan_q3)),
+    border=NA, col=rgb(96, 11, 64, 128/2, max=255))
+  lines(a$genomes, a$pan_avg, col=rgb(96,11,64,max=255), lty=2)
+  lines(a$genomes, a$pan_q2, col=rgb(96,11,64,max=255), lty=1)
+  # Legend
+  legend("topleft",
+    legend=c("pangenome","core genome","Inter-Quartile","Median","Average"),
+    pch=c(16,16,15,NA,NA),lty=c(NA,NA,NA,1,2), pt.cex=c(1,1,2,NA,NA),
+    col=c(rgb(96,11,64,max=255), rgb(0,121,166,max=255),
+      rgb(0.5,0.5,0.5,166/255), rep(rgb(0.5,0.5,0.5),2)), bty="n")
+  dev.off()
+}
+
+plot_core_pan(argv[1], argv[2])

metadata

@@ -1,103 +1,103 @@
 --- !ruby/object:Gem::Specification
 name: miga-base
 version: !ruby/object:Gem::Version
-  version: 0.2.6.5
+  version: 0.3.0.0
 platform: ruby
 authors:
 - Luis M. Rodriguez-R
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2017-05-22 00:00:00.000000000 Z
+date: 2017-05-29 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rest-client
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ~>
       - !ruby/object:Gem::Version
         version: '1.7'
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ~>
       - !ruby/object:Gem::Version
         version: '1.7'
 - !ruby/object:Gem::Dependency
   name: daemons
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ~>
       - !ruby/object:Gem::Version
         version: '1.2'
-    - - ">="
+    - - '>='
       - !ruby/object:Gem::Version
         version: 1.2.4
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ~>
       - !ruby/object:Gem::Version
         version: '1.2'
-    - - ">="
+    - - '>='
       - !ruby/object:Gem::Version
         version: 1.2.4
 - !ruby/object:Gem::Dependency
   name: json
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ~>
       - !ruby/object:Gem::Version
         version: '1.8'
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ~>
       - !ruby/object:Gem::Version
         version: '1.8'
 - !ruby/object:Gem::Dependency
   name: sqlite3
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ~>
       - !ruby/object:Gem::Version
         version: '1.3'
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ~>
       - !ruby/object:Gem::Version
         version: '1.3'
 - !ruby/object:Gem::Dependency
   name: rake
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ~>
       - !ruby/object:Gem::Version
         version: '11'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ~>
       - !ruby/object:Gem::Version
         version: '11'
 - !ruby/object:Gem::Dependency
   name: test-unit
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ~>
       - !ruby/object:Gem::Version
         version: '3'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ~>
       - !ruby/object:Gem::Version
         version: '3'
 description: Microbial Genomes Atlas
@@ -108,32 +108,6 @@ extensions: []
 extra_rdoc_files:
 - README.md
 files:
-- Gemfile
-- LICENSE
-- README.md
-- Rakefile
-- actions/add_result.rb
-- actions/add_taxonomy.rb
-- actions/create_dataset.rb
-- actions/create_project.rb
-- actions/daemon.rb
-- actions/date.rb
-- actions/download_dataset.rb
-- actions/find_datasets.rb
-- actions/import_datasets.rb
-- actions/index_taxonomy.rb
-- actions/list_datasets.rb
-- actions/list_files.rb
-- actions/plugins.rb
-- actions/project_info.rb
-- actions/result_stats.rb
-- actions/run_local.rb
-- actions/tax_distributions.rb
-- actions/unlink_dataset.rb
-- bin/miga
-- lib/miga.rb
-- lib/miga/_data/aai-intax.tsv.gz
-- lib/miga/_data/aai-novel.tsv.gz
 - lib/miga/common.rb
 - lib/miga/daemon.rb
 - lib/miga/dataset.rb
@@ -147,6 +121,18 @@ files:
 - lib/miga/tax_index.rb
 - lib/miga/taxonomy.rb
 - lib/miga/version.rb
+- lib/miga.rb
+- test/common_test.rb
+- test/daemon_test.rb
+- test/dataset_test.rb
+- test/metadata_test.rb
+- test/project_test.rb
+- test/remote_dataset_test.rb
+- test/tax_index_test.rb
+- test/taxonomy_test.rb
+- test/test_helper.rb
+- lib/miga/_data/aai-intax.tsv.gz
+- lib/miga/_data/aai-novel.tsv.gz
 - scripts/_distances_functions.bash
 - scripts/_distances_noref_nomulti.bash
 - scripts/_distances_ref_nomulti.bash
@@ -170,16 +156,8 @@ files:
 - scripts/subclades.bash
 - scripts/trimmed_fasta.bash
 - scripts/trimmed_reads.bash
-- test/common_test.rb
-- test/daemon_test.rb
-- test/dataset_test.rb
-- test/metadata_test.rb
-- test/project_test.rb
-- test/remote_dataset_test.rb
-- test/tax_index_test.rb
-- test/taxonomy_test.rb
-- test/test_helper.rb
 - utils/adapters.fa
+- utils/core-pan-plot.R
 - utils/index_metadata.rb
 - utils/mytaxa_scan.R
 - utils/mytaxa_scan.rb
@@ -189,6 +167,30 @@ files:
 - utils/subclades-compile.rb
 - utils/subclades-nj.R
 - utils/subclades.R
+- bin/miga
+- actions/add_result.rb
+- actions/add_taxonomy.rb
+- actions/create_dataset.rb
+- actions/create_project.rb
+- actions/daemon.rb
+- actions/date.rb
+- actions/download_dataset.rb
+- actions/find_datasets.rb
+- actions/import_datasets.rb
+- actions/index_taxonomy.rb
+- actions/list_datasets.rb
+- actions/list_files.rb
+- actions/plugins.rb
+- actions/project_info.rb
+- actions/result_stats.rb
+- actions/run_local.rb
+- actions/tax_distributions.rb
+- actions/test_taxonomy.rb
+- actions/unlink_dataset.rb
+- Gemfile
+- Rakefile
+- README.md
+- LICENSE
 homepage: http://enve-omics.ce.gatech.edu/miga
 licenses:
 - Artistic-2.0
@@ -197,25 +199,25 @@ post_install_message:
 rdoc_options:
 - lib
 - README.md
-- "--main"
+- --main
 - README.md
-- "--title"
+- --title
 - MiGA
 require_paths:
 - lib
 required_ruby_version: !ruby/object:Gem::Requirement
   requirements:
-  - - ">="
+  - - '>='
     - !ruby/object:Gem::Version
       version: '1.9'
 required_rubygems_version: !ruby/object:Gem::Requirement
   requirements:
-  - - ">="
+  - - '>='
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
 rubyforge_project: 
-rubygems_version: 2.6.8
+rubygems_version: 2.0.14
 signing_key: 
 specification_version: 4
 summary: MiGA