miga-base 0.3.3.0 → 0.3.3.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
-SHA256:
-  metadata.gz: 518860e2b5fbd03ec6887055a40f1d3fe5761f2bbab7550ccf575fa6a2b7b5bf
-  data.tar.gz: 620c88f6eca40d4f054191a871b52d1ec35bd0caf7dc580a8e27f2fcec7fa19f
+SHA1:
+  metadata.gz: b1b32b7800278dc330c5c8e01f4b94dfd1d97750
+  data.tar.gz: 2c3b6ef0e73568df8775fb98c65d454cdcf0f411
 SHA512:
-  metadata.gz: 957cf133b881d048a1804b276a331fa9f459275290b8108685e9338173f80300d145c01dc0e0c4ec6f1504248c26651799ebc4fb00bbee94ffc0be01247b051e
-  data.tar.gz: 338cf79cd3dd467b0997b51d6a4677b63eda00f55e957dd85b723b358eb9f207836b1e2df0d8258e2fef1db44081fd08289471b6e872a7f82e94712c844d8ec4
+  metadata.gz: 590a41c7bc94f5d36a53e0b9eb4f096211ccdae8724e63948480e0b57c8b7fa24a5779534868c7cb13405b3360f35be00d977728886ba7b7491ef5aeebb0bc0d
+  data.tar.gz: 64a273f14eea3aec6f9c8cfb388bdae4bdf2027d7ce95d89b1e8a27799e51420d2f58b42a58871037ee4a0e7f616f92e8ac485a12dbe226d7ab5cda99792f286

data/README.md

@@ -30,6 +30,13 @@ You have two options:
   [installation instructions](manual/part2/installation.md). Once you have MiGA
   installed, you can [deploy some examples](manual/part4.md).
 
+# How to cite MiGA
+
+> Rodriguez-R *et al*. 2018. The Microbial Genomes Atlas (MiGA) webserver:
+> taxonomic and gene diversity analysis of Archaea and Bacteria at the whole
+> genome level. *Nucleic Acids Research* 46(W1):W282-W288.
+> [doi:10.1093/nar/gky467](https://doi.org/10.1093/nar/gky467).
+
 # Authors
 
 Developed and maintained by [Luis M. Rodriguez-R][lrr]. MiGA is the result of a

data/actions/about.rb

@@ -12,7 +12,7 @@ OptionParser.new do |opt|
   opt.on("-m", "--metadata STRING",
     "Print name and metadata field only."
     ){ |v| o[:datum]=v }
-  opt.on("--tab STRING",
+  opt.on("--tab",
     "Returns a tab-delimited table."){ |v| o[:tabular] = v }
   opt_common(opt, o)
 end.parse!

data/actions/ls.rb

@@ -15,7 +15,7 @@ OptionParser.new do |opt|
   opt.on("-m", "--metadata STRING",
     "Print name and metadata field only. If set, ignores -i."
     ){ |v| o[:datum]=v }
-  opt.on("--tab STRING",
+  opt.on("--tab",
     "Returns a tab-delimited table."){ |v| o[:tabular] = v }
   opt.on("-s", "--silent",
     "No output and exit with non-zero status if the dataset list is empty."

data/actions/summary.rb

@@ -6,8 +6,10 @@
 o = {q:true, units:false, tabular:false}
 opts = OptionParser.new do |opt|
   opt_banner(opt)
-  opt_object(opt, o, [:project, :dataset_opt, :result_dataset])
-  opt.on("--tab STRING",
+  opt_object(opt, o, [:project, :dataset_opt])
+  opt_filter_datasets(opt, o)
+  opt_object(opt, o, [:result_dataset])
+  opt.on("--tab",
     "Returns a tab-delimited table."){ |v| o[:tabular] = v }
   opt.on("--key STRING",
     "Returns only the value of the requested key."){ |v| o[:key] = v }

data/lib/miga/remote_dataset/base.rb

@@ -0,0 +1,59 @@
+
+require 'restclient'
+require 'open-uri'
+
+class MiGA::RemoteDataset < MiGA::MiGA
+
+  # Class-level
+  class << self
+    def UNIVERSE ; @@UNIVERSE ; end
+  end
+
+end
+
+module MiGA::RemoteDataset::Base
+
+  @@_EUTILS = "https://eutils.ncbi.nlm.nih.gov/entrez/eutils/"
+
+  ##
+  # Structure of the different database Universes or containers. The structure
+  # is a Hash with universe names as keys as Symbol and values being a Hash with
+  # supported keys as Symbol:
+  # - +:dbs+ => Hash with keys being the database name and the values a Hash of
+  #   properties such as +stage+, +format+, and +map_to+.
+  # - +url+ => Pattern of the URL where the data can be obtained, where +%1$s+
+  #   is the name of the database, +%2$s+ is the IDs, and +%3$s+ is format.
+  # - +method+ => Method used to query the URL. Only +:rest+ is currently
+  #   supported.
+  # - +map_to_universe+ => Universe where results map to. Currently unsupported.
+  @@UNIVERSE = {
+    web:{
+      dbs: {
+        assembly:{stage: :assembly, format: :fasta},
+        assembly_gz:{stage: :assembly, format: :fasta_gz}
+      },
+      url: "%2$s",
+      method: :net
+    },
+    ebi:{
+      dbs: { embl:{stage: :assembly, format: :fasta} },
+      url: "http://www.ebi.ac.uk/Tools/dbfetch/dbfetch/%1$s/%2$s/%3$s",
+      method: :rest
+    },
+    ncbi:{
+      dbs: { nuccore:{stage: :assembly, format: :fasta} },
+      url: "#{@@_EUTILS}efetch.fcgi?db=%1$s&id=%2$s&rettype=%3$s&retmode=text",
+      method: :rest
+    },
+    ncbi_map:{
+      dbs: { assembly:{map_to: :nuccore, format: :text} },
+        # FIXME ncbi_map is intended to do internal NCBI mapping between
+        # databases.
+      url: "#{@@_EUTILS}elink.fcgi?dbfrom=%1$s&id=%2$s&db=%3$s - - - - -",
+      method: :rest,
+      map_to_universe: :ncbi
+    }
+  }
+
+end
+

data/lib/miga/remote_dataset.rb

@@ -1,55 +1,15 @@
 # @package MiGA
 # @license Artistic-2.0
 
-require "restclient"
-require "open-uri"
+require 'miga/remote_dataset/base'
 
 ##
 # MiGA representation of datasets with data in remote locations.
 class MiGA::RemoteDataset < MiGA::MiGA
-  # Class-level
 
-  @@_EUTILS = "https://eutils.ncbi.nlm.nih.gov/entrez/eutils/"
-  ##
-  # Structure of the different database Universes or containers. The structure
-  # is a Hash with universe names as keys as Symbol and values being a Hash with
-  # supported keys as Symbol:
-  # - +:dbs+ => Hash with keys being the database name and the values a Hash of
-  #   properties such as +stage+, +format+, and +map_to+.
-  # - +url+ => Pattern of the URL where the data can be obtained, where +%1$s+
-  #   is the name of the database, +%2$s+ is the IDs, and +%3$s+ is format.
-  # - +method+ => Method used to query the URL. Only +:rest+ is currently
-  #   supported.
-  # - +map_to_universe+ => Universe where results map to. Currently unsupported.
-  def self.UNIVERSE ; @@UNIVERSE ; end
-  @@UNIVERSE = {
-    web:{
-      dbs: {
-        assembly:{stage: :assembly, format: :fasta},
-        assembly_gz:{stage: :assembly, format: :fasta_gz}
-      },
-      url: "%2$s",
-      method: :net
-    },
-    ebi:{
-      dbs: { embl:{stage: :assembly, format: :fasta} },
-      url: "http://www.ebi.ac.uk/Tools/dbfetch/dbfetch/%1$s/%2$s/%3$s",
-      method: :rest
-    },
-    ncbi:{
-      dbs: { nuccore:{stage: :assembly, format: :fasta} },
-      url: "#{@@_EUTILS}efetch.fcgi?db=%1$s&id=%2$s&rettype=%3$s&retmode=text",
-      method: :rest
-    },
-    ncbi_map:{
-      dbs: { assembly:{map_to: :nuccore, format: :text} },
-        # FIXME ncbi_map is intended to do internal NCBI mapping between
-        # databases.
-      url: "#{@@_EUTILS}elink.fcgi?dbfrom=%1$s&id=%2$s&db=%3$s - - - - -",
-      method: :rest,
-      map_to_universe: :ncbi
-    }
-  }
+  include MiGA::RemoteDataset::Base
+
+  # Class-level
 
   ##
   # Download data from the +universe+ in the database +db+ with IDs +ids+ and

data/lib/miga/version.rb

@@ -10,7 +10,7 @@ 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 = [0.3, 3, 0]
+  VERSION = [0.3, 3, 1]
   
   ##
   # Nickname for the current major.minor version.
@@ -25,7 +25,7 @@ module MiGA
   CITATION = "Rodriguez-R et al (2018). " +
     "The Microbial Genomes Atlas (MiGA) webserver: taxonomic and gene " +
     "diversity analysis of Archaea and Bacteria at the whole genome level. " +
-    "Nucleic Acids Research, gky467. DOI: 10.1093/nar/gky467."
+    "Nucleic Acids Research 46(W1):W282-W288. doi:10.1093/nar/gky467."
 
 end
 

data/scripts/clade_finding.bash

@@ -11,36 +11,8 @@ cd "$PROJECT/data/10.clades/01.find"
 # Initialize
 miga date > "miga-project.start"
 
-# Markov-cluster genomes by ANI
-gunzip -c ../../09.distances/03.ani/miga-project.txt.gz | tail -n+2 \
-  | awk -F"\\t" '$4>=90{print $2"'"\\t"'"$3"'"\\t"'"$4}' \
-  > genome-genome.aai90.rbm
-ogs.mcl.rb -d . -o miga-project.aai90-clades -t "$CORES" -i \
-  -f "(\\S+)-(\\S+)\\.aai90\\.rbm"
-rm genome-genome.aai90.rbm
-gunzip -c ../../09.distances/02.aai/miga-project.txt.gz | tail -n+2 \
-  | awk -F"\\t" '$4>=95{print $2"'"\\t"'"$3"'"\\t"'"$4}' \
-  > genome-genome.ani95.rbm
-ogs.mcl.rb -d . -o miga-project.ani95-clades -t "$CORES" -b \
-  -f "(\\S+)-(\\S+)\\.ani95\\.rbm"
-rm genome-genome.ani95.rbm
-
-# Propose clade projects
-tail -n +2 miga-project.ani95-clades | tr "," "\\t" | awk 'NF >= 5' \
-  > miga-project.proposed-clades
-
-# Run R code (except in projects type clade)
-if [[ $(miga project_info -P "$PROJECT" -m type) != "clade" ]] ; then
-  "$MIGA/utils/subclades.R" \
-    ../../09.distances/02.aai/miga-project.txt.gz \
-    miga-project "$CORES"
-  mv miga-project.nwk miga-project.aai.nwk
-  
-  # Compile
-  ruby "$MIGA/utils/subclades-compile.rb" . \
-    >  miga-project.class.tsv \
-    2> miga-project.class.nwk
-fi
+# Run
+ruby -I "$MIGA/lib" "$MIGA/utils/subclades.rb" "$PROJECT" "$SCRIPT"
 
 # Finalize
 miga date > "miga-project.done"

data/scripts/subclades.bash

@@ -12,15 +12,7 @@ cd "$PROJECT/data/10.clades/02.ani"
 miga date > "miga-project.start"
 
 # Run R code
-"$MIGA/utils/subclades.R" \
-  ../../09.distances/03.ani/miga-project.txt.gz \
-  miga-project "$CORES"
-mv miga-project.nwk miga-project.ani.nwk
-
-# Compile
-ruby "$MIGA/utils/subclades-compile.rb" . \
-  >  miga-project.class.tsv \
-  2> miga-project.class.nwk
+ruby -I "$MIGA/lib" "$MIGA/utils/subclades.rb" "$PROJECT" "$SCRIPT"
 
 # Finalize
 miga date > "miga-project.done"

data/utils/distance/runner.rb

@@ -30,8 +30,10 @@ class MiGA::DistanceRunner
     if opts[:run_taxonomy] && project.metadata[:ref_project]
       @home = File.expand_path('05.taxonomy', @home)
       @ref_project = MiGA::Project.load(project.metadata[:ref_project])
+      raise "Cannot load reference project: #{project.metadata[:ref_project]}" if @ref_project.nil?
+    else
+      @ref_project = project
     end
-    @ref_project ||= project
     [:haai_p, :aai_p, :ani_p, :distances_checkpoint].each do |m|
       @opts[m] ||= ref_project.metadata[m]
     end

data/utils/enveomics/Pipelines/assembly.pbs/FastA.N50.pl

@@ -1 +1 @@
-utils/enveomics/Pipelines/assembly.pbs/../../Scripts/FastA.N50.pl
+../../Scripts/FastA.N50.pl

data/utils/enveomics/Pipelines/assembly.pbs/FastA.filterN.pl

@@ -1 +1 @@
-utils/enveomics/Pipelines/assembly.pbs/../../Scripts/FastA.filterN.pl
+../../Scripts/FastA.filterN.pl

data/utils/enveomics/Pipelines/assembly.pbs/FastA.length.pl

@@ -1 +1 @@
-utils/enveomics/Pipelines/assembly.pbs/../../Scripts/FastA.length.pl
+../../Scripts/FastA.length.pl

data/utils/enveomics/Pipelines/blast.pbs/FastA.split.pl

@@ -1 +1 @@
-utils/enveomics/Pipelines/blast.pbs/../../Scripts/FastA.split.pl
+../../Scripts/FastA.split.pl

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

@@ -1 +1 @@
-utils/enveomics/Scripts/lib/../../enveomics.R
+../../enveomics.R

data/utils/subclade/base.rb

@@ -0,0 +1,6 @@
+
+require 'zlib'
+require 'miga'
+
+class MiGA::SubcladeRunner
+end

data/utils/subclade/pipeline.rb

@@ -0,0 +1,54 @@
+
+# High-end pipelines for SubcladeRunner
+module MiGA::SubcladeRunner::Pipeline
+
+  # Run species-level clusterings using ANI>95% / AAI>90%
+  def cluster_species
+    tasks = {ani95: [:ani_distances, 95.0], aai90: [:aai_distances, 90.0]}
+    tasks.each do |k, par|
+      # Final output
+      ogs_file = "miga-project.#{k}-clades"
+      next if File.size? ogs_file
+      
+      # Build ABC files
+      abc_path = tmp_file("#{k}.abc")
+      ofh = File.open(abc_path, 'w')
+      metric_res = project.result(par[0]) or raise "Incomplete step #{par[0]}"
+      Zlib::GzipReader.open(metric_res.file_path(:matrix)) do |ifh|
+        ifh.each_line do |ln|
+          next if ln =~ /^metric\t/
+          r = ln.chomp.split("\t")
+          ofh.puts "G>#{r[1]}\tG>#{r[2]}\t#{r[3]}" if r[3].to_f >= par[1]
+        end
+      end
+      ofh.close
+      # Cluster genomes
+      `ogs.mcl.rb -o '#{ogs_file}' --abc '#{abc_path}' -t '#{opts[:thr]}'`
+    end
+    # Propose clades
+    ofh = File.open('miga-project.proposed-clades', 'w')
+    File.open('miga-project.ani95-clades', 'r') do |ifh|
+      ifh.each_line do |ln|
+        next if $.==1
+        r = ln.chomp.split(',')
+        ofh.puts r.join("\t") if r.size >= 5
+      end
+    end
+    ofh.close
+  end
+
+  def subclades metric
+    src = File.expand_path('utils/subclades.R', MiGA::MiGA.root_path)
+    step = :"#{metric}_distances"
+    metric_res = project.result(step) or raise "Incomplete step #{step}"
+    matrix = metric_res.file_path(:matrix)
+    `Rscript '#{src}' '#{matrix}' miga-project '#{opts[:thr]}'`
+    File.rename('miga-project.nwk',"miga-project.#{metric}.nwk") if
+          File.exist? 'miga-project.nwk'
+  end
+
+  def compile
+    src = File.expand_path('utils/subclades-compile.rb', MiGA::MiGA.root_path)
+    `ruby '#{src}' '.' 'miga-project.class'`
+  end
+end

data/utils/subclade/runner.rb

@@ -0,0 +1,51 @@
+
+require_relative 'base.rb'
+require_relative 'temporal.rb'
+require_relative 'pipeline.rb'
+
+
+class MiGA::SubcladeRunner
+
+  include MiGA::SubcladeRunner::Temporal
+  include MiGA::SubcladeRunner::Pipeline
+
+  attr_reader :project, :step, :opts, :home, :tmp
+
+  def initialize(project_path, step, opts_hash={})
+    @opts = opts_hash
+    @project = MiGA::Project.load(project_path) or
+          raise "No project at #{project_path}"
+    @step = step.to_sym
+    clades_dir = File.expand_path('data/10.clades', project.path)
+    @home = File.expand_path(@step==:clade_finding ? '01.find' : '02.ani',
+          clades_dir)
+    @opts[:thr] ||= ENV.fetch("CORES"){ 2 }.to_i
+  end
+
+  # Launch the appropriate analysis
+  def go!
+    return if project.type == :metagenomes
+    Dir.chdir home
+    Dir.mktmpdir do |tmp_dir|
+      @tmp = tmp_dir
+      create_temporals
+      step==:clade_finding ? go_clade_finding! : go_subclades!
+    end
+  end
+
+  # Launch analysis for clade_finding
+  def go_clade_finding!
+    cluster_species
+    unless project.is_clade?
+      subclades :aai
+      compile
+    end
+  end
+
+  # Launch analysis for subclades
+  def go_subclades!
+    subclades :ani
+    compile
+  end
+
+end

data/utils/subclade/temporal.rb

@@ -0,0 +1,14 @@
+
+require 'tmpdir'
+
+module MiGA::SubcladeRunner::Temporal
+
+  # Create the empty temporal structure
+  def create_temporals
+  end
+
+  # Path to the +file+ in the temporal directory
+  def tmp_file(file)
+    File.expand_path(file, tmp)
+  end
+end

data/utils/subclades-compile.rb

@@ -1,16 +1,15 @@
 #!/usr/bin/env ruby
-#
-# @author  Luis M. Rodriguez-R
-# @update  Jan-15-2016
-# @license artistic license 2.0
-#
 
 $:.push File.expand_path(File.dirname(__FILE__) + "/lib")
-dir = ARGV.shift or abort "Usage: #{$0} <classif.dir>"
+dir = ARGV.shift
+out = ARGV.shift or abort "Usage: #{$0} <classif.dir> <out.base>"
 
 def read_classif(dir, classif={})
    classif_file = File.expand_path("miga-project.classif", dir)
    return classif unless File.exist? classif_file
+   ready = File.expand_path('miga-project.ready', dir)
+   File.size?(ready) or raise "Incomplete recursion found at #{dir}"
+   File.unlink ready
    fh = File.open(classif_file, "r")
    klass = []
    while ln = fh.gets
@@ -44,7 +43,7 @@ end
 
 c = read_classif(dir)
 max_depth = c.values.map{|i| i.count}.max
-c.each do |k,v|
-   puts ([k] + v + ["0"]*(max_depth-v.count)).join("\t")
+File.open("#{out}.tsv", 'w') do |fh|
+  c.each { |k,v| fh.puts ([k] + v + ["0"]*(max_depth-v.count)).join("\t") }
 end
-$stderr.puts print_tree(c) + ";"
+File.open("#{out}.nwk", 'w') { |fh| fh.puts print_tree(c) + ";" }

data/utils/subclades.R

@@ -13,24 +13,82 @@ suppressPackageStartupMessages(library(parallel))
 suppressPackageStartupMessages(library(enveomics.R))
 
 #= Main function
-subclades <- function(ani_file, out_base, thr=1, ani=c()) {
+subclades <- function(ani_file, out_base, thr=1, ani.d=dist(0)) {
   say("==> Out base:", out_base, "<==")
-  
-  # Input arguments
-  if(missing(ani_file)){
-    a <- as.data.frame(ani)
+
+  # Normalize input matrix
+  dist_rdata = paste(out_base, "dist.rdata", sep=".")
+  if(!missing(ani_file)){
+    if(length(ani.d)==0 && !file.exists(dist_rdata)){
+      # Read from ani_file
+      a <- read.table(gzfile(ani_file), sep="\t", header=TRUE, as.is=TRUE)
+      if(nrow(a)==0){
+        generate_empty_files(out_base)
+        return(NULL)
+      }
+      say("Distances")
+      a$d <- 1 - (a$value/100)
+      ani.d <- enve.df2dist(a, 'a', 'b', 'd', default.d=max(a$d)*1.2)
+      save(ani.d, file=dist_rdata)
+    }
+  }
+
+  # Read result if the subclade is ready, run it otherwise
+  if(file.exists(paste(out_base,"classif",sep="."))){
+    say("Loading")
+    ani.medoids <- read.table(paste(out_base, "medoids", sep="."),
+          sep=' ', as.is=TRUE)[,1]
+    a <- read.table(paste(out_base,"classif",sep="."), sep="\t", as.is=TRUE)
+    ani.types <- a[,2]
+    names(ani.types) <- a[,1]
+    if(length(ani.d)==0) load(dist_rdata)
   }else{
-    a <- read.table(gzfile(ani_file), sep="\t", header=TRUE, as.is=TRUE)
+    res <- subclade_clustering(out_base, thr, ani.d, dist_rdata)
+    if(length(res)==0) return(NULL)
+    ani.medoids <- res[['ani.medoids']]
+    ani.types <- res[['ani.types']]
+    ani.d <- res[['ani.d']]
   }
-  if(nrow(a)==0){
-    generate_empty_files(out_base)
-    return(NULL)
+
+  # Recursive search
+  say("Recursive search")
+  for(i in 1:length(ani.medoids)){
+    medoid <- ani.medoids[i]
+    ds_f <- names(ani.types)[ ani.types==i ]
+    say("Analyzing subclade", i, "with medoid:", medoid)
+    dir_f <- paste(out_base, ".sc-", i, sep="")
+    if(!dir.exists(dir_f)) dir.create(dir_f)
+    write.table(ds_f,
+      paste(out_base, ".sc-", i, "/miga-project.all",sep=""),
+      quote=FALSE, col.names=FALSE, row.names=FALSE)
+    if(length(ds_f) > 8L){
+      ani_subset <- as.dist(as.matrix(ani.d)[ds_f, ds_f])
+      subclades(out_base=paste(out_base, ".sc-", i, "/miga-project", sep=""),
+        thr=thr, ani.d=ani_subset)
+    }
   }
   
+  # Declare recursion up-to-here complete
+  write.table(date(), paste(out_base, 'ready', sep='.'),
+        quote=FALSE, row.names=FALSE, col.names=FALSE)
+}
+
+#= Heavy-lifter
+subclade_clustering <- function(out_base, thr, ani.d, dist_rdata) {
   # Get ANI distances
-  say("Distances")
-  a$d <- 1-a$value/100
-  ani.d <- enve.df2dist(data.frame(a$a, a$b, a$d), default.d=max(a$d)*1.2)
+  if(length(ani.d) > 0){
+    # Just use ani.d (and save in dist_rdata_
+    save(ani.d, file=dist_rdata)
+  }else if(file.exists(dist_rdata)){
+    # Read from dist_rdata
+    load(dist_rdata)
+  }else{
+    stop("Cannot find input matrix", out_base)
+  }
+  if(length(labels(ani.d)) <= 8L) return(list())
+  
+  # Build tree
+  say("Tree")
   ani.ph <- bionj(ani.d)
   express.ori <- options('expressions')$expressions
   if(express.ori < ani.ph$Nnode*4){
@@ -75,7 +133,6 @@ subclades <- function(ani_file, out_base, thr=1, ani=c()) {
   say("Text report")
   write.table(ani.medoids, paste(out_base, "medoids", sep="."),
     quote=FALSE, col.names=FALSE, row.names=FALSE)
-  save(ani.d, file=paste(out_base, "dist.rdata", sep="."))
   classif <- cbind(names(ani.types), ani.types, ani.medoids[ ani.types ], NA)
   ani.d.m <- 100 - as.matrix(ani.d)*100
   for(j in 1:nrow(classif)){
@@ -83,27 +140,18 @@ subclades <- function(ani_file, out_base, thr=1, ani=c()) {
   }
   write.table(classif, paste(out_base,"classif",sep="."),
     quote=FALSE, col.names=FALSE, row.names=FALSE, sep="\t")
-
-  # Recursive search
-  say("Recursive search")
-  for(i in 1:top.n){
-    medoid <- ani.medoids[i]
-    ds_f <- names(ani.types)[ ani.types==i ]
-    say("Analyzing subclade", i, "with medoid:", medoid)
-    dir.create(paste(out_base, ".sc-", i, sep=""))
-    write.table(ds_f,
-      paste(out_base, ".sc-", i, "/miga-project.all",sep=""),
-      quote=FALSE, col.names=FALSE, row.names=FALSE)
-    if(length(ds_f) > 5){
-      a_f <- a[ (a$a %in% ds_f) & (a$b %in% ds_f), ]
-      subclades(out_base=paste(out_base, ".sc-", i, "/miga-project", sep=""),
-        thr=thr, ani=a_f)
-    }
-  }
+  
+  # Return data
+  say("Cluster ready")
+  return(list(
+    ani.medoids=ani.medoids,
+    ani.types=ani.types,
+    ani.d=ani.d
+  ))
 }
 
 #= Helper functions
-say <- function(...) { cat("[", date(), "]", ..., "\n") }
+say <- function(...) { message(paste("[",date(),"]",...,"\n"),appendLF=FALSE) }
 
 generate_empty_files <- function(out_base) {
   pdf(paste(out_base, ".pdf", sep=""), 7, 12)
@@ -182,6 +230,7 @@ ggplotColours <- function(n=6, h=c(0, 360)+15, alpha=1){
 }
 
 #= Main
+options(warn=1)
 subclades(ani_file=argv[1], out_base=argv[2],
   thr=ifelse(is.na(argv[3]), 1, as.numeric(argv[3])))
 

data/utils/subclades.rb

@@ -0,0 +1,9 @@
+#!/usr/bin/env ruby
+
+require_relative 'subclade/runner.rb'
+
+project = ARGV.shift
+step = ARGV.shift
+opts = Hash[ ARGV.map{ |i| i.split("=",2).tap{ |j| j[0] = j[0].to_sym } } ]
+runner = MiGA::SubcladeRunner.new(project, step, opts)
+runner.go!

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: miga-base
 version: !ruby/object:Gem::Version
-  version: 0.3.3.0
+  version: 0.3.3.1
 platform: ruby
 authors:
 - Luis M. Rodriguez-R
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2018-08-17 00:00:00.000000000 Z
+date: 2018-08-29 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rest-client
@@ -161,6 +161,7 @@ files:
 - lib/miga/project/plugins.rb
 - lib/miga/project/result.rb
 - lib/miga/remote_dataset.rb
+- lib/miga/remote_dataset/base.rb
 - lib/miga/result.rb
 - lib/miga/result/base.rb
 - lib/miga/result/dates.rb
@@ -472,9 +473,14 @@ files:
 - utils/plot-taxdist.R
 - utils/ref-tree.R
 - utils/requirements.txt
+- utils/subclade/base.rb
+- utils/subclade/pipeline.rb
+- utils/subclade/runner.rb
+- utils/subclade/temporal.rb
 - utils/subclades-compile.rb
 - utils/subclades-nj.R
 - utils/subclades.R
+- utils/subclades.rb
 homepage: http://enve-omics.ce.gatech.edu/miga
 licenses:
 - Artistic-2.0
@@ -501,7 +507,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
       version: '0'
 requirements: []
 rubyforge_project: 
-rubygems_version: 2.7.6
+rubygems_version: 2.5.2.3
 signing_key: 
 specification_version: 4
 summary: MiGA