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 != 1 && 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 )
73 if cla.get_number_of_files == 4
74 mapfile = cla.get_file_name( 0 )
75 input = cla.get_file_name( 1 )
76 output = cla.get_file_name( 2 )
77 list_file = cla.get_file_name( 3 )
78 elsif cla.get_number_of_files == 3
79 input = cla.get_file_name( 0 )
80 output = cla.get_file_name( 1 )
81 list_file = cla.get_file_name( 2 )
82 elsif cla.get_number_of_files == 1
83 input = cla.get_file_name( 0 )
85 if input.downcase.end_with?( ".fasta" )
86 i = input[ 0 .. input.length - 6 ]
90 output = i + "_ni.fasta"
91 list_file = i + ".nim"
95 allowed_opts = Array.new
96 allowed_opts.push( EXTRACT_TAXONOMY_OPTION )
98 disallowed = cla.validate_allowed_options_as_str( allowed_opts )
99 if ( disallowed.length > 0 )
100 Util.fatal_error( PRG_NAME, "unknown option(s): " + disallowed )
103 extract_taxonomy = false
104 if ( cla.is_option_set?( EXTRACT_TAXONOMY_OPTION ) )
105 extract_taxonomy = true
108 if ( File.exists?( output ) )
109 Util.fatal_error( PRG_NAME, "outfile [" + output + "] already exists" )
111 if ( File.exists?( list_file ) )
112 Util.fatal_error( PRG_NAME, "list file [" + list_file + "] already exists" )
114 if ( !File.exists?( input) )
115 Util.fatal_error( PRG_NAME, "infile [" + input + "] does not exist" )
117 if ( mapfile != nil && !File.exists?( mapfile ) )
118 Util.fatal_error( PRG_NAME, "mapfile [" + mapfile + "] does not exist" )
121 fasta_like = Util.looks_like_fasta?( input )
125 puts( "Map file : " + mapfile )
127 puts( "Input alignment : " + input )
128 puts( "Output alignment: " + output )
129 puts( "Name list : " + list_file )
131 puts( "Format : Fasta" )
133 puts( "Format : Phylip like" )
135 if ( extract_taxonomy )
136 puts( "Extract taxonomy: true" )
140 species_map = Hash.new
142 File.open( mapfile ) do | file |
143 while line = file.gets
144 if ( line =~/(.+)#(.+)/ || line =~/(.+)\s+(.+)/ )
145 species_map[ $1 ] = $2
146 Util.print_message( PRG_NAME, "mapping: " + $1 + ' => ' + $2 )
155 msa = f.create_msa_from_file( input, FastaParser.new() )
157 msa = f.create_msa_from_file( input, GeneralMsaParser.new() )
159 rescue Exception => e
160 Util.fatal_error( PRG_NAME, "failed to read file: " + e.to_s )
163 if ( msa == nil || msa.get_number_of_seqs() < 1 )
164 Util.fatal_error( PRG_NAME, "failed to read MSA" )
167 Util.check_file_for_writability( list_file )
168 rescue Exception => e
169 Util.fatal_error( PRG_NAME, "error: " + e.to_, STDOUT )
172 #removed = msa.remove_redundant_sequences!( true )
174 # Util.print_message( PRG_NAME, "going to ignore the following " + removed.size.to_s + " redundant sequences:" )
175 # removed.each { | seq_name |
178 # Util.print_message( PRG_NAME, "will process " + msa.get_number_of_seqs.to_s + " non redundant sequences" )
181 lf = File.open( list_file, "a" )
182 for i in 0 ... msa.get_number_of_seqs
183 seq = msa.get_sequence( i )
184 seq.set_name( Util::normalize_seq_name( modify_name( seq.get_name(), i, lf, species_map, extract_taxonomy ), 10 ) )
190 w = FastaWriter.new()
192 w = PhylipSequentialWriter.new()
194 w.set_max_name_length( 10 )
197 io.write_to_file( msa, output, w )
198 rescue Exception => e
199 Util.fatal_error( PRG_NAME, "failed to write file: " + e.to_s )
202 if ( @taxonomies.length > 0 )
203 Util.print_message( PRG_NAME, "number of unique taxonomies: " + @taxonomies.length.to_s )
205 Util.print_message( PRG_NAME, "wrote: " + list_file )
206 Util.print_message( PRG_NAME, "wrote: " + output )
207 Util.print_message( PRG_NAME, "OK" )
212 def modify_name( desc, counter, file, species_map, extract_taxonomy )
215 # if desc =~ /^>?\s*\S{1,10}_([0-9A-Z]{3,5})/
216 if desc =~ /^>?\s*\S{1,10}_([A-Z]{3,5})/
217 new_desc = counter.to_s( 16 ) + "_" + $1
219 new_desc = counter.to_s( 16 )
220 elsif extract_taxonomy
221 if ( desc.count( "[" ) != desc.count( "]" ) )
222 Util.fatal_error( PRG_NAME, "illegal bracket count in: " + desc )
225 species_map.each_key do | key |
226 if desc =~ /[\b|_]#{key}\b/ # Added boundaries to prevent e.g. RAT matching ARATH.
227 species = species_map[ key ]
228 new_desc = counter.to_s( 16 ) + "_" + species
233 if desc =~/.*\[(\S{3,}?)\]/
237 species.gsub!( /\s+/, " " )
238 species.gsub!( /-/, "" )
239 species.gsub!( /\)/, "" )
240 species.gsub!( /\(/, "" )
241 species.gsub!( /\'/, "" )
242 if species =~ /\S+\s\S+/ || species =~ /\S{3,5}/
243 if species =~ /(\S+)\s(\S+)/
244 code = $1[ 0..2 ] + $2[ 0..1 ]
245 elsif species =~ /\S{3,5}/
247 elsif species.count( " " ) > 2
248 species =~ /(\S+)\s+(\S+)\s+(\S+)$/
252 code = code[ 0 ] + third_last[ 0 ] + second_last[ 0 ] + last[ 0 ] + last[ last.size - 1 ]
253 elsif species.count( " " ) > 1
254 species =~ /(\S+)\s+(\S+)$/
257 code = code[ 0..1 ] + second_last[ 0 ] + last[ 0 ] + last[ last.size - 1 ]
259 new_desc = counter.to_s( 16 ) + "_" + code
260 if @taxonomies.has_key?( code )
261 if ( !@taxonomies.has_value?( species ) )
262 Util.fatal_error( PRG_NAME, "code [#{code}] is not unique in [#{desc}]" )
265 if ( @taxonomies.has_value?( species ) )
266 Util.fatal_error( PRG_NAME, "genome [#{species}] is not unique in [#{desc}]" )
268 @taxonomies[ code ] = species
272 Util.fatal_error( PRG_NAME, "illegal format [#{species}] in: " + desc )
275 Util.fatal_error( PRG_NAME, "illegal format in: " + desc )
281 species_map.each_key do | key |
283 species = species_map[ key ]
284 species = species.gsub( /\s+/, "" )
285 species = species.gsub( /_/, " " )
287 if species =~ /(\S+)\s+(\S+)/
288 species = $1[0..2] + $2[0..1]
290 species = species.gsub( /\s+/, "" )
291 species = species.slice(0, 5)
297 Util.fatal_error( PRG_NAME, "species not found in: " + desc )
299 new_desc = counter.to_s( 16 ) + "_" + species
303 Util.fatal_error( PRG_NAME, "failed to extract species from: " + desc )
306 file.print( new_desc + ": " + desc + " [" + my_species + "]" + "\n" )
308 file.print( new_desc + ": " + desc + "\n" )
313 end # class TaxonomyProcessor