miga-base 0.7.9.0 → 0.7.12.1

data/lib/miga/cli/action/classify_wf.rb

@@ -42,7 +42,9 @@ class MiGA::Cli::Action::ClassifyWf < MiGA::Cli::Action
         '--no-summaries',
         'Do not generate intermediate step summaries'
       ) { |v| cli[:summaries] = v }
-      opts_for_wf(opt, 'Input genome assemblies (nucleotides, FastA)')
+      opts_for_wf(
+        opt, 'Input genome assemblies (nucleotides, FastA)', qual: false
+      )
     end
   end
 

data/lib/miga/cli/action/derep_wf.rb

@@ -19,6 +19,10 @@ class MiGA::Cli::Action::DerepWf < MiGA::Cli::Action
         'Use Average Amino Acid Identity (AAI) as genome similarity metric',
         'By default: Use Average Nucleotide Identity (ANI)'
       ) { cli[:metric] = :aai }
+      opt.on(
+        '--ani',
+        'Use Average Nucleotide Identity (ANI) as similarity metric (default)'
+      ) { cli[:metric] = :ani }
       opt.on(
         '--threshold FLOAT', Float,
         "Metric threshold (%) to dereplicate. By default: #{cli[:threshold]}"

data/lib/miga/cli/action/edit.rb

@@ -17,18 +17,21 @@ class MiGA::Cli::Action::Edit < MiGA::Cli::Action
         'Activate dataset; requires -D'
       ) { |v| cli[:activate] = v }
       opt.on(
-        '--inactivate',
-        'Inactivate dataset; requires -D'
-      ) { |v| cli[:activate] = !v }
+        '--inactivate [reason]',
+        'Inactivate dataset; requires -D',
+        'The argument is optional: reason to inactivate dataset'
+      ) { |v| cli[:activate] = false ; cli[:reason] = v }
     end
   end
 
   def perform
     obj = cli.load_project_or_dataset
     unless cli[:activate].nil?
-      cli.ensure_par({ dataset: '-D' },
-                     '%<name>s is mandatory with --[in-]activate: please provide %<flag>s')
-      cli[:activate] ? obj.activate! : obj.inactivate!
+      cli.ensure_par(
+        { dataset: '-D' },
+        '%<name>s is mandatory with --[in-]activate: please provide %<flag>s'
+      )
+      cli[:activate] ? obj.activate! : obj.inactivate!(cli[:reason])
     end
     cli.add_metadata(obj)
     obj.save

data/lib/miga/cli/action/quality_wf.rb

@@ -15,7 +15,10 @@ class MiGA::Cli::Action::QualityWf < MiGA::Cli::Action
         '-m', '--mytaxa-scan',
         'Perform MyTaxa scan analysis'
       ) { |v| cli[:mytaxa] = v }
-      opts_for_wf(opt, 'Input genome assemblies (nucleotides, FastA)')
+      opts_for_wf(
+        opt, 'Input genome assemblies (nucleotides, FastA)',
+        qual: false
+      )
     end
   end
 

data/lib/miga/cli/action/stats.rb

@@ -38,7 +38,7 @@ class MiGA::Cli::Action::Stats < MiGA::Cli::Action
     end
     if cli[:key].nil?
       r[:stats].each do |k, v|
-        k_n = k == :g_c_content ? 'G+C content' : k.to_s.unmiga_name.capitalize
+        k_n = k.to_s.unmiga_name.sub(/^./, &:upcase)
         cli.puts "#{k_n}: #{v.is_a?(Array) ? v.join(' ') : v}"
       end
     else

data/lib/miga/cli/action/wf.rb

@@ -15,7 +15,7 @@ module MiGA::Cli::Action::Wf
 
   def opts_for_wf(opt, files_desc, params = {})
     {
-      multi: false, cleanup: true, project_type: false, ncbi: true
+      multi: false, cleanup: true, project_type: false, ncbi: true, qual: true
     }.each { |k, v| params[k] = v if params[k].nil? }
     opt.on(
       '-o', '--out_dir PATH',
@@ -40,6 +40,13 @@ module MiGA::Cli::Action::Wf
         'Only download complete genomes, not drafts'
       ) { |v| cli[:ncbi_draft] = v }
     end
+    if params[:qual]
+      opt.on(
+        '--min-qual FLOAT', Float,
+        'Minimum genome quality to include in analysis',
+        'By default: 50.0'
+      ) { |v| cli[:min_qual] = v }
+    end
     if params[:cleanup]
       opt.on(
         '-c', '--clean',
@@ -125,7 +132,7 @@ module MiGA::Cli::Action::Wf
              ]) unless MiGA::Project.exist? cli[:outdir]
     # Define project metadata
     p = cli.load_project(:outdir, '-o')
-    [:haai_p, :aai_p, :ani_p, :ess_coll].each { |i| p_metadata[i] = cli[i] }
+    %i[haai_p aai_p ani_p ess_coll min_qual].each { |i| p_metadata[i] = cli[i] }
     p_metadata[:type] = cli[:project_type]
     transfer_metadata(p, p_metadata)
     # Download datasets
@@ -159,9 +166,10 @@ module MiGA::Cli::Action::Wf
                  '-P', cli[:outdir],
                  '-r', r,
                  '-o', File.expand_path("#{r}.tsv", cli[:outdir]),
-                 '--tab'
+                 '--tab', '--ref', '--active'
                ])
     end
+    call_cli(['browse', '-P', cli[:outdir]])
   end
 
   def cleanup

data/lib/miga/cli/base.rb

@@ -11,39 +11,40 @@ module MiGA::Cli::Base
     preproc_wf: 'Preprocess input genomes or metagenomes',
     index_wf: 'Generate distance indexing of input genomes',
     # Projects
-    new: 'Creates an empty MiGA project',
-    about: 'Displays information about a MiGA project',
-    doctor: 'Performs consistency checks on a MiGA project',
-    get_db: 'Downloads a pre-indexed database',
+    new: 'Create an empty MiGA project',
+    about: 'Display information about a MiGA project',
+    doctor: 'Perform consistency checks on a MiGA project',
+    get_db: 'Download a pre-indexed database',
+    browse: 'Explore a project locally using a web browser',
     # Datasets
-    add: 'Creates a dataset in a MiGA project',
-    get: 'Downloads a dataset from public databases into a MiGA project',
-    ncbi_get: 'Downloads all genomes in a taxon from NCBI into a MiGA project',
-    rm: 'Removes a dataset from an MiGA project',
-    find: 'Finds unregistered datasets based on result files',
+    add: 'Create a dataset in a MiGA project',
+    get: 'Download a dataset from public databases into a MiGA project',
+    ncbi_get: 'Download all genomes in a taxon from NCBI into a MiGA project',
+    rm: 'Remove a dataset from an MiGA project',
+    find: 'Find unregistered datasets based on result files',
     ln: 'Link datasets (including results) from one project to another',
-    ls: 'Lists all registered datasets in an MiGA project',
-    archive: 'Generates a tar-ball with all files from select datasets',
+    ls: 'List all registered datasets in an MiGA project',
+    archive: 'Generate a tar-ball with all files from select datasets',
     # Results
-    add_result: 'Registers a result',
-    stats: 'Extracts statistics for the given result',
-    files: 'Lists registered files from the results of a dataset or project',
-    run: 'Executes locally one step analysis producing the given result',
-    summary: 'Generates a summary table for the statistics of all datasets',
-    next_step: 'Returns the next task to run in a dataset or project',
+    add_result: 'Register a result',
+    stats: 'Extract statistics for the given result',
+    files: 'List registered files from the results of a dataset or project',
+    run: 'Execute locally one step analysis producing the given result',
+    summary: 'Generate a summary table for the statistics of all datasets',
+    next_step: 'Return the next task to run in a dataset or project',
     # Objects (Datasets or Projects)
-    edit: 'Edits the metadata of a dataset or project',
+    edit: 'Edit the metadata of a dataset or project',
     # System
     init: 'Initialize MiGA to process new projects',
-    daemon: 'Controls the daemon of a MiGA project',
-    lair: 'Controls groups of daemons for several MiGA projects',
-    date: 'Returns the current date in standard MiGA format',
-    console: 'Opens an IRB console with MiGA',
+    daemon: 'Control the daemon of a MiGA project',
+    lair: 'Control groups of daemons for several MiGA projects',
+    date: 'Return the current date in standard MiGA format',
+    console: 'Open an IRB console with MiGA',
     # Taxonomy
-    tax_set: 'Registers taxonomic information for datasets',
-    tax_test: 'Returns test of taxonomic distributions for query datasets',
-    tax_index: 'Creates a taxonomy-indexed list of the datasets',
-    tax_dist: 'Estimates distributions of distance by taxonomy',
+    tax_set: 'Register taxonomic information for datasets',
+    tax_test: 'Return test of taxonomic distributions for query datasets',
+    tax_index: 'Create a taxonomy-indexed list of the datasets',
+    tax_dist: 'Estimate distributions of distance by taxonomy',
   }
 
   @@TASK_ALIAS = {

data/lib/miga/common/format.rb

@@ -68,15 +68,20 @@ module MiGA::Common::Format
   # a FastA or FastQ file (supports gzipped files). The +format+ must be a
   # Symbol, one of +:fasta+ or +:fastq+. Additional estimations can be
   # controlled via the +opts+ Hash. Supported options include:
-  # - +:n50+: If true, it also returns the N50 and the median (in bp)
-  # - +:gc+: If true, it also returns the G+C content (in %)
-  # - +:x+: If true, it also returns the undetermined bases content (in %)
+  # - +:n50+: Include the N50 and the median (in bp)
+  # - +:gc+: Include the G+C content (in %)
+  # - +:x+: Include the undetermined bases content (in %)
+  # - +:skew+: Include G-C and A-T sequence skew (in %; forces gc: true).
+  #   See definition used here in DOI:10.1177/117693430700300006
   def seqs_length(file, format, opts = {})
+    opts[:gc] = true if opts[:skew]
     fh = file =~ /\.gz/ ? Zlib::GzipReader.open(file) : File.open(file, 'r')
     l = []
     gc = 0
     xn = 0
-    i = 0 # <- Zlib::GzipReader doesn't set `$.`
+    t  = 0
+    c  = 0
+    i  = 0 # <- Zlib::GzipReader doesn't set `$.`
     fh.each_line do |ln|
       i += 1
       if (format == :fasta and ln =~ /^>/) or
@@ -86,6 +91,10 @@ module MiGA::Common::Format
         l[l.size - 1] += ln.chomp.size
         gc += ln.scan(/[GCgc]/).count if opts[:gc]
         xn += ln.scan(/[XNxn]/).count if opts[:x]
+        if opts[:skew]
+          t += ln.scan(/[Tt]/).count
+          c += ln.scan(/[Cc]/).count
+        end
       end
     end
     fh.close
@@ -97,6 +106,12 @@ module MiGA::Common::Format
     o[:sd]  = Math.sqrt o[:var]
     o[:gc]  = 100.0 * gc / o[:tot] if opts[:gc]
     o[:x]   = 100.0 * xn / o[:tot] if opts[:x]
+    if opts[:skew]
+      at = o[:tot] - gc
+      o[:at_skew] = 100.0 * (2 * t - at) / at
+      o[:gc_skew] = 100.0 * (2 * c - gc) / gc
+    end
+
     if opts[:n50]
       l.sort!
       thr = o[:tot] / 2
@@ -132,9 +147,14 @@ class String
   end
 
   ##
-  # Replace underscores by spaces or dots (depending on context).
+  # Replace underscores by spaces or other symbols depending on context
   def unmiga_name
-    gsub(/_(str|sp|subsp|pv)__/, '_\\1._').tr('_', ' ')
+    gsub(/_(str|sp|subsp|pv)__/, '_\\1._')
+      .gsub(/g_c_(content)/, 'G+C \\1')
+      .gsub(/g_c_(skew)/, 'G-C \\1')
+      .gsub(/a_t_(skew)/, 'A-T \\1')
+      .gsub(/x_content/, &:capitalize)
+      .tr('_', ' ')
   end
 
   ##

data/lib/miga/daemon.rb

@@ -72,6 +72,7 @@ class MiGA::Daemon < MiGA::MiGA
     say '-----------------------------------'
     say 'MiGA:%s launched' % project.name
     say '-----------------------------------'
+    recalculate_status!
     load_status
     say 'Configuration options:'
     say @runopts.to_s
@@ -99,6 +100,7 @@ class MiGA::Daemon < MiGA::MiGA
   end
 
   def recalculate_status!
+    say 'Recalculating status for all datasets'
     project.each_dataset(&:recalculate_status)
   end
 
@@ -158,8 +160,8 @@ class MiGA::Daemon < MiGA::MiGA
   end
 
   ##
-  # Traverse datasets, and returns boolean indicating if at any datasets
-  # are incomplete
+  # Traverse datasets, and returns boolean indicating if at any reference
+  # datasets are incomplete
   def check_datasets
     l_say(2, 'Checking datasets')
     o = false
@@ -167,7 +169,7 @@ class MiGA::Daemon < MiGA::MiGA
       next unless ds.status == :incomplete
       next if ds.next_preprocessing(false).nil?
 
-      o = true
+      o = true if ds.ref?
       queue_job(:d, ds)
     end
     o
@@ -183,7 +185,7 @@ class MiGA::Daemon < MiGA::MiGA
     return if project.dataset_names.empty?
 
     # Double-check if all datasets are ready
-    return unless project.done_preprocessing?(false)
+    return unless project.done_preprocessing?
 
     # Queue project-level job
     to_run = project.next_task(nil, false)

data/lib/miga/dataset.rb

@@ -97,7 +97,10 @@ class MiGA::Dataset < MiGA::MiGA
 
   ##
   # Inactivate a dataset. This halts automated processing by the daemon
-  def inactivate!
+  # 
+  # If given, the +reason+ string is saved as a metadata +:warn+ entry
+  def inactivate!(reason = nil)
+    metadata[:warn] = "Inactive: #{reason}" unless reason.nil?
     metadata[:inactive] = true
     metadata.save
     pull_hook :on_inactivate
@@ -107,6 +110,7 @@ class MiGA::Dataset < MiGA::MiGA
   # Activate a dataset. This removes the +:inactive+ flag
   def activate!
     metadata[:inactive] = nil
+    metadata[:warn] = nil if metadata[:warn] && metadata[:warn] =~ /^Inactive: /
     metadata.save
     pull_hook :on_activate
   end

data/lib/miga/dataset/base.rb

@@ -35,8 +35,8 @@ module MiGA::Dataset::Base
     mytaxa: '07.annotation/02.taxonomy/01.mytaxa',
     mytaxa_scan: '07.annotation/03.qa/02.mytaxa_scan',
     # Distances (for single-species datasets)
-    distances: '09.distances',
     taxonomy: '09.distances/05.taxonomy',
+    distances: '09.distances',
     # General statistics
     stats: '90.stats'
   }
@@ -66,7 +66,7 @@ module MiGA::Dataset::Base
   @@PREPROCESSING_TASKS = [
     :raw_reads, :trimmed_reads, :read_quality, :trimmed_fasta,
     :assembly, :cds, :essential_genes, :ssu, :mytaxa, :mytaxa_scan,
-    :distances, :taxonomy, :stats
+    :taxonomy, :distances, :stats
   ]
 
   ##
@@ -77,7 +77,7 @@ module MiGA::Dataset::Base
   ##
   # Tasks to be executed only in datasets that are not multi-organism. These
   # tasks are ignored for multi-organism datasets or for unknown types.
-  @@ONLY_NONMULTI_TASKS = [:mytaxa_scan, :distances, :taxonomy]
+  @@ONLY_NONMULTI_TASKS = [:mytaxa_scan, :taxonomy, :distances]
   @@_ONLY_NONMULTI_TASKS_H = Hash[@@ONLY_NONMULTI_TASKS.map { |i| [i, true] }]
 
   ##

data/lib/miga/dataset/hooks.rb

@@ -52,15 +52,15 @@ module MiGA::Dataset::Hooks
   end
 
   ##
-  # Run +cmd+ in the command-line with {{variables}}: dataset, project, miga,
-  # object (as defined for the event, if any)
+  # Run +cmd+ in the command-line with {{variables}}:
+  # dataset, project, project_name, miga, object (if defined for the event)
   # - +hook_args+: +[cmd]+
   # - +event_args+: +[object (optional)]+
   def hook_run_cmd(hook_args, event_args)
     Process.wait(
       spawn hook_args.first.miga_variables(
-        dataset: name, project: project.path, miga: MiGA::MiGA.root_path,
-        object: event_args.first
+        dataset: name, project: project.path, project_name: project.name,
+        miga: MiGA::MiGA.root_path, object: event_args.first
       )
     )
   end

data/lib/miga/dataset/result.rb

@@ -26,15 +26,24 @@ module MiGA::Dataset::Result
   # The values are symbols:
   # - empty: the dataset has no data
   # - inactive: the dataset is inactive
+  # - upstream: the task is upstream from dataset's input
   # - force: forced to ignore by metadata
   # - project: incompatible project
   # - noref: incompatible dataset, only for reference
   # - multi: incompatible dataset, only for multi
   # - nonmulti: incompatible dataset, only for nonmulti
+  # - complete: the task is already complete
   # - execute: do not ignore, execute the task
   def why_ignore(task)
-    if !active?
+    if !get_result(task).nil?
+      :complete
+    elsif !active?
       :inactive
+    elsif first_preprocessing.nil?
+      :empty
+    elsif @@PREPROCESSING_TASKS.index(task) <
+          @@PREPROCESSING_TASKS.index(first_preprocessing)
+      :upstream
     elsif !metadata["run_#{task}"].nil?
       metadata["run_#{task}"] ? :execute : :force
     elsif task == :taxonomy && project.metadata[:ref_project].nil?
@@ -56,7 +65,7 @@ module MiGA::Dataset::Result
   # initial input. Passes +save+ to #add_result.
   def first_preprocessing(save = false)
     @first_processing ||= @@PREPROCESSING_TASKS.find do |t|
-      !ignore_task?(t) && !add_result(t, save).nil?
+      !add_result(t, save).nil?
     end
   end
 
@@ -70,7 +79,7 @@ module MiGA::Dataset::Result
         false
       elsif add_result(t, save).nil?
         if (metadata["_try_#{t}"] || 0) > (project.metadata[:max_try] || 10)
-          inactivate!
+          inactivate! "Too many errors in step #{t}"
           false
         else
           true
@@ -121,17 +130,12 @@ module MiGA::Dataset::Result
   # - complete: a task with registered results
   # - pending: a task queued to be performed
   def result_status(task)
-    if first_preprocessing.nil?
-      :ignore_empty
-    elsif !get_result(task).nil?
-      :complete
-    elsif @@PREPROCESSING_TASKS.index(task) <
-          @@PREPROCESSING_TASKS.index(first_preprocessing)
-      :-
-    elsif ignore_task?(task)
-      :"ignore_#{why_ignore task}"
-    else
-      :pending
+    reason = why_ignore(task)
+    case reason
+    when :upstream; :-
+    when :execute; :pending
+    when :complete; :complete
+    else; :"ignore_#{reason}"
     end
   end
 

data/lib/miga/lair.rb

@@ -111,7 +111,7 @@ class MiGA::Lair < MiGA::MiGA
 
         yield(project)
       elsif Dir.exist? f
-        each_project(f) { |project| yield(project) }
+        each_project(f) { |p| yield(p) }
       end
     end
   end

data/lib/miga/project/dataset.rb

@@ -134,12 +134,10 @@ module MiGA::Project::Dataset
   ##
   # Are all the datasets in the project preprocessed? Save intermediate results
   # if +save+ (until the first incomplete dataset is reached).
-  def done_preprocessing?(save = true)
-    dataset_names.each do |dn|
-      ds = dataset(dn)
-      return false if ds.is_ref? and not ds.done_preprocessing?(save)
+  def done_preprocessing?(save = false)
+    !each_dataset.any? do |d|
+      d.ref? && d.active? && !d.done_preprocessing?(save)
     end
-    true
   end
 
   ##