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 )
80 disallowed = cla.validate_allowed_options_as_str( allowed_opts )
81 if ( disallowed.length > 0 )
82 Util.fatal_error( PRG_NAME,
83 "unknown option(s): " + disallowed,
87 domain_id = cla.get_file_name( 0 )
88 hmmsearch_output = cla.get_file_name( 1 )
89 fasta_sequence_file = cla.get_file_name( 2 )
90 outfile = cla.get_file_name( 3 )
93 if ( cla.is_option_set?( ADD_POSITION_OPTION ) )
98 if ( cla.is_option_set?( TRIM_OPTION ) )
102 add_domain_number = false
103 add_domain_number_as_letter = false
104 add_domain_number_as_digit = false
106 if ( cla.is_option_set?( ADD_DOMAIN_NUMBER_OPTION ) )
107 add_domain_number = true
109 if ( cla.is_option_set?( ADD_DOMAIN_NUMBER_OPTION_AS_LETTER ) )
110 add_domain_number_as_letter = true
112 if ( cla.is_option_set?( ADD_DOMAIN_NUMBER_OPTION_AS_DIGIT ) )
113 add_domain_number_as_digit = true
117 if cla.is_option_set? ADD_SPECIES
121 if ( add_domain_number_as_letter && add_domain_number_as_digit )
122 puts( "attempt to add domain number as letter and digit at the same time" )
127 e_value_threshold = -1.0
128 if ( cla.is_option_set?( E_VALUE_THRESHOLD_OPTION ) )
130 e_value_threshold = cla.get_option_value_as_float( E_VALUE_THRESHOLD_OPTION )
131 rescue ArgumentError => e
132 Forester::Util.fatal_error( PRG_NAME, "error: " + e.to_s, STDOUT )
134 if ( e_value_threshold < 0.0 )
135 Forester::Util.fatal_error( PRG_NAME, "attempt to use a negative E-value threshold", STDOUT )
139 length_threshold = -1
140 if ( cla.is_option_set?( LENGTH_THRESHOLD_OPTION ) )
142 length_threshold = cla.get_option_value_as_int( LENGTH_THRESHOLD_OPTION )
143 rescue ArgumentError => e
144 Forester::Util.fatal_error( PRG_NAME, "error: " + e.to_s, STDOUT )
146 if ( length_threshold < 0)
147 Forester::Util.fatal_error( PRG_NAME, "attempt to use a negative length threshold", STDOUT )
154 puts( "Domain : " + domain_id )
155 log << "Domain : " + domain_id + ld
156 puts( "Hmmscan outputfile : " + hmmsearch_output )
157 log << "Hmmscan outputfile : " + hmmsearch_output + ld
158 puts( "Fasta sequencefile (complete sequences): " + fasta_sequence_file )
159 log << "Fasta sequencefile (complete sequences): " + fasta_sequence_file + ld
160 puts( "Outputfile : " + outfile )
161 log << "Outputfile : " + outfile + ld
162 puts( "Passed sequences outfile (fasta) : " + outfile + PASSED_SEQS_SUFFIX )
163 log << "Passed sequences outfile (fasta) : " + outfile + PASSED_SEQS_SUFFIX + ld
164 puts( "Failed sequences outfile (fasta) : " + outfile + FAILED_SEQS_SUFFIX )
165 log << "Failed sequences outfile (fasta) : " + outfile + FAILED_SEQS_SUFFIX + ld
166 puts( "Logfile : " + outfile + LOG_FILE_SUFFIX )
167 log << "Logfile : " + outfile + LOG_FILE_SUFFIX + ld
168 if ( e_value_threshold >= 0.0 )
169 puts( "E-value threshold : " + e_value_threshold.to_s )
170 log << "E-value threshold : " + e_value_threshold.to_s + ld
172 puts( "E-value threshold : no threshold" )
173 log << "E-value threshold : no threshold" + ld
175 if ( length_threshold > 0 )
176 puts( "Length threshold : " + length_threshold.to_s )
177 log << "Length threshold : " + length_threshold.to_s + ld
179 puts( "Length threshold : no threshold" )
180 log << "Length threshold : no threshold" + ld
184 puts( "Trim last 2 chars : true" )
185 log << "Trim last 2 chars : true" + ld
187 puts( "Trim names : false" )
188 log << "Trim names : false" + ld
193 puts( "Add positions (rel to complete seq) to extracted domains: true" )
194 log << "Add positions (rel to complete seq) to extracted domains: true" + ld
196 puts( "Add positions (rel to complete seq) to extracted domains: false" )
197 log << "Add positions (rel to complete seq) to extracted domains: false" + ld
200 if ( add_domain_number || add_domain_number_as_digit || add_domain_number_as_letter )
201 puts( "Add numbers to extracted domains (in case of more than one domain per complete seq): true" )
202 log << "Add numbers to extracted domains (in case of more than one domain per complete seq): true" + ld
204 puts( "Add numbers to extracted domains (in case of more than one domain per complete seq): false" )
205 log << "Add numbers to extracted domains (in case of more than one domain per complete seq): false" + ld
212 parser = HmmscanDomainExtractor.new()
213 domain_count = parser.parse( domain_id,
217 outfile + PASSED_SEQS_SUFFIX,
218 outfile + FAILED_SEQS_SUFFIX,
223 add_domain_number_as_digit,
224 add_domain_number_as_letter,
228 rescue ArgumentError, IOError, StandardError => e
229 Util.fatal_error( PRG_NAME, "error: " + e.to_s, STDOUT )
230 rescue Exception => e
231 Util.fatal_error( PRG_NAME, "unexpected exception!: " + e.to_s, STDOUT )
235 Util.print_message( PRG_NAME, "extracted a total of " + domain_count.to_s + " domains" )
236 Util.print_message( PRG_NAME, "wrote; " + outfile )
237 Util.print_message( PRG_NAME, "wrote: " + outfile + LOG_FILE_SUFFIX )
238 Util.print_message( PRG_NAME, "(wrote: " + outfile + PASSED_SEQS_SUFFIX + ")" )
239 Util.print_message( PRG_NAME, "(wrote: " + outfile + FAILED_SEQS_SUFFIX + ")" )
242 f = File.open( outfile + LOG_FILE_SUFFIX, 'a' )
245 rescue Exception => e
246 Util.fatal_error( PRG_NAME, "error: " + e.to_s )
250 Util.print_message( PRG_NAME, "OK" )
259 puts( " " + PRG_NAME + ".rb [options] <domain> <hmmscan outputfile> <file containing complete sequences in fasta format> <outputfile>" )
261 puts( " options: -" + E_VALUE_THRESHOLD_OPTION + "=<f>: E-value threshold, default is no threshold" )
262 puts( " -" + LENGTH_THRESHOLD_OPTION + "=<i>: length threshold, default is no threshold" )
263 puts( " -" + ADD_POSITION_OPTION + ": to add positions (rel to complete seq) to extracted domains" )
264 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\")" )
265 puts( " -" + ADD_DOMAIN_NUMBER_OPTION_AS_DIGIT + ": to add numbers to extracted domains as digit (example \"domain2\")" )
266 puts( " -" + ADD_DOMAIN_NUMBER_OPTION_AS_LETTER + ": to add numbers to extracted domains as letter (example \"domaina\")" )
267 puts( " -" + TRIM_OPTION + ": to remove the last 2 characters from sequence names" )
268 puts( " -" + ADD_SPECIES + ": to add species [in brackets]" )
272 end # class DomainSequenceExtractor