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"
32 EXTRACT_TAXONOMY_OPTION = "t"
35 @taxonomies = Hash.new()
40 Util.print_program_information( PRG_NAME,
49 if ( ARGV == nil || ( ARGV.length != 3 && ARGV.length != 4 && ARGV.length != 5 && ARGV.length != 6 ) )
50 puts( "Usage: #{PRG_NAME}.rb [options] [input map file] <input sequences> <output sequences> <output id list>" )
52 puts( " options: -" + EXTRACT_TAXONOMY_OPTION + ": to extract taxonomy information from bracketed expression" )
58 cla = CommandLineArguments.new( ARGV )
59 rescue ArgumentError => e
60 Util.fatal_error( PRG_NAME, "error: " + e.to_s )
68 if cla.get_number_of_files == 4
69 mapfile = cla.get_file_name( 0 )
70 input = cla.get_file_name( 1 )
71 output = cla.get_file_name( 2 )
72 list_file = cla.get_file_name( 3 )
73 elsif cla.get_number_of_files == 3
74 input = cla.get_file_name( 0 )
75 output = cla.get_file_name( 1 )
76 list_file = cla.get_file_name( 2 )
80 allowed_opts = Array.new
81 allowed_opts.push( EXTRACT_TAXONOMY_OPTION )
83 disallowed = cla.validate_allowed_options_as_str( allowed_opts )
84 if ( disallowed.length > 0 )
85 Util.fatal_error( PRG_NAME, "unknown option(s): " + disallowed )
88 extract_taxonomy = false
89 if ( cla.is_option_set?( EXTRACT_TAXONOMY_OPTION ) )
90 extract_taxonomy = true
93 if ( File.exists?( output ) )
94 Util.fatal_error( PRG_NAME, "outfile [" + output + "] already exists" )
96 if ( File.exists?( list_file ) )
97 Util.fatal_error( PRG_NAME, "list file [" + list_file + "] already exists" )
99 if ( !File.exists?( input) )
100 Util.fatal_error( PRG_NAME, "infile [" + input + "] does not exist" )
102 if ( mapfile != nil && !File.exists?( mapfile ) )
103 Util.fatal_error( PRG_NAME, "mapfile [" + mapfile + "] does not exist" )
106 fasta_like = Util.looks_like_fasta?( input )
110 puts( "Map file : " + mapfile )
112 puts( "Input alignment : " + input )
113 puts( "Output alignment: " + output )
114 puts( "Name list : " + list_file )
116 puts( "Format : Fasta" )
118 puts( "Format : Phylip like" )
120 if ( extract_taxonomy )
121 puts( "Extract taxonomy: true" )
125 species_map = Hash.new
127 File.open( mapfile ) do | file |
128 while line = file.gets
129 if ( line =~/(.+)#(.+)/ || line =~/(.+)\s+(.+)/ )
130 species_map[ $1 ] = $2
131 Util.print_message( PRG_NAME, "mapping: " + $1 + ' => ' + $2 )
140 msa = f.create_msa_from_file( input, FastaParser.new() )
142 msa = f.create_msa_from_file( input, GeneralMsaParser.new() )
144 rescue Exception => e
145 Util.fatal_error( PRG_NAME, "failed to read file: " + e.to_s )
148 if ( msa == nil || msa.get_number_of_seqs() < 1 )
149 Util.fatal_error( PRG_NAME, "failed to read MSA" )
152 Util.check_file_for_writability( list_file )
153 rescue Exception => e
154 Util.fatal_error( PRG_NAME, "error: " + e.to_, STDOUT )
157 #removed = msa.remove_redundant_sequences!( true )
159 # Util.print_message( PRG_NAME, "going to ignore the following " + removed.size.to_s + " redundant sequences:" )
160 # removed.each { | seq_name |
163 # Util.print_message( PRG_NAME, "will process " + msa.get_number_of_seqs.to_s + " non redundant sequences" )
166 lf = File.open( list_file, "a" )
167 for i in 0 ... msa.get_number_of_seqs
168 seq = msa.get_sequence( i )
169 seq.set_name( Util::normalize_seq_name( modify_name( seq.get_name(), i, lf, species_map, extract_taxonomy ), 10 ) )
175 w = FastaWriter.new()
177 w = PhylipSequentialWriter.new()
179 w.set_max_name_length( 10 )
182 io.write_to_file( msa, output, w )
183 rescue Exception => e
184 Util.fatal_error( PRG_NAME, "failed to write file: " + e.to_s )
187 if ( @taxonomies.length > 0 )
188 Util.print_message( PRG_NAME, "number of unique taxonomies: " + @taxonomies.length.to_s )
190 Util.print_message( PRG_NAME, "wrote: " + list_file )
191 Util.print_message( PRG_NAME, "wrote: " + output )
192 Util.print_message( PRG_NAME, "OK" )
197 def modify_name( desc, counter, file, species_map, extract_taxonomy )
200 if desc =~ /^>?\s*\S{1,10}_([0-9A-Z]{3,5})/
201 new_desc = counter.to_s( 16 ) + "_" + $1
202 elsif extract_taxonomy
203 if ( desc.count( "[" ) != desc.count( "]" ) )
204 Util.fatal_error( PRG_NAME, "illegal bracket count in: " + desc )
207 species_map.each_key do | key |
208 if desc =~ /[\b|_]#{key}\b/ # Added boundaries to prevent e.g. RAT matching ARATH.
209 species = species_map[ key ]
210 new_desc = counter.to_s( 16 ) + "_" + species
215 if desc =~/.*\[(\S{3,}?)\]/
219 species.gsub!( /\s+/, " " )
220 species.gsub!( /-/, "" )
221 species.gsub!( /\)/, "" )
222 species.gsub!( /\(/, "" )
223 species.gsub!( /\'/, "" )
224 if species =~ /\S+\s\S+/ || species =~ /\S{3,5}/
225 if species =~ /(\S+)\s(\S+)/
226 code = $1[ 0..2 ] + $2[ 0..1 ]
227 elsif species =~ /\S{3,5}/
229 elsif species.count( " " ) > 2
230 species =~ /(\S+)\s+(\S+)\s+(\S+)$/
234 code = code[ 0 ] + third_last[ 0 ] + second_last[ 0 ] + last[ 0 ] + last[ last.size - 1 ]
235 elsif species.count( " " ) > 1
236 species =~ /(\S+)\s+(\S+)$/
239 code = code[ 0..1 ] + second_last[ 0 ] + last[ 0 ] + last[ last.size - 1 ]
241 new_desc = counter.to_s( 16 ) + "_" + code
242 if @taxonomies.has_key?( code )
243 if ( !@taxonomies.has_value?( species ) )
244 Util.fatal_error( PRG_NAME, "code [#{code}] is not unique in [#{desc}]" )
247 if ( @taxonomies.has_value?( species ) )
248 Util.fatal_error( PRG_NAME, "genome [#{species}] is not unique in [#{desc}]" )
250 @taxonomies[ code ] = species
254 Util.fatal_error( PRG_NAME, "illegal format [#{species}] in: " + desc )
257 Util.fatal_error( PRG_NAME, "illegal format in: " + desc )
263 species_map.each_key do | key |
265 species = species_map[ key ]
266 species = species.gsub( /\s+/, "" )
267 species = species.gsub( /_/, " " )
269 if species =~ /(\S+)\s+(\S+)/
270 species = $1[0..2] + $2[0..1]
272 species = species.gsub( /\s+/, "" )
273 species = species.slice(0, 5)
279 Util.fatal_error( PRG_NAME, "species not found in: " + desc )
281 new_desc = counter.to_s( 16 ) + "_" + species
285 Util.fatal_error( PRG_NAME, "failed to extract species from: " + desc )
288 file.print( new_desc + ": " + desc + " [" + my_species + "]" + "\n" )
290 file.print( new_desc + ": " + desc + "\n" )
295 end # class TaxonomyProcessor