scbi_fasta 0.1.8
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- data/History.txt +22 -0
- data/Manifest.txt +16 -0
- data/PostInstall.txt +7 -0
- data/README.rdoc +93 -0
- data/Rakefile +26 -0
- data/lib/scbi_fasta.rb +12 -0
- data/lib/scbi_fasta/fasta_file.rb +92 -0
- data/lib/scbi_fasta/fasta_qual_file.rb +288 -0
- data/script/console +10 -0
- data/script/destroy +14 -0
- data/script/generate +14 -0
- data/test/test_fasta_file.rb +65 -0
- data/test/test_fasta_qual_file.rb +161 -0
- data/test/test_helper.rb +4 -0
- metadata +83 -0
data/History.txt
ADDED
@@ -0,0 +1,22 @@
|
|
1
|
+
=== 0.1.8 2011-06-13
|
2
|
+
|
3
|
+
first rubygems published
|
4
|
+
|
5
|
+
=== 0.1.3 2010-09-03
|
6
|
+
|
7
|
+
* 1 minor fix, 3 enhancements:
|
8
|
+
* remove comments from sequence names
|
9
|
+
* FastaQualFile can be used without qual file
|
10
|
+
* added FastaFile
|
11
|
+
* added tests
|
12
|
+
|
13
|
+
=== 0.0.2 2010-05-14
|
14
|
+
|
15
|
+
* 1 improvement:
|
16
|
+
* speed improvement in FastaQualFile
|
17
|
+
|
18
|
+
=== 0.0.1 2010-05-14
|
19
|
+
|
20
|
+
* 1 major enhancement:
|
21
|
+
* Initial release of FastaQualFile
|
22
|
+
|
data/Manifest.txt
ADDED
@@ -0,0 +1,16 @@
|
|
1
|
+
History.txt
|
2
|
+
Manifest.txt
|
3
|
+
PostInstall.txt
|
4
|
+
README.rdoc
|
5
|
+
Rakefile
|
6
|
+
lib/scbi_fasta.rb
|
7
|
+
lib/scbi_fasta/fasta_qual_file.rb
|
8
|
+
lib/scbi_fasta/fasta_file.rb
|
9
|
+
script/console
|
10
|
+
script/destroy
|
11
|
+
script/generate
|
12
|
+
test/test_helper.rb
|
13
|
+
test/test_fasta_file.rb
|
14
|
+
test/test_fasta_qual_file.rb
|
15
|
+
|
16
|
+
|
data/PostInstall.txt
ADDED
data/README.rdoc
ADDED
@@ -0,0 +1,93 @@
|
|
1
|
+
= scbi_fasta
|
2
|
+
|
3
|
+
* http://www.scbi.uma.es/downloads
|
4
|
+
|
5
|
+
== DESCRIPTION:
|
6
|
+
|
7
|
+
scbi_fasta is a ruby gem to read FASTA+QUAL files (DNA/RNA sequences).
|
8
|
+
|
9
|
+
|
10
|
+
== FEATURES/PROBLEMS:
|
11
|
+
|
12
|
+
* Read FASTA files with associated QUAL files
|
13
|
+
* Quality values can be automatically splitted
|
14
|
+
* Iteration over large files without extra memory usage
|
15
|
+
|
16
|
+
|
17
|
+
== SYNOPSIS:
|
18
|
+
|
19
|
+
=== Reading a FASTA+QUAL with iterator:
|
20
|
+
|
21
|
+
require 'scbi_fasta'
|
22
|
+
|
23
|
+
# open file in sanger mode
|
24
|
+
fqr=FastaQualFile.new('file1.fasta','file1.fasta.qual')
|
25
|
+
|
26
|
+
|
27
|
+
fqr.each do |name,seq_fasta,qual,comments|
|
28
|
+
|
29
|
+
puts name
|
30
|
+
puts seq_fasta
|
31
|
+
puts qual
|
32
|
+
puts comments
|
33
|
+
end
|
34
|
+
|
35
|
+
fqr.close
|
36
|
+
|
37
|
+
=== Reading a FASTA one sequence at a time:
|
38
|
+
|
39
|
+
require 'scbi_fasta'
|
40
|
+
|
41
|
+
# open file
|
42
|
+
fqr=FastaQualFile.new('file1.fasta','file1.fasta.qual')
|
43
|
+
|
44
|
+
|
45
|
+
begin
|
46
|
+
|
47
|
+
# read one sequence
|
48
|
+
name,seq_fasta,qual,comments=fqr.next_seq
|
49
|
+
|
50
|
+
# name will be nil if there are not more sequences available
|
51
|
+
if !name.nil?
|
52
|
+
puts name
|
53
|
+
puts seq_fasta
|
54
|
+
puts qual
|
55
|
+
puts comments
|
56
|
+
end
|
57
|
+
end until name.nil?
|
58
|
+
|
59
|
+
fqr.close
|
60
|
+
|
61
|
+
|
62
|
+
== REQUIREMENTS:
|
63
|
+
|
64
|
+
* This is a standalone gem.
|
65
|
+
|
66
|
+
== INSTALL:
|
67
|
+
|
68
|
+
* gem install scbi_fasta
|
69
|
+
|
70
|
+
== LICENSE:
|
71
|
+
|
72
|
+
(The MIT License)
|
73
|
+
|
74
|
+
Copyright (c) 2010 Almudena Bocinos, Dario Guerrero
|
75
|
+
|
76
|
+
Permission is hereby granted, free of charge, to any person obtaining
|
77
|
+
a copy of this software and associated documentation files (the
|
78
|
+
'Software'), to deal in the Software without restriction, including
|
79
|
+
without limitation the rights to use, copy, modify, merge, publish,
|
80
|
+
distribute, sublicense, and/or sell copies of the Software, and to
|
81
|
+
permit persons to whom the Software is furnished to do so, subject to
|
82
|
+
the following conditions:
|
83
|
+
|
84
|
+
The above copyright notice and this permission notice shall be
|
85
|
+
included in all copies or substantial portions of the Software.
|
86
|
+
|
87
|
+
THE SOFTWARE IS PROVIDED 'AS IS', WITHOUT WARRANTY OF ANY KIND,
|
88
|
+
EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
|
89
|
+
MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
|
90
|
+
IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
|
91
|
+
CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
|
92
|
+
TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
|
93
|
+
SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
|
data/Rakefile
ADDED
@@ -0,0 +1,26 @@
|
|
1
|
+
require 'rubygems'
|
2
|
+
gem 'hoe', '>= 2.1.0'
|
3
|
+
require 'hoe'
|
4
|
+
require 'fileutils'
|
5
|
+
require './lib/scbi_fasta'
|
6
|
+
|
7
|
+
Hoe.plugin :newgem
|
8
|
+
# Hoe.plugin :website
|
9
|
+
# Hoe.plugin :cucumberfeatures
|
10
|
+
|
11
|
+
# Generate all the Rake tasks
|
12
|
+
# Run 'rake -T' to see list of generated tasks (from gem root directory)
|
13
|
+
$hoe = Hoe.spec 'scbi_fasta' do
|
14
|
+
self.developer 'Almudena Bocinos', 'alkoke@gmail.com'
|
15
|
+
self.post_install_message = 'PostInstall.txt' # TODO remove if post-install message not required
|
16
|
+
self.rubyforge_name = self.name # TODO this is default value
|
17
|
+
# self.extra_deps = [['activesupport','>= 2.0.2']]
|
18
|
+
|
19
|
+
end
|
20
|
+
|
21
|
+
require 'newgem/tasks'
|
22
|
+
Dir['tasks/**/*.rake'].each { |t| load t }
|
23
|
+
|
24
|
+
# TODO - want other tests/tasks run by default? Add them to the list
|
25
|
+
# remove_task :default
|
26
|
+
task :default => [:spec, :features, :redocs]
|
data/lib/scbi_fasta.rb
ADDED
@@ -0,0 +1,12 @@
|
|
1
|
+
$:.unshift(File.dirname(__FILE__)) unless
|
2
|
+
$:.include?(File.dirname(__FILE__)) || $:.include?(File.expand_path(File.dirname(__FILE__)))
|
3
|
+
|
4
|
+
$: << File.join(File.dirname(__FILE__),File.basename(__FILE__,File.extname(__FILE__)))
|
5
|
+
|
6
|
+
require 'fasta_qual_file'
|
7
|
+
require 'fasta_file'
|
8
|
+
|
9
|
+
|
10
|
+
module ScbiFasta
|
11
|
+
VERSION = '0.1.8'
|
12
|
+
end
|
@@ -0,0 +1,92 @@
|
|
1
|
+
########################################################
|
2
|
+
# Author: Almudena Bocinos Rioboo
|
3
|
+
# Use: Deprecated. The class FastaReader has been substituted by FastaQualReader
|
4
|
+
# Reads a fasta file and fires events to process it
|
5
|
+
#
|
6
|
+
########################################################
|
7
|
+
|
8
|
+
class FastaFile
|
9
|
+
|
10
|
+
attr :num_seqs
|
11
|
+
|
12
|
+
|
13
|
+
# Initialize instance
|
14
|
+
def initialize(file_name)
|
15
|
+
if File.exist?(file_name)
|
16
|
+
@num_seqs = 0
|
17
|
+
|
18
|
+
@file_name = file_name
|
19
|
+
@file = File.open(file_name)
|
20
|
+
else
|
21
|
+
raise "File #{file_name} doesn't exists"
|
22
|
+
|
23
|
+
end
|
24
|
+
|
25
|
+
end
|
26
|
+
|
27
|
+
|
28
|
+
# Scans a file, firing events to process content
|
29
|
+
def each
|
30
|
+
|
31
|
+
#init variables
|
32
|
+
seq_name = '';
|
33
|
+
seq_fasta = '';
|
34
|
+
seq_found = false;
|
35
|
+
|
36
|
+
|
37
|
+
# for each line of the file
|
38
|
+
@file.each do |line|
|
39
|
+
|
40
|
+
line.chomp!;
|
41
|
+
# if line is name
|
42
|
+
if line =~ /^>/
|
43
|
+
|
44
|
+
#process previous sequence
|
45
|
+
if seq_found
|
46
|
+
@num_seqs=@num_seqs+1
|
47
|
+
yield(seq_name,seq_fasta);
|
48
|
+
end
|
49
|
+
|
50
|
+
#get only name
|
51
|
+
|
52
|
+
#get only name
|
53
|
+
line.gsub!(/^>\s*/,'');
|
54
|
+
|
55
|
+
line =~ /(^[^\s]+)/
|
56
|
+
|
57
|
+
# remove comments
|
58
|
+
seq_name = $1
|
59
|
+
|
60
|
+
seq_fasta='';
|
61
|
+
seq_found = true;
|
62
|
+
|
63
|
+
else
|
64
|
+
|
65
|
+
line.strip! if !line.empty?
|
66
|
+
|
67
|
+
#add line to fasta of seq
|
68
|
+
seq_fasta+=line;
|
69
|
+
|
70
|
+
seq_fasta.strip! if !seq_fasta.empty?
|
71
|
+
|
72
|
+
end
|
73
|
+
|
74
|
+
end
|
75
|
+
|
76
|
+
# when found EOF, process last sequence
|
77
|
+
if seq_found
|
78
|
+
@num_seqs=@num_seqs+1
|
79
|
+
yield(seq_name,seq_fasta);
|
80
|
+
end
|
81
|
+
end
|
82
|
+
|
83
|
+
|
84
|
+
def close
|
85
|
+
@file.close
|
86
|
+
end
|
87
|
+
|
88
|
+
def with_qual?
|
89
|
+
false
|
90
|
+
end
|
91
|
+
|
92
|
+
end
|
@@ -0,0 +1,288 @@
|
|
1
|
+
|
2
|
+
########################################################
|
3
|
+
# Author: Almudena Bocinos Rioboo
|
4
|
+
#
|
5
|
+
# Reads a fasta file and qual file, and then fires events to process it
|
6
|
+
#
|
7
|
+
#
|
8
|
+
#
|
9
|
+
########################################################
|
10
|
+
|
11
|
+
class DifferentNamesException < RuntimeError
|
12
|
+
end
|
13
|
+
|
14
|
+
class DifferentSizesException < RuntimeError
|
15
|
+
end
|
16
|
+
|
17
|
+
class FastaQualFile
|
18
|
+
|
19
|
+
attr_accessor :num_seqs, :end_fasta
|
20
|
+
|
21
|
+
#------------------------------------
|
22
|
+
# Initialize instance
|
23
|
+
#------------------------------------
|
24
|
+
def initialize(fasta_file_name,qual_file_name='',qual_to_array=false)
|
25
|
+
|
26
|
+
|
27
|
+
if !File.exist?(fasta_file_name)
|
28
|
+
raise "File #{fasta_file_name} doesn't exists"
|
29
|
+
end
|
30
|
+
|
31
|
+
@with_qual = true
|
32
|
+
|
33
|
+
if qual_file_name.nil? or qual_file_name.empty? or !File.exist?(qual_file_name)
|
34
|
+
@with_qual = false
|
35
|
+
#raise "File #{qual_file_name} doesn't exists"
|
36
|
+
end
|
37
|
+
|
38
|
+
|
39
|
+
@num_seqs = 0 ;
|
40
|
+
|
41
|
+
@seq_name = '';
|
42
|
+
@file_fasta = File.open(fasta_file_name) ;
|
43
|
+
@end_fasta=false;
|
44
|
+
|
45
|
+
if @with_qual
|
46
|
+
|
47
|
+
@qual_to_array = qual_to_array
|
48
|
+
|
49
|
+
@seq_qual_name='';
|
50
|
+
@file_qual = File.open(qual_file_name) ;
|
51
|
+
@end_qual=false;
|
52
|
+
end
|
53
|
+
|
54
|
+
end
|
55
|
+
|
56
|
+
def close
|
57
|
+
@file_fasta.close
|
58
|
+
|
59
|
+
@file_qual.close if @with_qual
|
60
|
+
end
|
61
|
+
|
62
|
+
|
63
|
+
#------------------------------------
|
64
|
+
# Scans all sequences
|
65
|
+
#------------------------------------
|
66
|
+
def each
|
67
|
+
|
68
|
+
rewind
|
69
|
+
|
70
|
+
n,f,q=next_seq
|
71
|
+
while (!n.nil?)
|
72
|
+
|
73
|
+
if @with_qual
|
74
|
+
yield(n,f,q)
|
75
|
+
else
|
76
|
+
yield(n,f)
|
77
|
+
end
|
78
|
+
|
79
|
+
n,f,q=next_seq
|
80
|
+
end
|
81
|
+
|
82
|
+
rewind
|
83
|
+
|
84
|
+
end
|
85
|
+
|
86
|
+
|
87
|
+
def rewind
|
88
|
+
|
89
|
+
@num_seqs = 0 ;
|
90
|
+
|
91
|
+
@seq_name = '';
|
92
|
+
@file_fasta.pos=0
|
93
|
+
@end_fasta=false;
|
94
|
+
|
95
|
+
if @with_qual
|
96
|
+
@seq_qual_name='';
|
97
|
+
@file_qual.pos=0
|
98
|
+
@end_qual=false;
|
99
|
+
end
|
100
|
+
|
101
|
+
end
|
102
|
+
|
103
|
+
#------------------------------------
|
104
|
+
# Scans a file, firing events to process content
|
105
|
+
#------------------------------------
|
106
|
+
def next_seq
|
107
|
+
|
108
|
+
#init variables
|
109
|
+
res = nil
|
110
|
+
|
111
|
+
# envia on_process_sequence
|
112
|
+
if ((!@end_fasta) && (!@with_qual or !@end_qual))
|
113
|
+
|
114
|
+
name_f,fasta=read_fasta
|
115
|
+
|
116
|
+
if @with_qual
|
117
|
+
|
118
|
+
name_q,qual=read_qual
|
119
|
+
|
120
|
+
if (name_f!=name_q)
|
121
|
+
raise DifferentNamesException.new, "Sequence(<#{name_f}>) and qual(<#{name_q}>) names differs. Sequence will be ignored."
|
122
|
+
else
|
123
|
+
@num_seqs=@num_seqs+1
|
124
|
+
|
125
|
+
#storage a string of qualities in an array of qualities
|
126
|
+
# a_qual = qual.strip.split(/\s/).map{|e| e.to_i}
|
127
|
+
# if ((!a_qual.nil?) && (!a_qual.empty?) && (fasta.size==a_qual.size))
|
128
|
+
|
129
|
+
if fasta.length == qual.count(' ') + 1
|
130
|
+
if @qual_to_array
|
131
|
+
a_qual = qual.strip.split(/\s/).map{|e| e.to_i}
|
132
|
+
res =[name_f,fasta,a_qual]
|
133
|
+
else
|
134
|
+
res =[name_f,fasta,qual]
|
135
|
+
end
|
136
|
+
|
137
|
+
else #if (!a_qual.empty?)
|
138
|
+
raise DifferentSizesException.new, "Sequence(<#{name_f}>) and qual(<#{name_q}>) sizes differs (#{fasta.length},#{qual.count(' ')} ). Sequence will be ignored."
|
139
|
+
end
|
140
|
+
|
141
|
+
end
|
142
|
+
|
143
|
+
else # without qual
|
144
|
+
res =[name_f,fasta]
|
145
|
+
end
|
146
|
+
end
|
147
|
+
|
148
|
+
return res
|
149
|
+
end
|
150
|
+
|
151
|
+
def with_qual?
|
152
|
+
@with_qual
|
153
|
+
end
|
154
|
+
|
155
|
+
|
156
|
+
private
|
157
|
+
|
158
|
+
#------------------------------------
|
159
|
+
# Read one sequence fasta
|
160
|
+
#------------------------------------
|
161
|
+
|
162
|
+
def read_fasta
|
163
|
+
seq_fasta='';
|
164
|
+
res = nil
|
165
|
+
|
166
|
+
# mientras hay lineas en el fichero
|
167
|
+
while (!@file_fasta.eof)
|
168
|
+
# $LOG.debug "Total fasta #{@seq_fasta} Line fast #{@line_fasta} "
|
169
|
+
|
170
|
+
# lee una linea
|
171
|
+
line_fasta = @file_fasta.readline; line_fasta.chomp!
|
172
|
+
# si llega a fin de ficheor, poner eof
|
173
|
+
@end_fasta=@file_fasta.eof
|
174
|
+
|
175
|
+
# si la linea es una nueva secuencia
|
176
|
+
if ((line_fasta =~ /^>/))
|
177
|
+
|
178
|
+
# puede ocurrir que antes ya se hubiese leido una secuencia, entonces devolverla
|
179
|
+
if !@seq_name.empty?
|
180
|
+
# ya había leido una
|
181
|
+
#puts "leida #{@seq_name} + #{@seq_fasta}"
|
182
|
+
res = [@seq_name,seq_fasta]
|
183
|
+
end
|
184
|
+
|
185
|
+
|
186
|
+
#get only name
|
187
|
+
line_fasta.gsub!(/^>\s*/,'');
|
188
|
+
|
189
|
+
line_fasta =~ /(^[^\s]+)/
|
190
|
+
# remove comments
|
191
|
+
|
192
|
+
@seq_name = $1
|
193
|
+
|
194
|
+
seq_fasta='';
|
195
|
+
|
196
|
+
# si hay algo que devolver, romper bucle
|
197
|
+
if res
|
198
|
+
break
|
199
|
+
end
|
200
|
+
|
201
|
+
else # no es una linea de nombre, añadir al fasta
|
202
|
+
line_fasta.strip! if !line_fasta.empty?
|
203
|
+
#add line to fasta of seq
|
204
|
+
seq_fasta+=line_fasta;
|
205
|
+
seq_fasta.strip! if !seq_fasta.empty?
|
206
|
+
end
|
207
|
+
|
208
|
+
|
209
|
+
end
|
210
|
+
|
211
|
+
|
212
|
+
#si no hay más secuencias hay que devolver la última, CASO EOF
|
213
|
+
if res.nil? and !@seq_name.empty?
|
214
|
+
res= [@seq_name,seq_fasta]
|
215
|
+
end
|
216
|
+
|
217
|
+
return res
|
218
|
+
end
|
219
|
+
|
220
|
+
#----
|
221
|
+
|
222
|
+
#------------------------------------
|
223
|
+
# Read one sequence qual
|
224
|
+
#------------------------------------
|
225
|
+
def read_qual
|
226
|
+
seq_qual='';
|
227
|
+
res = nil
|
228
|
+
|
229
|
+
# mientras hay lineas en el fichero
|
230
|
+
while (!@file_qual.eof)
|
231
|
+
# $LOG.debug "Total fasta #{@seq_qual} Line fast #{@line_qual} "
|
232
|
+
|
233
|
+
# lee una linea
|
234
|
+
line_qual = @file_qual.readline; line_qual.chomp!
|
235
|
+
# si llega a fin de ficheor, poner eof
|
236
|
+
@end_qual=@file_qual.eof
|
237
|
+
|
238
|
+
# si la linea es una nueva secuencia
|
239
|
+
if ((line_qual =~ /^>/))
|
240
|
+
|
241
|
+
# puede ocurrir que antes ya se hubiese leido una secuencia, entonces devolverla
|
242
|
+
if !@seq_qual_name.empty?
|
243
|
+
# ya había leido una
|
244
|
+
#puts "leida #{@seq_name} + #{@seq_qual}"
|
245
|
+
|
246
|
+
seq_qual.gsub!(/\s\s+/,' ')
|
247
|
+
res = [@seq_qual_name,seq_qual]
|
248
|
+
end
|
249
|
+
|
250
|
+
|
251
|
+
#get only name
|
252
|
+
#get only name
|
253
|
+
line_qual.gsub!(/^>\s*/,'');
|
254
|
+
|
255
|
+
line_qual =~ /(^[^\s]+)/
|
256
|
+
# remove comments
|
257
|
+
|
258
|
+
@seq_qual_name = $1
|
259
|
+
seq_qual='';
|
260
|
+
|
261
|
+
# si hay algo que devolver, romper bucle
|
262
|
+
if res
|
263
|
+
break
|
264
|
+
end
|
265
|
+
|
266
|
+
else # no es una linea de nombre, añadir al fasta
|
267
|
+
line_qual.strip! if !line_qual.empty?
|
268
|
+
#add line to qual of seq
|
269
|
+
seq_qual=seq_qual+' '+line_qual
|
270
|
+
seq_qual.strip! if !seq_qual.empty?
|
271
|
+
end
|
272
|
+
|
273
|
+
end
|
274
|
+
|
275
|
+
|
276
|
+
#si no hay más secuencias hay que devolver la última, CASO EOF
|
277
|
+
if res.nil? and !@seq_qual_name.empty?
|
278
|
+
seq_qual.gsub!(/\s\s+/,' ')
|
279
|
+
res= [@seq_qual_name,seq_qual]
|
280
|
+
end
|
281
|
+
|
282
|
+
return res
|
283
|
+
end
|
284
|
+
|
285
|
+
|
286
|
+
|
287
|
+
|
288
|
+
end
|
data/script/console
ADDED
@@ -0,0 +1,10 @@
|
|
1
|
+
#!/usr/bin/env ruby
|
2
|
+
# File: script/console
|
3
|
+
irb = RUBY_PLATFORM =~ /(:?mswin|mingw)/ ? 'irb.bat' : 'irb'
|
4
|
+
|
5
|
+
libs = " -r irb/completion"
|
6
|
+
# Perhaps use a console_lib to store any extra methods I may want available in the cosole
|
7
|
+
# libs << " -r #{File.dirname(__FILE__) + '/../lib/console_lib/console_logger.rb'}"
|
8
|
+
libs << " -r #{File.dirname(__FILE__) + '/../lib/scbi_fasta.rb'}"
|
9
|
+
puts "Loading scbi_fasta gem"
|
10
|
+
exec "#{irb} #{libs} --simple-prompt"
|
data/script/destroy
ADDED
@@ -0,0 +1,14 @@
|
|
1
|
+
#!/usr/bin/env ruby
|
2
|
+
APP_ROOT = File.expand_path(File.join(File.dirname(__FILE__), '..'))
|
3
|
+
|
4
|
+
begin
|
5
|
+
require 'rubigen'
|
6
|
+
rescue LoadError
|
7
|
+
require 'rubygems'
|
8
|
+
require 'rubigen'
|
9
|
+
end
|
10
|
+
require 'rubigen/scripts/destroy'
|
11
|
+
|
12
|
+
ARGV.shift if ['--help', '-h'].include?(ARGV[0])
|
13
|
+
RubiGen::Base.use_component_sources! [:rubygems, :newgem, :newgem_theme, :test_unit]
|
14
|
+
RubiGen::Scripts::Destroy.new.run(ARGV)
|
data/script/generate
ADDED
@@ -0,0 +1,14 @@
|
|
1
|
+
#!/usr/bin/env ruby
|
2
|
+
APP_ROOT = File.expand_path(File.join(File.dirname(__FILE__), '..'))
|
3
|
+
|
4
|
+
begin
|
5
|
+
require 'rubigen'
|
6
|
+
rescue LoadError
|
7
|
+
require 'rubygems'
|
8
|
+
require 'rubigen'
|
9
|
+
end
|
10
|
+
require 'rubigen/scripts/generate'
|
11
|
+
|
12
|
+
ARGV.shift if ['--help', '-h'].include?(ARGV[0])
|
13
|
+
RubiGen::Base.use_component_sources! [:rubygems, :newgem, :newgem_theme, :test_unit]
|
14
|
+
RubiGen::Scripts::Generate.new.run(ARGV)
|
@@ -0,0 +1,65 @@
|
|
1
|
+
require File.dirname(__FILE__) + '/test_helper.rb'
|
2
|
+
|
3
|
+
class TestFastaFile < Test::Unit::TestCase
|
4
|
+
|
5
|
+
def setup
|
6
|
+
@test_file='/tmp/fbinfile';
|
7
|
+
|
8
|
+
@seq_fasta='ACTG'
|
9
|
+
@seq_qual=[25]
|
10
|
+
@seq_name='SEQ'
|
11
|
+
|
12
|
+
end
|
13
|
+
|
14
|
+
|
15
|
+
def fill_file(n)
|
16
|
+
f=File.new(@test_file+'.fasta','w')
|
17
|
+
q=File.new(@test_file+'.qual','w')
|
18
|
+
|
19
|
+
n.times do |c|
|
20
|
+
i = c+1
|
21
|
+
name = ">#{@seq_name+i.to_s} comments"
|
22
|
+
|
23
|
+
f.puts(name)
|
24
|
+
f.puts(@seq_fasta*i)
|
25
|
+
|
26
|
+
q.puts(name)
|
27
|
+
q.puts((@seq_qual*i*@seq_fasta.length).join(' '))
|
28
|
+
|
29
|
+
|
30
|
+
end
|
31
|
+
|
32
|
+
f.close
|
33
|
+
q.close
|
34
|
+
end
|
35
|
+
|
36
|
+
def test_each
|
37
|
+
|
38
|
+
# make new file and fill with data
|
39
|
+
fill_file(100)
|
40
|
+
|
41
|
+
fqr=FastaFile.new(@test_file+'.fasta')
|
42
|
+
|
43
|
+
i=1
|
44
|
+
|
45
|
+
|
46
|
+
fqr.each do |n,s|
|
47
|
+
|
48
|
+
#puts n,s.length,q.split(' ').length
|
49
|
+
assert(@seq_name+i.to_s==n)
|
50
|
+
assert(@seq_fasta*i==s)
|
51
|
+
|
52
|
+
i+=1
|
53
|
+
end
|
54
|
+
|
55
|
+
fqr.close
|
56
|
+
|
57
|
+
rescue Exception => e
|
58
|
+
puts "failed in #{n} , #{s}"
|
59
|
+
puts "expected #{@seq_name+i.to_s} , #{@seq_fasta*i==s}"
|
60
|
+
|
61
|
+
puts e.message, e.backtrace
|
62
|
+
|
63
|
+
end
|
64
|
+
|
65
|
+
end
|
@@ -0,0 +1,161 @@
|
|
1
|
+
require File.dirname(__FILE__) + '/test_helper.rb'
|
2
|
+
|
3
|
+
class TestFastaQualFile < Test::Unit::TestCase
|
4
|
+
|
5
|
+
|
6
|
+
def setup
|
7
|
+
@test_file='/tmp/fbinfile';
|
8
|
+
|
9
|
+
@seq_fasta='ACTG'
|
10
|
+
@seq_qual=[25]
|
11
|
+
@seq_name='SEQ'
|
12
|
+
|
13
|
+
end
|
14
|
+
|
15
|
+
|
16
|
+
def fill_file(n)
|
17
|
+
f=File.new(@test_file+'.fasta','w')
|
18
|
+
q=File.new(@test_file+'.qual','w')
|
19
|
+
|
20
|
+
n.times do |c|
|
21
|
+
i = c+1
|
22
|
+
name = ">#{@seq_name+i.to_s} comments"
|
23
|
+
|
24
|
+
f.puts(name)
|
25
|
+
f.puts(@seq_fasta*i)
|
26
|
+
|
27
|
+
q.puts(name)
|
28
|
+
q.puts((@seq_qual*i*@seq_fasta.length).join(' '))
|
29
|
+
|
30
|
+
|
31
|
+
end
|
32
|
+
|
33
|
+
f.close
|
34
|
+
q.close
|
35
|
+
end
|
36
|
+
|
37
|
+
def test_next_seq
|
38
|
+
|
39
|
+
# make new file and fill with data
|
40
|
+
fill_file(100)
|
41
|
+
|
42
|
+
|
43
|
+
fqr=FastaQualFile.new(@test_file+'.fasta',@test_file+'.qual')
|
44
|
+
n,s,q = ['']*3
|
45
|
+
i=0
|
46
|
+
|
47
|
+
100.times do |c|
|
48
|
+
i = c+1
|
49
|
+
n,s,q = fqr.next_seq
|
50
|
+
|
51
|
+
#puts n,s.length,q.split(' ').length
|
52
|
+
assert(@seq_name+i.to_s==n)
|
53
|
+
assert(@seq_fasta*i==s)
|
54
|
+
assert((@seq_qual*i*@seq_fasta.length).join(' ')==q)
|
55
|
+
|
56
|
+
end
|
57
|
+
|
58
|
+
fqr.close
|
59
|
+
|
60
|
+
rescue Exception => e
|
61
|
+
puts "failed in #{n} , #{s}, #{q}"
|
62
|
+
puts "expected #{@seq_name+i.to_s} , #{@seq_fasta*i==s}, #{(@seq_qual*i*@seq_fasta.length).join(' ')}"
|
63
|
+
puts e.message, e.backtrace
|
64
|
+
|
65
|
+
end
|
66
|
+
|
67
|
+
def test_each
|
68
|
+
|
69
|
+
# make new file and fill with data
|
70
|
+
fill_file(100)
|
71
|
+
|
72
|
+
|
73
|
+
fqr=FastaQualFile.new(@test_file+'.fasta',@test_file+'.qual')
|
74
|
+
|
75
|
+
i=1
|
76
|
+
|
77
|
+
fqr.each do |n,s,q|
|
78
|
+
|
79
|
+
#puts n,s.length,q.split(' ').length
|
80
|
+
assert(@seq_name+i.to_s==n)
|
81
|
+
assert(@seq_fasta*i==s)
|
82
|
+
assert((@seq_qual*i*@seq_fasta.length).join(' ')==q)
|
83
|
+
|
84
|
+
i+=1
|
85
|
+
end
|
86
|
+
|
87
|
+
fqr.close
|
88
|
+
|
89
|
+
rescue Exception => e
|
90
|
+
puts "failed in #{n} , #{s}, #{q}"
|
91
|
+
puts "expected #{@seq_name+i.to_s} , #{@seq_fasta*i==s}, #{(@seq_qual*i*@seq_fasta.length).join(' ')}"
|
92
|
+
puts e.message, e.backtrace
|
93
|
+
|
94
|
+
end
|
95
|
+
|
96
|
+
def test_next_seq_only_fasta
|
97
|
+
|
98
|
+
# make new file and fill with data
|
99
|
+
fill_file(100)
|
100
|
+
|
101
|
+
|
102
|
+
fqr=FastaQualFile.new(@test_file+'.fasta')
|
103
|
+
n,s = ['']*2
|
104
|
+
i=0
|
105
|
+
|
106
|
+
100.times do |c|
|
107
|
+
i = c+1
|
108
|
+
n,s,q = fqr.next_seq
|
109
|
+
|
110
|
+
#puts n,s.length,q.split(' ').length
|
111
|
+
assert(@seq_name+i.to_s==n)
|
112
|
+
assert(@seq_fasta*i==s)
|
113
|
+
assert(q.nil?)
|
114
|
+
|
115
|
+
end
|
116
|
+
|
117
|
+
fqr.close
|
118
|
+
|
119
|
+
rescue Exception => e
|
120
|
+
puts "failed in #{n} , #{s}"
|
121
|
+
puts "expected #{@seq_name+i.to_s} , #{@seq_fasta*i==s}"
|
122
|
+
puts e.message, e.backtrace
|
123
|
+
|
124
|
+
end
|
125
|
+
|
126
|
+
|
127
|
+
def test_each_only_fasta
|
128
|
+
|
129
|
+
count=100
|
130
|
+
# make new file and fill with data
|
131
|
+
fill_file(count)
|
132
|
+
|
133
|
+
|
134
|
+
fqr=FastaQualFile.new(@test_file+'.fasta')
|
135
|
+
|
136
|
+
i=1
|
137
|
+
|
138
|
+
fqr.each do |n,s,q|
|
139
|
+
|
140
|
+
#puts n,s.length,q.split(' ').length
|
141
|
+
assert(@seq_name+i.to_s==n)
|
142
|
+
assert(@seq_fasta*i==s)
|
143
|
+
|
144
|
+
assert(q.nil?)
|
145
|
+
|
146
|
+
i+=1
|
147
|
+
end
|
148
|
+
|
149
|
+
assert(i==count+1)
|
150
|
+
|
151
|
+
fqr.close
|
152
|
+
|
153
|
+
rescue Exception => e
|
154
|
+
puts "failed in #{n} , #{s}"
|
155
|
+
puts "expected #{@seq_name+i.to_s} , #{@seq_fasta*i==s}"
|
156
|
+
puts e.message, e.backtrace
|
157
|
+
|
158
|
+
end
|
159
|
+
|
160
|
+
|
161
|
+
end
|
data/test/test_helper.rb
ADDED
metadata
ADDED
@@ -0,0 +1,83 @@
|
|
1
|
+
--- !ruby/object:Gem::Specification
|
2
|
+
name: scbi_fasta
|
3
|
+
version: !ruby/object:Gem::Version
|
4
|
+
prerelease:
|
5
|
+
version: 0.1.8
|
6
|
+
platform: ruby
|
7
|
+
authors:
|
8
|
+
- Almudena Bocinos
|
9
|
+
autorequire:
|
10
|
+
bindir: bin
|
11
|
+
cert_chain: []
|
12
|
+
|
13
|
+
date: 2011-06-13 00:00:00 Z
|
14
|
+
dependencies:
|
15
|
+
- !ruby/object:Gem::Dependency
|
16
|
+
name: hoe
|
17
|
+
prerelease: false
|
18
|
+
requirement: &id001 !ruby/object:Gem::Requirement
|
19
|
+
none: false
|
20
|
+
requirements:
|
21
|
+
- - ">="
|
22
|
+
- !ruby/object:Gem::Version
|
23
|
+
version: 2.8.0
|
24
|
+
type: :development
|
25
|
+
version_requirements: *id001
|
26
|
+
description: scbi_fasta is a ruby gem to read FASTA+QUAL files (DNA/RNA sequences).
|
27
|
+
email:
|
28
|
+
- alkoke@gmail.com
|
29
|
+
executables: []
|
30
|
+
|
31
|
+
extensions: []
|
32
|
+
|
33
|
+
extra_rdoc_files:
|
34
|
+
- History.txt
|
35
|
+
- Manifest.txt
|
36
|
+
- PostInstall.txt
|
37
|
+
files:
|
38
|
+
- History.txt
|
39
|
+
- Manifest.txt
|
40
|
+
- PostInstall.txt
|
41
|
+
- README.rdoc
|
42
|
+
- Rakefile
|
43
|
+
- lib/scbi_fasta.rb
|
44
|
+
- lib/scbi_fasta/fasta_qual_file.rb
|
45
|
+
- lib/scbi_fasta/fasta_file.rb
|
46
|
+
- script/console
|
47
|
+
- script/destroy
|
48
|
+
- script/generate
|
49
|
+
- test/test_helper.rb
|
50
|
+
- test/test_fasta_file.rb
|
51
|
+
- test/test_fasta_qual_file.rb
|
52
|
+
homepage: http://www.scbi.uma.es/downloads
|
53
|
+
licenses: []
|
54
|
+
|
55
|
+
post_install_message: PostInstall.txt
|
56
|
+
rdoc_options:
|
57
|
+
- --main
|
58
|
+
- README.rdoc
|
59
|
+
require_paths:
|
60
|
+
- lib
|
61
|
+
required_ruby_version: !ruby/object:Gem::Requirement
|
62
|
+
none: false
|
63
|
+
requirements:
|
64
|
+
- - ">="
|
65
|
+
- !ruby/object:Gem::Version
|
66
|
+
version: "0"
|
67
|
+
required_rubygems_version: !ruby/object:Gem::Requirement
|
68
|
+
none: false
|
69
|
+
requirements:
|
70
|
+
- - ">="
|
71
|
+
- !ruby/object:Gem::Version
|
72
|
+
version: "0"
|
73
|
+
requirements: []
|
74
|
+
|
75
|
+
rubyforge_project: scbi_fasta
|
76
|
+
rubygems_version: 1.7.2
|
77
|
+
signing_key:
|
78
|
+
specification_version: 3
|
79
|
+
summary: scbi_fasta is a ruby gem to read FASTA+QUAL files (DNA/RNA sequences).
|
80
|
+
test_files:
|
81
|
+
- test/test_fasta_file.rb
|
82
|
+
- test/test_fasta_qual_file.rb
|
83
|
+
- test/test_helper.rb
|