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'
34 LOG_FILE_SUFFIX = '_domain_seq_extr.log'
35 PASSED_SEQS_SUFFIX = '_domain_seq_extr_passed'
36 FAILED_SEQS_SUFFIX = '_domain_seq_extr_failed'
37 HELP_OPTION_1 = 'help'
42 Util.print_program_information( PRG_NAME,
51 ld = Constants::LINE_DELIMITER
54 cla = CommandLineArguments.new( ARGV )
56 Util.fatal_error( PRG_NAME, "error: " + $!, STDOUT )
59 if ( cla.is_option_set?( HELP_OPTION_1 ) ||
60 cla.is_option_set?( HELP_OPTION_2 ) )
65 if ( cla.get_number_of_files != 4 )
70 allowed_opts = Array.new
71 allowed_opts.push( E_VALUE_THRESHOLD_OPTION )
72 allowed_opts.push( ADD_POSITION_OPTION )
73 allowed_opts.push( ADD_DOMAIN_NUMBER_OPTION )
74 allowed_opts.push( LENGTH_THRESHOLD_OPTION )
75 allowed_opts.push( ADD_DOMAIN_NUMBER_OPTION_AS_DIGIT )
76 allowed_opts.push( ADD_DOMAIN_NUMBER_OPTION_AS_LETTER )
77 allowed_opts.push( TRIM_OPTION )
79 disallowed = cla.validate_allowed_options_as_str( allowed_opts )
80 if ( disallowed.length > 0 )
81 Util.fatal_error( PRG_NAME,
82 "unknown option(s): " + disallowed,
86 domain_id = cla.get_file_name( 0 )
87 hmmsearch_output = cla.get_file_name( 1 )
88 fasta_sequence_file = cla.get_file_name( 2 )
89 outfile = cla.get_file_name( 3 )
92 if ( cla.is_option_set?( ADD_POSITION_OPTION ) )
97 if ( cla.is_option_set?( TRIM_OPTION ) )
101 add_domain_number = false
102 add_domain_number_as_letter = false
103 add_domain_number_as_digit = false
105 if ( cla.is_option_set?( ADD_DOMAIN_NUMBER_OPTION ) )
106 add_domain_number = true
108 if ( cla.is_option_set?( ADD_DOMAIN_NUMBER_OPTION_AS_LETTER ) )
109 add_domain_number_as_letter = true
111 if ( cla.is_option_set?( ADD_DOMAIN_NUMBER_OPTION_AS_DIGIT ) )
112 add_domain_number_as_digit = true
115 if ( add_domain_number_as_letter && add_domain_number_as_digit )
116 puts( "attempt to add domain number as letter and digit at the same time" )
121 e_value_threshold = -1.0
122 if ( cla.is_option_set?( E_VALUE_THRESHOLD_OPTION ) )
124 e_value_threshold = cla.get_option_value_as_float( E_VALUE_THRESHOLD_OPTION )
125 rescue ArgumentError => e
126 Forester::Util.fatal_error( PRG_NAME, "error: " + e.to_s, STDOUT )
128 if ( e_value_threshold < 0.0 )
129 Forester::Util.fatal_error( PRG_NAME, "attempt to use a negative E-value threshold", STDOUT )
133 length_threshold = -1
134 if ( cla.is_option_set?( LENGTH_THRESHOLD_OPTION ) )
136 length_threshold = cla.get_option_value_as_int( LENGTH_THRESHOLD_OPTION )
137 rescue ArgumentError => e
138 Forester::Util.fatal_error( PRG_NAME, "error: " + e.to_s, STDOUT )
140 if ( length_threshold < 0)
141 Forester::Util.fatal_error( PRG_NAME, "attempt to use a negative length threshold", STDOUT )
148 puts( "Domain : " + domain_id )
149 log << "Domain : " + domain_id + ld
150 puts( "Hmmscan outputfile : " + hmmsearch_output )
151 log << "Hmmscan outputfile : " + hmmsearch_output + ld
152 puts( "Fasta sequencefile (complete sequences): " + fasta_sequence_file )
153 log << "Fasta sequencefile (complete sequences): " + fasta_sequence_file + ld
154 puts( "Outputfile : " + outfile )
155 log << "Outputfile : " + outfile + ld
156 puts( "Passed sequences outfile (fasta) : " + outfile + PASSED_SEQS_SUFFIX )
157 log << "Passed sequences outfile (fasta) : " + outfile + PASSED_SEQS_SUFFIX + ld
158 puts( "Failed sequences outfile (fasta) : " + outfile + FAILED_SEQS_SUFFIX )
159 log << "Failed sequences outfile (fasta) : " + outfile + FAILED_SEQS_SUFFIX + ld
160 puts( "Logfile : " + outfile + LOG_FILE_SUFFIX )
161 log << "Logfile : " + outfile + LOG_FILE_SUFFIX + ld
162 if ( e_value_threshold >= 0.0 )
163 puts( "E-value threshold : " + e_value_threshold.to_s )
164 log << "E-value threshold : " + e_value_threshold.to_s + ld
166 puts( "E-value threshold : no threshold" )
167 log << "E-value threshold : no threshold" + ld
169 if ( length_threshold > 0 )
170 puts( "Length threshold : " + length_threshold.to_s )
171 log << "Length threshold : " + length_threshold.to_s + ld
173 puts( "Length threshold : no threshold" )
174 log << "Length threshold : no threshold" + ld
178 puts( "Trim last 2 chars : true" )
179 log << "Trim last 2 chars : true" + ld
181 puts( "Trim names : false" )
182 log << "Trim names : false" + ld
187 puts( "Add positions (rel to complete seq) to extracted domains: true" )
188 log << "Add positions (rel to complete seq) to extracted domains: true" + ld
190 puts( "Add positions (rel to complete seq) to extracted domains: false" )
191 log << "Add positions (rel to complete seq) to extracted domains: false" + ld
194 if ( add_domain_number || add_domain_number_as_digit || add_domain_number_as_letter )
195 puts( "Add numbers to extracted domains (in case of more than one domain per complete seq): true" )
196 log << "Add numbers to extracted domains (in case of more than one domain per complete seq): true" + ld
198 puts( "Add numbers to extracted domains (in case of more than one domain per complete seq): false" )
199 log << "Add numbers to extracted domains (in case of more than one domain per complete seq): false" + ld
206 parser = HmmscanDomainExtractor.new()
207 domain_count = parser.parse( domain_id,
211 outfile + PASSED_SEQS_SUFFIX,
212 outfile + FAILED_SEQS_SUFFIX,
217 add_domain_number_as_digit,
218 add_domain_number_as_letter,
221 rescue ArgumentError, IOError, StandardError => e
222 Util.fatal_error( PRG_NAME, "error: " + e.to_s, STDOUT )
223 rescue Exception => e
224 Util.fatal_error( PRG_NAME, "unexpected exception!: " + e.to_s, STDOUT )
228 Util.print_message( PRG_NAME, "extracted a total of " + domain_count.to_s + " domains" )
229 Util.print_message( PRG_NAME, "wrote; " + outfile )
230 Util.print_message( PRG_NAME, "wrote: " + outfile + LOG_FILE_SUFFIX )
231 Util.print_message( PRG_NAME, "(wrote: " + outfile + PASSED_SEQS_SUFFIX + ")" )
232 Util.print_message( PRG_NAME, "(wrote: " + outfile + FAILED_SEQS_SUFFIX + ")" )
235 f = File.open( outfile + LOG_FILE_SUFFIX, 'a' )
238 rescue Exception => e
239 Util.fatal_error( PRG_NAME, "error: " + e.to_s )
243 Util.print_message( PRG_NAME, "OK" )
252 puts( " " + PRG_NAME + ".rb [options] <domain> <hmmscan outputfile> <file containing complete sequences in fasta format> <outputfile>" )
254 puts( " options: -" + E_VALUE_THRESHOLD_OPTION + "=<f>: E-value threshold, default is no threshold" )
255 puts( " -" + LENGTH_THRESHOLD_OPTION + "=<i>: length threshold, default is no threshold" )
256 puts( " -" + ADD_POSITION_OPTION + ": to add positions (rel to complete seq) to extracted domains" )
257 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\")" )
258 puts( " -" + ADD_DOMAIN_NUMBER_OPTION_AS_DIGIT + ": to add numbers to extracted domains as digit (example \"domain2\")" )
259 puts( " -" + ADD_DOMAIN_NUMBER_OPTION_AS_LETTER + ": to add numbers to extracted domains as letter (example \"domaina\")" )
260 puts( " -" + TRIM_OPTION + ": to remove the last 2 characters from sequence names" )
264 end # class DomainSequenceExtractor