2 # = lib/evo/apps/taxonomy_processor - TaxonomyProcessor class
4 # Copyright:: Copyright (C) 2006-2007 Christian M. Zmasek
5 # License:: GNU Lesser General Public License (LGPL)
7 # $Id: taxonomy_processor.rb,v 1.26 2010/12/13 19:00:11 cmzmasek Exp $
10 require 'lib/evo/util/util'
11 require 'lib/evo/msa/msa_factory'
12 require 'lib/evo/msa/msa'
13 require 'lib/evo/io/msa_io'
14 require 'lib/evo/io/parser/fasta_parser'
15 require 'lib/evo/io/parser/general_msa_parser'
16 require 'lib/evo/io/writer/fasta_writer'
17 require 'lib/evo/io/writer/phylip_sequential_writer'
18 require 'lib/evo/util/command_line_arguments'
22 class TaxonomyProcessor
25 PRG_DATE = "2010.02.24"
26 PRG_DESC = "replacement of species names in multiple sequence files"
28 COPYRIGHT = "2010 Christian M Zmasek"
29 CONTACT = "phylosoft@gmail.com"
30 WWW = "www.phylosoft.org"
34 EXTRACT_TAXONOMY_OPTION = "t"
37 @taxonomies = Hash.new()
42 Util.print_program_information( PRG_NAME,
51 if ( ARGV == nil || ( ARGV.length != 3 && ARGV.length != 4 && ARGV.length != 5 && ARGV.length != 6 ) )
52 puts( "Usage: #{PRG_NAME}.rb [options] [input map file] <input sequences> <output sequences> <output id list>" )
54 puts( " options: -" + EXTRACT_TAXONOMY_OPTION + ": to extract taxonomy information from bracketed expression" )
60 cla = CommandLineArguments.new( ARGV )
61 rescue ArgumentError => e
62 Util.fatal_error( PRG_NAME, "error: " + e.to_s )
70 if cla.get_number_of_files == 4
71 mapfile = cla.get_file_name( 0 )
72 input = cla.get_file_name( 1 )
73 output = cla.get_file_name( 2 )
74 list_file = cla.get_file_name( 3 )
75 elsif cla.get_number_of_files == 3
76 input = cla.get_file_name( 0 )
77 output = cla.get_file_name( 1 )
78 list_file = cla.get_file_name( 2 )
82 allowed_opts = Array.new
83 allowed_opts.push( EXTRACT_TAXONOMY_OPTION )
85 disallowed = cla.validate_allowed_options_as_str( allowed_opts )
86 if ( disallowed.length > 0 )
87 Util.fatal_error( PRG_NAME, "unknown option(s): " + disallowed )
90 extract_taxonomy = false
91 if ( cla.is_option_set?( EXTRACT_TAXONOMY_OPTION ) )
92 extract_taxonomy = true
95 if ( File.exists?( output ) )
96 Util.fatal_error( PRG_NAME, "outfile [" + output + "] already exists" )
98 if ( File.exists?( list_file ) )
99 Util.fatal_error( PRG_NAME, "list file [" + list_file + "] already exists" )
101 if ( !File.exists?( input) )
102 Util.fatal_error( PRG_NAME, "infile [" + input + "] does not exist" )
104 if ( mapfile != nil && !File.exists?( mapfile ) )
105 Util.fatal_error( PRG_NAME, "mapfile [" + mapfile + "] does not exist" )
108 fasta_like = Util.looks_like_fasta?( input )
112 puts( "Map file : " + mapfile )
114 puts( "Input alignment : " + input )
115 puts( "Output alignment: " + output )
116 puts( "Name list : " + list_file )
118 puts( "Format : Fasta" )
120 puts( "Format : Phylip like" )
122 if ( extract_taxonomy )
123 puts( "Extract taxonomy: true" )
127 species_map = Hash.new
129 File.open( mapfile ) do | file |
130 while line = file.gets
131 if ( line =~/(.+)#(.+)/ || line =~/(.+)\s+(.+)/ )
132 species_map[ $1 ] = $2
133 Util.print_message( PRG_NAME, "mapping: " + $1 + ' => ' + $2 )
142 msa = f.create_msa_from_file( input, FastaParser.new() )
144 msa = f.create_msa_from_file( input, GeneralMsaParser.new() )
146 rescue Exception => e
147 Util.fatal_error( PRG_NAME, "failed to read file: " + e.to_s )
150 if ( msa == nil || msa.get_number_of_seqs() < 1 )
151 Util.fatal_error( PRG_NAME, "failed to read MSA" )
154 Util.check_file_for_writability( list_file )
155 rescue Exception => e
156 Util.fatal_error( PRG_NAME, "error: " + e.to_, STDOUT )
159 #removed = msa.remove_redundant_sequences!( true )
161 # Util.print_message( PRG_NAME, "going to ignore the following " + removed.size.to_s + " redundant sequences:" )
162 # removed.each { | seq_name |
165 # Util.print_message( PRG_NAME, "will process " + msa.get_number_of_seqs.to_s + " non redundant sequences" )
168 lf = File.open( list_file, "a" )
169 for i in 0 ... msa.get_number_of_seqs
170 seq = msa.get_sequence( i )
171 seq.set_name( Util::normalize_seq_name( modify_name( seq.get_name(), i, lf, species_map, extract_taxonomy ), 10 ) )
177 w = FastaWriter.new()
179 w = PhylipSequentialWriter.new()
181 w.set_max_name_length( 10 )
184 io.write_to_file( msa, output, w )
185 rescue Exception => e
186 Util.fatal_error( PRG_NAME, "failed to write file: " + e.to_s )
189 if ( @taxonomies.length > 0 )
190 Util.print_message( PRG_NAME, "number of unique taxonomies: " + @taxonomies.length.to_s )
192 Util.print_message( PRG_NAME, "wrote: " + list_file )
193 Util.print_message( PRG_NAME, "wrote: " + output )
194 Util.print_message( PRG_NAME, "OK" )
199 def modify_name( desc, counter, file, species_map, extract_taxonomy )
202 if desc =~ /^>?\s*\S{1,10}_([0-9A-Z]{3,5})/
203 new_desc = counter.to_s( 16 ) + "_" + $1
205 new_desc = counter.to_s( 16 )
206 elsif extract_taxonomy
207 if ( desc.count( "[" ) != desc.count( "]" ) )
208 Util.fatal_error( PRG_NAME, "illegal bracket count in: " + desc )
211 species_map.each_key do | key |
212 if desc =~ /[\b|_]#{key}\b/ # Added boundaries to prevent e.g. RAT matching ARATH.
213 species = species_map[ key ]
214 new_desc = counter.to_s( 16 ) + "_" + species
219 if desc =~/.*\[(\S{3,}?)\]/
223 species.gsub!( /\s+/, " " )
224 species.gsub!( /-/, "" )
225 species.gsub!( /\)/, "" )
226 species.gsub!( /\(/, "" )
227 species.gsub!( /\'/, "" )
228 if species =~ /\S+\s\S+/ || species =~ /\S{3,5}/
229 if species =~ /(\S+)\s(\S+)/
230 code = $1[ 0..2 ] + $2[ 0..1 ]
231 elsif species =~ /\S{3,5}/
233 elsif species.count( " " ) > 2
234 species =~ /(\S+)\s+(\S+)\s+(\S+)$/
238 code = code[ 0 ] + third_last[ 0 ] + second_last[ 0 ] + last[ 0 ] + last[ last.size - 1 ]
239 elsif species.count( " " ) > 1
240 species =~ /(\S+)\s+(\S+)$/
243 code = code[ 0..1 ] + second_last[ 0 ] + last[ 0 ] + last[ last.size - 1 ]
245 new_desc = counter.to_s( 16 ) + "_" + code
246 if @taxonomies.has_key?( code )
247 if ( !@taxonomies.has_value?( species ) )
248 Util.fatal_error( PRG_NAME, "code [#{code}] is not unique in [#{desc}]" )
251 if ( @taxonomies.has_value?( species ) )
252 Util.fatal_error( PRG_NAME, "genome [#{species}] is not unique in [#{desc}]" )
254 @taxonomies[ code ] = species
258 Util.fatal_error( PRG_NAME, "illegal format [#{species}] in: " + desc )
261 Util.fatal_error( PRG_NAME, "illegal format in: " + desc )
267 species_map.each_key do | key |
269 species = species_map[ key ]
270 species = species.gsub( /\s+/, "" )
271 species = species.gsub( /_/, " " )
273 if species =~ /(\S+)\s+(\S+)/
274 species = $1[0..2] + $2[0..1]
276 species = species.gsub( /\s+/, "" )
277 species = species.slice(0, 5)
283 Util.fatal_error( PRG_NAME, "species not found in: " + desc )
285 new_desc = counter.to_s( 16 ) + "_" + species
289 Util.fatal_error( PRG_NAME, "failed to extract species from: " + desc )
292 file.print( new_desc + ": " + desc + " [" + my_species + "]" + "\n" )
294 file.print( new_desc + ": " + desc + "\n" )
299 end # class TaxonomyProcessor