2 # = lib/evo/apps/domain_sequence_extractor.rb - DomainSequenceExtractor class
4 # Copyright:: Copyright (C) 2012 Christian M. Zmasek
5 # License:: GNU Lesser General Public License (LGPL)
10 require 'lib/evo/util/constants'
11 require 'lib/evo/util/util'
12 require 'lib/evo/util/command_line_arguments'
13 require 'lib/evo/io/parser/hmmscan_domain_extractor'
17 class DomainSequenceExtractor
21 PRG_DESC = "extraction of domain sequences from hmmscan output"
23 COPYRIGHT = "2012 Christian M Zmasek"
24 CONTACT = "phylosoft@gmail.com"
25 WWW = "www.phylosoft.org"
27 E_VALUE_THRESHOLD_OPTION = 'e'
28 LENGTH_THRESHOLD_OPTION = 'l'
29 ADD_POSITION_OPTION = 'p'
30 ADD_DOMAIN_NUMBER_OPTION = 'd'
31 ADD_DOMAIN_NUMBER_OPTION_AS_DIGIT = 'dd'
32 ADD_DOMAIN_NUMBER_OPTION_AS_LETTER = 'dl'
35 LOG_FILE_SUFFIX = '_domain_seq_extr.log'
36 PASSED_SEQS_SUFFIX = '_domain_seq_extr_passed'
37 FAILED_SEQS_SUFFIX = '_domain_seq_extr_failed'
38 HELP_OPTION_1 = 'help'
43 Util.print_program_information( PRG_NAME,
52 ld = Constants::LINE_DELIMITER
55 cla = CommandLineArguments.new( ARGV )
57 Util.fatal_error( PRG_NAME, "error: " + $!, STDOUT )
60 if ( cla.is_option_set?( HELP_OPTION_1 ) ||
61 cla.is_option_set?( HELP_OPTION_2 ) )
66 if ( cla.get_number_of_files != 4 )
71 allowed_opts = Array.new
72 allowed_opts.push( E_VALUE_THRESHOLD_OPTION )
73 allowed_opts.push( ADD_POSITION_OPTION )
74 allowed_opts.push( ADD_DOMAIN_NUMBER_OPTION )
75 allowed_opts.push( LENGTH_THRESHOLD_OPTION )
76 allowed_opts.push( ADD_DOMAIN_NUMBER_OPTION_AS_DIGIT )
77 allowed_opts.push( ADD_DOMAIN_NUMBER_OPTION_AS_LETTER )
78 allowed_opts.push( TRIM_OPTION )
79 allowed_opts.push( ADD_SPECIES )
81 disallowed = cla.validate_allowed_options_as_str( allowed_opts )
82 if ( disallowed.length > 0 )
83 Util.fatal_error( PRG_NAME,
84 "unknown option(s): " + disallowed,
88 domain_id = cla.get_file_name( 0 )
89 hmmsearch_output = cla.get_file_name( 1 )
90 fasta_sequence_file = cla.get_file_name( 2 )
91 outfile = cla.get_file_name( 3 )
94 if ( cla.is_option_set?( ADD_POSITION_OPTION ) )
99 if ( cla.is_option_set?( TRIM_OPTION ) )
103 add_domain_number = false
104 add_domain_number_as_letter = false
105 add_domain_number_as_digit = false
107 if ( cla.is_option_set?( ADD_DOMAIN_NUMBER_OPTION ) )
108 add_domain_number = true
110 if ( cla.is_option_set?( ADD_DOMAIN_NUMBER_OPTION_AS_LETTER ) )
111 add_domain_number_as_letter = true
113 if ( cla.is_option_set?( ADD_DOMAIN_NUMBER_OPTION_AS_DIGIT ) )
114 add_domain_number_as_digit = true
118 if cla.is_option_set? ADD_SPECIES
122 if ( add_domain_number_as_letter && add_domain_number_as_digit )
123 puts( "attempt to add domain number as letter and digit at the same time" )
128 e_value_threshold = -1.0
129 if ( cla.is_option_set?( E_VALUE_THRESHOLD_OPTION ) )
131 e_value_threshold = cla.get_option_value_as_float( E_VALUE_THRESHOLD_OPTION )
132 rescue ArgumentError => e
133 Forester::Util.fatal_error( PRG_NAME, "error: " + e.to_s, STDOUT )
135 if ( e_value_threshold < 0.0 )
136 Forester::Util.fatal_error( PRG_NAME, "attempt to use a negative E-value threshold", STDOUT )
140 length_threshold = -1
141 if ( cla.is_option_set?( LENGTH_THRESHOLD_OPTION ) )
143 length_threshold = cla.get_option_value_as_int( LENGTH_THRESHOLD_OPTION )
144 rescue ArgumentError => e
145 Forester::Util.fatal_error( PRG_NAME, "error: " + e.to_s, STDOUT )
147 if ( length_threshold < 0)
148 Forester::Util.fatal_error( PRG_NAME, "attempt to use a negative length threshold", STDOUT )
155 puts( "Domain : " + domain_id )
156 log << "Domain : " + domain_id + ld
157 puts( "Hmmscan outputfile : " + hmmsearch_output )
158 log << "Hmmscan outputfile : " + hmmsearch_output + ld
159 puts( "Fasta sequencefile (complete sequences): " + fasta_sequence_file )
160 log << "Fasta sequencefile (complete sequences): " + fasta_sequence_file + ld
161 puts( "Outputfile : " + outfile )
162 log << "Outputfile : " + outfile + ld
163 puts( "Passed sequences outfile (fasta) : " + outfile + PASSED_SEQS_SUFFIX )
164 log << "Passed sequences outfile (fasta) : " + outfile + PASSED_SEQS_SUFFIX + ld
165 puts( "Failed sequences outfile (fasta) : " + outfile + FAILED_SEQS_SUFFIX )
166 log << "Failed sequences outfile (fasta) : " + outfile + FAILED_SEQS_SUFFIX + ld
167 puts( "Logfile : " + outfile + LOG_FILE_SUFFIX )
168 log << "Logfile : " + outfile + LOG_FILE_SUFFIX + ld
169 if ( e_value_threshold >= 0.0 )
170 puts( "E-value threshold : " + e_value_threshold.to_s )
171 log << "E-value threshold : " + e_value_threshold.to_s + ld
173 puts( "E-value threshold : no threshold" )
174 log << "E-value threshold : no threshold" + ld
176 if ( length_threshold > 0 )
177 puts( "Length threshold : " + length_threshold.to_s )
178 log << "Length threshold : " + length_threshold.to_s + ld
180 puts( "Length threshold : no threshold" )
181 log << "Length threshold : no threshold" + ld
185 puts( "Trim last 2 chars : true" )
186 log << "Trim last 2 chars : true" + ld
188 puts( "Trim names : false" )
189 log << "Trim names : false" + ld
194 puts( "Add positions (rel to complete seq) to extracted domains: true" )
195 log << "Add positions (rel to complete seq) to extracted domains: true" + ld
197 puts( "Add positions (rel to complete seq) to extracted domains: false" )
198 log << "Add positions (rel to complete seq) to extracted domains: false" + ld
201 if ( add_domain_number || add_domain_number_as_digit || add_domain_number_as_letter )
202 puts( "Add numbers to extracted domains (in case of more than one domain per complete seq): true" )
203 log << "Add numbers to extracted domains (in case of more than one domain per complete seq): true" + ld
205 puts( "Add numbers to extracted domains (in case of more than one domain per complete seq): false" )
206 log << "Add numbers to extracted domains (in case of more than one domain per complete seq): false" + ld
213 parser = HmmscanDomainExtractor.new()
214 domain_count = parser.parse( domain_id,
218 outfile + PASSED_SEQS_SUFFIX,
219 outfile + FAILED_SEQS_SUFFIX,
224 add_domain_number_as_digit,
225 add_domain_number_as_letter,
229 rescue ArgumentError, IOError, StandardError => e
230 Util.fatal_error( PRG_NAME, "error: " + e.to_s, STDOUT )
231 rescue Exception => e
232 Util.fatal_error( PRG_NAME, "unexpected exception!: " + e.to_s, STDOUT )
236 Util.print_message( PRG_NAME, "extracted a total of " + domain_count.to_s + " domains" )
237 Util.print_message( PRG_NAME, "wrote; " + outfile )
238 Util.print_message( PRG_NAME, "wrote: " + outfile + LOG_FILE_SUFFIX )
239 Util.print_message( PRG_NAME, "(wrote: " + outfile + PASSED_SEQS_SUFFIX + ")" )
240 Util.print_message( PRG_NAME, "(wrote: " + outfile + FAILED_SEQS_SUFFIX + ")" )
243 f = File.open( outfile + LOG_FILE_SUFFIX, 'a' )
246 rescue Exception => e
247 Util.fatal_error( PRG_NAME, "error: " + e.to_s )
251 Util.print_message( PRG_NAME, "OK" )
260 puts( " " + PRG_NAME + ".rb [options] <domain> <hmmscan outputfile> <file containing complete sequences in fasta format> <outputfile>" )
262 puts( " options: -" + E_VALUE_THRESHOLD_OPTION + "=<f>: E-value threshold, default is no threshold" )
263 puts( " -" + LENGTH_THRESHOLD_OPTION + "=<i>: length threshold, default is no threshold" )
264 puts( " -" + ADD_POSITION_OPTION + ": to add positions (rel to complete seq) to extracted domains" )
265 puts( " -" + ADD_DOMAIN_NUMBER_OPTION + ": to add numbers to extracted domains (in case of more than one domain per complete seq) (example \"domain~2-3\")" )
266 puts( " -" + ADD_DOMAIN_NUMBER_OPTION_AS_DIGIT + ": to add numbers to extracted domains as digit (example \"domain2\")" )
267 puts( " -" + ADD_DOMAIN_NUMBER_OPTION_AS_LETTER + ": to add numbers to extracted domains as letter (example \"domaina\")" )
268 puts( " -" + TRIM_OPTION + ": to remove the last 2 characters from sequence names" )
269 puts( " -" + ADD_SPECIES + ": to add species [in brackets]" )
273 end # class DomainSequenceExtractor