scbi_distributed_blast 0.0.3 → 0.0.4

data/History.txt

@@ -1,3 +1,7 @@
+=== 0.0.4 2011-07-12
+
+updated documentation
+
 === 0.0.3 2011-07-07
 
 prerelease

data/README.rdoc

@@ -4,7 +4,7 @@
 
 == DESCRIPTION:
 
-scbi_distributed_blast is a simple distribution mechanism for blast+ made on top of scbi_mapreduce. With scbi_distributed_blast you can perform distributed blasts using a cluster or a set of machines of your network. It uses the version of blast+ that you have installed.
+scbi_distributed_blast is a simple distribution mechanism for blast+ made on top of scbi_mapreduce. With scbi_distributed_blast you can perform distributed blasts using a cluster, a set of machines of your network or your own multi-core personal computer. It uses the same blast+ that you have already installed.
  
 == FEATURES:
 
@@ -16,15 +16,15 @@ scbi_distributed_blast is a simple distribution mechanism for blast+ made on top
 
 Once installed, scbi_distributed_blast is very easy to use. To launch it locally in your own personal computer using 8 cores, you can do:
   
-  $> scbi_distributed_blast -s 10.0.0 -w 8 'full_blast_cmd'
+  $> scbi_distributed_blast -w 8 'full_blast_cmd'
   
 Where full_blast_cmd is the blast+ cmd that you would write to execute your desired blast search. Eg.:
 
-  $> scbi_distributed_blast -s 10.0.0 -w 8  'blastn -task blastn-short -db my_db.fasta -query ~/seqs/sample.fasta -outfmt 6 -out output_file'
+  $> scbi_distributed_blast -w 8  'blastn -task blastn-short -db my_db.fasta -query ~/seqs/sample.fasta -outfmt 6 -out output_file'
 
 Sequences are sent in chunks of 100, but you can change this value by using the -g parameter:
 
-  $> scbi_distributed_blast -s 10.0.0 -w 8 -g 200 'blastn -task blastn-short -db my_db.fasta -query ~/seqs/sample.fasta -outfmt 6 -out output_file'
+  $> scbi_distributed_blast -w 8 -g 200 'blastn -task blastn-short -db my_db.fasta -query ~/seqs/sample.fasta -outfmt 6 -out output_file'
 
 To get additional help:
 

data/bin/scbi_distributed_blast

@@ -8,6 +8,7 @@ require 'scbi_distributed_blast'
 require 'logger'
 require 'optparse'
 
+# retrieve environment variables
 if ENV['SCBI_DISTRIBUTED_BLAST_INIT'] && File.exists?(ENV['SCBI_DISTRIBUTED_BLAST_INIT'])
   $INIT_FILE=File.expand_path(ENV['SCBI_DISTRIBUTED_BLAST_INIT'])
 elsif File.exists?(File.join('~','scbi_distributed_blast_init_env'))
@@ -17,8 +18,8 @@ else
 end
 
 
+# parse input options
 options = {}
-
 optparse = OptionParser.new do |opts|
 
   # Set a banner, displayed at the top
@@ -27,11 +28,13 @@ optparse = OptionParser.new do |opts|
 
   # Define the options, and what they do
   
+  # server ip
   options[:server_ip] = '0.0.0.0'
   opts.on( '-s', '--server IP', 'Server ip. You can use a partial ip to select the apropriate interface' ) do |server_ip|
     options[:server_ip] = server_ip
   end
 
+  # server port
   options[:port] = 0 # any free port
   opts.on( '-p', '--port PORT', 'Server port. If set to 0, an arbitrary empty port will be used') do |port|
     options[:port] = port.to_i
@@ -55,17 +58,18 @@ optparse = OptionParser.new do |opts|
     end
   end
 
+  # chunk size
   options[:chunk_size] = 100
   opts.on( '-g', '--group_size chunk_size', 'Group sequences in chunks of size <chunk_size>' ) do |cs|
     options[:chunk_size] = cs.to_i
   end
   
+  # log_file
   options[:log_file] = STDOUT
   opts.on( '-l', '--log_file file', 'Define a log file. STDOUT by default' ) do |cs|
     options[:log_file] = cs
   end
   
-
   # This displays the help screen, all programs are
   # assumed to have this option.
   opts.on_tail( '-h', '--help', 'Display this screen' ) do
@@ -78,8 +82,10 @@ end
 # parse options and remove from ARGV
 optparse.parse!
 
+# use remaining command line arguments as blast_cmd
 blast_cmd = ARGV.join(' ')
 
+# enable logger
 $LOG = Logger.new(options[:log_file])
 $LOG.datetime_format = "%Y-%m-%d %H:%M:%S"
 
@@ -89,6 +95,7 @@ $LOG.info "Original Blast+ CMD: #{blast_cmd}"
 input_file = nil
 output_file = nil
 
+# extract query and output files from blast_cmd
 if blast_cmd.upcase.index('-QUERY')
   params=blast_cmd.split(' -')
   params.reverse_each do |param|
@@ -109,19 +116,13 @@ if blast_cmd.upcase.index('-QUERY')
       params.delete(param)
       # break
     end
-      
-      
   end
   
   blast_cmd=params.join(' -')
 end
 
-# puts "BLASTCMD: #{blast_cmd}"
-# puts "Input file: #{input_file}"
-# puts "Output file: #{output_file}"
-
-if !input_file.nil? and File.exists?(File.expand_path(input_file))
-  
+# check if query is defined
+if !input_file.nil? and File.exists?(File.expand_path(input_file))  
   $LOG.info "Query input file: #{input_file}"
 else
   $LOG.error "No input file specified in blast command (-query parameter)"

data/lib/scbi_distributed_blast/scbi_dblast_manager.rb

@@ -12,13 +12,17 @@ class ScbiDblastManager < ScbiMapreduce::WorkManager
   # worker connection, and thus, all global variables here should be
   # class variables (starting with @@)
   def self.init_work_manager(input_file, blast_cmd, output_file)
+    # save blast_cmd
     @@blast_cmd=blast_cmd
+    
+    # define output file
     if output_file.nil?
       @@output_file=STDOUT
     else
       @@output_file=File.open(output_file,'w')
     end
 
+    # open input file in fasta
     @@fqr = FastaQualFile.new(input_file)
     
   end
@@ -26,6 +30,7 @@ class ScbiDblastManager < ScbiMapreduce::WorkManager
   # end_work_manager is executed at the end, when all the process is done.
   # You can use it to close files opened in init_work_manager
   def self.end_work_manager
+    # close opened files
     @@fqr.close
     @@output_file.close if @@output_file!=STDOUT
   end
@@ -33,6 +38,7 @@ class ScbiDblastManager < ScbiMapreduce::WorkManager
   # worker_initial_config is used to send initial parameters to workers.
   # The method is executed once per each worker
   def worker_initial_config
+    # send blast_cmd to workers
     {:blast_cmd=>@@blast_cmd}
   end
 
@@ -41,6 +47,7 @@ class ScbiDblastManager < ScbiMapreduce::WorkManager
   # This method must return the work data or nil if no more data is available
   def next_work
     
+    # read next sequence from inputfile
     n,f = @@fqr.next_seq
 
     if n.nil?
@@ -55,8 +62,8 @@ class ScbiDblastManager < ScbiMapreduce::WorkManager
   # work_received is executed each time a worker has finished a job.
   # Here you can write results down to disk, perform some aggregated statistics, etc...
   def work_received(results)
+    # write results to disk
     @@output_file.puts results
-    # write_data_to_disk(results)
   end
 
 end

data/lib/scbi_distributed_blast.rb

@@ -7,5 +7,5 @@ require 'scbi_distributed_blast/scbi_dblast_manager.rb'
 ROOT_PATH=File.join(File.dirname(__FILE__),'scbi_distributed_blast')
 
 module ScbiDistributedBlast
-   VERSION = '0.0.3'
+   VERSION = '0.0.4'
 end

metadata

@@ -2,7 +2,7 @@
 name: scbi_distributed_blast
 version: !ruby/object:Gem::Version 
   prerelease: 
-  version: 0.0.3
+  version: 0.0.4
 platform: ruby
 authors: 
 - Dario Guerrero
@@ -10,7 +10,7 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2011-07-08 00:00:00 Z
+date: 2011-07-12 00:00:00 Z
 dependencies: 
 - !ruby/object:Gem::Dependency 
   name: scbi_mapreduce
@@ -56,7 +56,7 @@ dependencies:
         version: 2.8.0
   type: :development
   version_requirements: *id004
-description: scbi_distributed_blast is a simple distribution mechanism for blast+ made on top of scbi_mapreduce. With scbi_distributed_blast you can perform distributed blasts using a cluster or a set of machines of your network. It uses the version of blast+ that you have installed.
+description: scbi_distributed_blast is a simple distribution mechanism for blast+ made on top of scbi_mapreduce. With scbi_distributed_blast you can perform distributed blasts using a cluster, a set of machines of your network or your own multi-core personal computer. It uses the same blast+ that you have already installed.
 email: 
 - dariogf@gmail.com
 executables: