BioDSL 1.0.0
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +7 -0
- data/.gitignore +10 -0
- data/BioDSL.gemspec +64 -0
- data/LICENSE +339 -0
- data/README.md +205 -0
- data/Rakefile +94 -0
- data/examples/fastq_to_fasta.rb +8 -0
- data/lib/BioDSL/cary.rb +242 -0
- data/lib/BioDSL/command.rb +133 -0
- data/lib/BioDSL/commands/add_key.rb +110 -0
- data/lib/BioDSL/commands/align_seq_mothur.rb +194 -0
- data/lib/BioDSL/commands/analyze_residue_distribution.rb +222 -0
- data/lib/BioDSL/commands/assemble_pairs.rb +336 -0
- data/lib/BioDSL/commands/assemble_seq_idba.rb +230 -0
- data/lib/BioDSL/commands/assemble_seq_ray.rb +345 -0
- data/lib/BioDSL/commands/assemble_seq_spades.rb +252 -0
- data/lib/BioDSL/commands/classify_seq.rb +217 -0
- data/lib/BioDSL/commands/classify_seq_mothur.rb +226 -0
- data/lib/BioDSL/commands/clip_primer.rb +318 -0
- data/lib/BioDSL/commands/cluster_otus.rb +181 -0
- data/lib/BioDSL/commands/collapse_otus.rb +170 -0
- data/lib/BioDSL/commands/collect_otus.rb +150 -0
- data/lib/BioDSL/commands/complement_seq.rb +117 -0
- data/lib/BioDSL/commands/count.rb +135 -0
- data/lib/BioDSL/commands/count_values.rb +149 -0
- data/lib/BioDSL/commands/degap_seq.rb +253 -0
- data/lib/BioDSL/commands/dereplicate_seq.rb +168 -0
- data/lib/BioDSL/commands/dump.rb +157 -0
- data/lib/BioDSL/commands/filter_rrna.rb +239 -0
- data/lib/BioDSL/commands/genecall.rb +237 -0
- data/lib/BioDSL/commands/grab.rb +535 -0
- data/lib/BioDSL/commands/index_taxonomy.rb +226 -0
- data/lib/BioDSL/commands/mask_seq.rb +175 -0
- data/lib/BioDSL/commands/mean_scores.rb +168 -0
- data/lib/BioDSL/commands/merge_pair_seq.rb +175 -0
- data/lib/BioDSL/commands/merge_table.rb +225 -0
- data/lib/BioDSL/commands/merge_values.rb +113 -0
- data/lib/BioDSL/commands/plot_heatmap.rb +233 -0
- data/lib/BioDSL/commands/plot_histogram.rb +306 -0
- data/lib/BioDSL/commands/plot_matches.rb +282 -0
- data/lib/BioDSL/commands/plot_residue_distribution.rb +278 -0
- data/lib/BioDSL/commands/plot_scores.rb +285 -0
- data/lib/BioDSL/commands/random.rb +153 -0
- data/lib/BioDSL/commands/read_fasta.rb +222 -0
- data/lib/BioDSL/commands/read_fastq.rb +414 -0
- data/lib/BioDSL/commands/read_table.rb +329 -0
- data/lib/BioDSL/commands/reverse_seq.rb +113 -0
- data/lib/BioDSL/commands/slice_align.rb +400 -0
- data/lib/BioDSL/commands/slice_seq.rb +151 -0
- data/lib/BioDSL/commands/sort.rb +223 -0
- data/lib/BioDSL/commands/split_pair_seq.rb +220 -0
- data/lib/BioDSL/commands/split_values.rb +165 -0
- data/lib/BioDSL/commands/trim_primer.rb +314 -0
- data/lib/BioDSL/commands/trim_seq.rb +192 -0
- data/lib/BioDSL/commands/uchime_ref.rb +170 -0
- data/lib/BioDSL/commands/uclust.rb +286 -0
- data/lib/BioDSL/commands/unique_values.rb +145 -0
- data/lib/BioDSL/commands/usearch_global.rb +171 -0
- data/lib/BioDSL/commands/usearch_local.rb +171 -0
- data/lib/BioDSL/commands/write_fasta.rb +207 -0
- data/lib/BioDSL/commands/write_fastq.rb +191 -0
- data/lib/BioDSL/commands/write_table.rb +419 -0
- data/lib/BioDSL/commands/write_tree.rb +167 -0
- data/lib/BioDSL/commands.rb +31 -0
- data/lib/BioDSL/config.rb +55 -0
- data/lib/BioDSL/csv.rb +307 -0
- data/lib/BioDSL/debug.rb +42 -0
- data/lib/BioDSL/fasta.rb +133 -0
- data/lib/BioDSL/fastq.rb +77 -0
- data/lib/BioDSL/filesys.rb +137 -0
- data/lib/BioDSL/fork.rb +145 -0
- data/lib/BioDSL/hamming.rb +128 -0
- data/lib/BioDSL/helpers/aux_helper.rb +44 -0
- data/lib/BioDSL/helpers/email_helper.rb +66 -0
- data/lib/BioDSL/helpers/history_helper.rb +40 -0
- data/lib/BioDSL/helpers/log_helper.rb +55 -0
- data/lib/BioDSL/helpers/options_helper.rb +405 -0
- data/lib/BioDSL/helpers/status_helper.rb +132 -0
- data/lib/BioDSL/helpers.rb +35 -0
- data/lib/BioDSL/html_report.rb +200 -0
- data/lib/BioDSL/math.rb +55 -0
- data/lib/BioDSL/mummer.rb +216 -0
- data/lib/BioDSL/pipeline.rb +354 -0
- data/lib/BioDSL/seq/ambiguity.rb +66 -0
- data/lib/BioDSL/seq/assemble.rb +240 -0
- data/lib/BioDSL/seq/backtrack.rb +252 -0
- data/lib/BioDSL/seq/digest.rb +99 -0
- data/lib/BioDSL/seq/dynamic.rb +263 -0
- data/lib/BioDSL/seq/homopolymer.rb +59 -0
- data/lib/BioDSL/seq/kmer.rb +293 -0
- data/lib/BioDSL/seq/levenshtein.rb +113 -0
- data/lib/BioDSL/seq/translate.rb +109 -0
- data/lib/BioDSL/seq/trim.rb +188 -0
- data/lib/BioDSL/seq.rb +742 -0
- data/lib/BioDSL/serializer.rb +98 -0
- data/lib/BioDSL/stream.rb +113 -0
- data/lib/BioDSL/taxonomy.rb +691 -0
- data/lib/BioDSL/test.rb +42 -0
- data/lib/BioDSL/tmp_dir.rb +68 -0
- data/lib/BioDSL/usearch.rb +301 -0
- data/lib/BioDSL/verbose.rb +42 -0
- data/lib/BioDSL/version.rb +31 -0
- data/lib/BioDSL.rb +81 -0
- data/test/BioDSL/commands/test_add_key.rb +105 -0
- data/test/BioDSL/commands/test_align_seq_mothur.rb +99 -0
- data/test/BioDSL/commands/test_analyze_residue_distribution.rb +134 -0
- data/test/BioDSL/commands/test_assemble_pairs.rb +459 -0
- data/test/BioDSL/commands/test_assemble_seq_idba.rb +50 -0
- data/test/BioDSL/commands/test_assemble_seq_ray.rb +51 -0
- data/test/BioDSL/commands/test_assemble_seq_spades.rb +50 -0
- data/test/BioDSL/commands/test_classify_seq.rb +50 -0
- data/test/BioDSL/commands/test_classify_seq_mothur.rb +59 -0
- data/test/BioDSL/commands/test_clip_primer.rb +377 -0
- data/test/BioDSL/commands/test_cluster_otus.rb +128 -0
- data/test/BioDSL/commands/test_collapse_otus.rb +81 -0
- data/test/BioDSL/commands/test_collect_otus.rb +82 -0
- data/test/BioDSL/commands/test_complement_seq.rb +78 -0
- data/test/BioDSL/commands/test_count.rb +103 -0
- data/test/BioDSL/commands/test_count_values.rb +85 -0
- data/test/BioDSL/commands/test_degap_seq.rb +96 -0
- data/test/BioDSL/commands/test_dereplicate_seq.rb +92 -0
- data/test/BioDSL/commands/test_dump.rb +109 -0
- data/test/BioDSL/commands/test_filter_rrna.rb +128 -0
- data/test/BioDSL/commands/test_genecall.rb +50 -0
- data/test/BioDSL/commands/test_grab.rb +398 -0
- data/test/BioDSL/commands/test_index_taxonomy.rb +62 -0
- data/test/BioDSL/commands/test_mask_seq.rb +98 -0
- data/test/BioDSL/commands/test_mean_scores.rb +111 -0
- data/test/BioDSL/commands/test_merge_pair_seq.rb +115 -0
- data/test/BioDSL/commands/test_merge_table.rb +131 -0
- data/test/BioDSL/commands/test_merge_values.rb +83 -0
- data/test/BioDSL/commands/test_plot_heatmap.rb +185 -0
- data/test/BioDSL/commands/test_plot_histogram.rb +194 -0
- data/test/BioDSL/commands/test_plot_matches.rb +157 -0
- data/test/BioDSL/commands/test_plot_residue_distribution.rb +309 -0
- data/test/BioDSL/commands/test_plot_scores.rb +308 -0
- data/test/BioDSL/commands/test_random.rb +88 -0
- data/test/BioDSL/commands/test_read_fasta.rb +229 -0
- data/test/BioDSL/commands/test_read_fastq.rb +552 -0
- data/test/BioDSL/commands/test_read_table.rb +327 -0
- data/test/BioDSL/commands/test_reverse_seq.rb +79 -0
- data/test/BioDSL/commands/test_slice_align.rb +218 -0
- data/test/BioDSL/commands/test_slice_seq.rb +131 -0
- data/test/BioDSL/commands/test_sort.rb +128 -0
- data/test/BioDSL/commands/test_split_pair_seq.rb +164 -0
- data/test/BioDSL/commands/test_split_values.rb +95 -0
- data/test/BioDSL/commands/test_trim_primer.rb +329 -0
- data/test/BioDSL/commands/test_trim_seq.rb +150 -0
- data/test/BioDSL/commands/test_uchime_ref.rb +113 -0
- data/test/BioDSL/commands/test_uclust.rb +139 -0
- data/test/BioDSL/commands/test_unique_values.rb +98 -0
- data/test/BioDSL/commands/test_usearch_global.rb +123 -0
- data/test/BioDSL/commands/test_usearch_local.rb +125 -0
- data/test/BioDSL/commands/test_write_fasta.rb +159 -0
- data/test/BioDSL/commands/test_write_fastq.rb +166 -0
- data/test/BioDSL/commands/test_write_table.rb +411 -0
- data/test/BioDSL/commands/test_write_tree.rb +122 -0
- data/test/BioDSL/helpers/test_options_helper.rb +272 -0
- data/test/BioDSL/seq/test_assemble.rb +98 -0
- data/test/BioDSL/seq/test_backtrack.rb +176 -0
- data/test/BioDSL/seq/test_digest.rb +71 -0
- data/test/BioDSL/seq/test_dynamic.rb +133 -0
- data/test/BioDSL/seq/test_homopolymer.rb +58 -0
- data/test/BioDSL/seq/test_kmer.rb +134 -0
- data/test/BioDSL/seq/test_translate.rb +75 -0
- data/test/BioDSL/seq/test_trim.rb +101 -0
- data/test/BioDSL/test_cary.rb +176 -0
- data/test/BioDSL/test_command.rb +45 -0
- data/test/BioDSL/test_csv.rb +514 -0
- data/test/BioDSL/test_debug.rb +42 -0
- data/test/BioDSL/test_fasta.rb +154 -0
- data/test/BioDSL/test_fastq.rb +46 -0
- data/test/BioDSL/test_filesys.rb +145 -0
- data/test/BioDSL/test_fork.rb +85 -0
- data/test/BioDSL/test_math.rb +41 -0
- data/test/BioDSL/test_mummer.rb +79 -0
- data/test/BioDSL/test_pipeline.rb +187 -0
- data/test/BioDSL/test_seq.rb +790 -0
- data/test/BioDSL/test_serializer.rb +72 -0
- data/test/BioDSL/test_stream.rb +55 -0
- data/test/BioDSL/test_taxonomy.rb +336 -0
- data/test/BioDSL/test_test.rb +42 -0
- data/test/BioDSL/test_tmp_dir.rb +58 -0
- data/test/BioDSL/test_usearch.rb +33 -0
- data/test/BioDSL/test_verbose.rb +42 -0
- data/test/helper.rb +82 -0
- data/www/command.html.haml +14 -0
- data/www/css.html.haml +55 -0
- data/www/input_files.html.haml +3 -0
- data/www/layout.html.haml +12 -0
- data/www/output_files.html.haml +3 -0
- data/www/overview.html.haml +15 -0
- data/www/pipeline.html.haml +4 -0
- data/www/png.html.haml +2 -0
- data/www/status.html.haml +9 -0
- data/www/time.html.haml +11 -0
- metadata +503 -0
@@ -0,0 +1,217 @@
|
|
1
|
+
# >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>><<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< #
|
2
|
+
# #
|
3
|
+
# Copyright (C) 2007-2015 Martin Asser Hansen (mail@maasha.dk). #
|
4
|
+
# #
|
5
|
+
# This program is free software; you can redistribute it and/or #
|
6
|
+
# modify it under the terms of the GNU General Public License #
|
7
|
+
# as published by the Free Software Foundation; either version 2 #
|
8
|
+
# of the License, or (at your option) any later version. #
|
9
|
+
# #
|
10
|
+
# This program is distributed in the hope that it will be useful, #
|
11
|
+
# but WITHOUT ANY WARRANTY; without even the implied warranty of #
|
12
|
+
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the #
|
13
|
+
# GNU General Public License for more details. #
|
14
|
+
# #
|
15
|
+
# You should have received a copy of the GNU General Public License #
|
16
|
+
# along with this program; if not, write to the Free Software #
|
17
|
+
# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, #
|
18
|
+
# USA. #
|
19
|
+
# #
|
20
|
+
# http://www.gnu.org/copyleft/gpl.html #
|
21
|
+
# #
|
22
|
+
# >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>><<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< #
|
23
|
+
# #
|
24
|
+
# This software is part of the BioDSL framework (www.BioDSL.org). #
|
25
|
+
# #
|
26
|
+
# >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>><<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< #
|
27
|
+
|
28
|
+
module BioDSL
|
29
|
+
# == Classify sequences in the stream.
|
30
|
+
#
|
31
|
+
# +classify_seq+ searches sequences in the stream against a pre-indexed
|
32
|
+
# (using +index_taxonomy+) database. The database consists a taxonomic tree
|
33
|
+
# index and indices for each taxonomic level saved in the following files
|
34
|
+
# (here using the prefix "taxonomy"):
|
35
|
+
#
|
36
|
+
# * taxonomy_tax_index.dat - return node for a given node id.
|
37
|
+
# * taxonomy_kmer_index.dat - return list of node ids for a given level
|
38
|
+
# and kmer.
|
39
|
+
#
|
40
|
+
# Each sequence is broken down into unique kmers of a given kmer_size
|
41
|
+
# overlapping with a given step_size - see +index_taxonomy+. Now, for each
|
42
|
+
# taxonomic level, starting from species all nodes for each kmer is looked
|
43
|
+
# up in the database. The nodes containing most kmers are considered hits.
|
44
|
+
# If there are no hits at a taxonomic level, we move to the next level. Hits
|
45
|
+
# are sorted according to how many kmers matched this particular node and a
|
46
|
+
# consensus taxonomy string is determined. Hits are also filtered with the
|
47
|
+
# following options:
|
48
|
+
#
|
49
|
+
# * hits_max - Include maximally this number of hits in the consensus.
|
50
|
+
# * best_only - Include only the best scoring hits in the consensus.
|
51
|
+
# That is if a hit consists of 344 kmers out of 345
|
52
|
+
# possible, only hits with 344 kmers are included.
|
53
|
+
# * coverage - Filter hits based on kmer coverage. If a hit contains
|
54
|
+
# fewer kmers than the total amount of kmers x coverage
|
55
|
+
# it will be filtered.
|
56
|
+
# * consensus - For a number of hits accept consensus at a given level
|
57
|
+
# if within this percentage.
|
58
|
+
#
|
59
|
+
# The output of +classify_seq+ are sequence type records with the
|
60
|
+
# additional keys:
|
61
|
+
#
|
62
|
+
# * TAXONOMY_HITS - The number of hits used in the consensus.
|
63
|
+
# * TAXONOMY - The taxonomy string.
|
64
|
+
#
|
65
|
+
# The consensus is determined from a list of taxonomic strings, i.e. the
|
66
|
+
# TAXONOMIC_HITS, and is composed of a consensus for each taxonomic level.
|
67
|
+
# E.g. for the kingdom level if 60% of the taxonomic strings indicate
|
68
|
+
# 'Bacteria' and the consensus is 50% then the consensus for the kingdom
|
69
|
+
# level will be reported as 'Bacteria(60)'. If the name at any level
|
70
|
+
# consists of multiple words they are treated independently. E.g if we have
|
71
|
+
# three taxonomic strings at the species level with the names:
|
72
|
+
#
|
73
|
+
# * Escherichia coli K-12
|
74
|
+
# * Escherichia coli sp. AC3432
|
75
|
+
# * Escherichia coli sp. AC1232
|
76
|
+
#
|
77
|
+
# The corresponding consensus for that level will be reported as
|
78
|
+
# 'Escherichia coli sp.(100/100/66)'. The forth word in the last two
|
79
|
+
# taxonomy strings (AC3432 and AC1232) have a consensus below 50% and are
|
80
|
+
# ignored.
|
81
|
+
#
|
82
|
+
# == Usage
|
83
|
+
#
|
84
|
+
# classify_seq(<dir: <dir>>[, prefix: <string>[, kmer_size: <uint>
|
85
|
+
# [, step_size: <uint>[, hits_max: <uint>[, consensus:
|
86
|
+
# <float>[, coverage: <float>[, best_only: <bool>]]]]]]])
|
87
|
+
#
|
88
|
+
# === Options
|
89
|
+
#
|
90
|
+
# * dir: <dir> - Directory containing taxonomy files.
|
91
|
+
# * prefix: <string> - Taxonomy files prefix (default="taxonomy").
|
92
|
+
# * kmer_size: <uint> - Kmer size (default=8).
|
93
|
+
# * step_size: <uint> - Step size (default=1).
|
94
|
+
# * hits_max: <uint> - Maximum hits to include in consensus (default=50).
|
95
|
+
# * consensus: <float> - Consensus cutoff (default=0.51).
|
96
|
+
# * coverage: <float> - Coverate cutoff (default=0.9).
|
97
|
+
# * best_only: <bool> - Only use best hits for consensus (default=true).
|
98
|
+
#
|
99
|
+
# == Examples
|
100
|
+
#
|
101
|
+
# To classify a bunch of OTU sequences in the file +otus.fna+ we do:
|
102
|
+
#
|
103
|
+
# BP.new.
|
104
|
+
# read_fasta(input: "otus.fna").
|
105
|
+
# classify_seq(dir: "RDP11_3").
|
106
|
+
# write_table(keys: [:SEQ_NAME, :TAXONOMY_HITS, :TAXONOMY]).
|
107
|
+
# run
|
108
|
+
#
|
109
|
+
# OTU_0 1 K#Bacteria(100);P#Proteobacteria(100);C#Gammaproteobacteria...
|
110
|
+
# OTU_1 1 K#Bacteria(100);P#Proteobacteria(100);C#Gammaproteobacteria...
|
111
|
+
# OTU_2 1 K#Bacteria(100);P#Proteobacteria(100);C#Gammaproteobacteria...
|
112
|
+
# OTU_3 1 K#Bacteria(100);P#Proteobacteria(100);C#Gammaproteobacteria...
|
113
|
+
# OTU_4 2 K#Bacteria(100);P#Fusobacteria(100);C#Fusobacteriia(100);O#...
|
114
|
+
class ClassifySeq
|
115
|
+
STATS = %i(records_in records_out sequences_in sequences_out residues_in
|
116
|
+
residues_out)
|
117
|
+
|
118
|
+
# Constructor for the ClassifySeq class.
|
119
|
+
#
|
120
|
+
# @param [Hash] options Options hash.
|
121
|
+
# @option options [String] :dir Directory path with indexes.
|
122
|
+
# @option options [String] :prefix Index prefix.
|
123
|
+
# @option options [Integer] :kmer_size Kmer size.
|
124
|
+
# @option options [Integer] :step_size Step size.
|
125
|
+
# @option options [Integer] :hits_max Max hits to report per sequence.
|
126
|
+
# @option options [Float] :consensus Taxonomy string consensus percent.
|
127
|
+
# @option options [Float] :coverage Kmer coverage filter percent.
|
128
|
+
# @option options [Boolean] :best_only Flag to report best hit only.
|
129
|
+
#
|
130
|
+
# @return [ClassifySeq] Returns an instance of the class.
|
131
|
+
def initialize(options)
|
132
|
+
@options = options
|
133
|
+
|
134
|
+
check_options
|
135
|
+
defaults
|
136
|
+
end
|
137
|
+
|
138
|
+
# Return a lambda for the ClassifySeq command.
|
139
|
+
#
|
140
|
+
# @return [Proc] Returns the command lambda.
|
141
|
+
def lmb
|
142
|
+
lambda do |input, output, status|
|
143
|
+
status_init(status, STATS)
|
144
|
+
|
145
|
+
@status[:sequences_in] = 0
|
146
|
+
|
147
|
+
search = BioDSL::Taxonomy::Search.new(@options)
|
148
|
+
|
149
|
+
input.each_with_index do |record, i|
|
150
|
+
@status[:records_in] += 1
|
151
|
+
|
152
|
+
classify_seq(record, i, search) if record.key? :SEQ
|
153
|
+
|
154
|
+
output << record
|
155
|
+
@status[:records_out] += 1
|
156
|
+
end
|
157
|
+
end
|
158
|
+
end
|
159
|
+
|
160
|
+
private
|
161
|
+
|
162
|
+
# Check options.
|
163
|
+
def check_options
|
164
|
+
options_allowed(@options, :dir, :prefix, :kmer_size, :step_size,
|
165
|
+
:hits_max, :consensus, :coverage, :best_only)
|
166
|
+
options_required(@options, :dir)
|
167
|
+
options_dirs_exist(@options, :dir)
|
168
|
+
options_allowed_values(@options, best_only: [nil, true, false])
|
169
|
+
|
170
|
+
run_assertions
|
171
|
+
end
|
172
|
+
|
173
|
+
# Run assertions.
|
174
|
+
def run_assertions
|
175
|
+
options_assert(@options, ':kmer_size > 0')
|
176
|
+
options_assert(@options, ':kmer_size <= 12')
|
177
|
+
options_assert(@options, ':step_size > 0')
|
178
|
+
options_assert(@options, ':step_size <= 12')
|
179
|
+
options_assert(@options, ':hits_max > 0')
|
180
|
+
options_assert(@options, ':consensus > 0')
|
181
|
+
options_assert(@options, ':consensus <= 1')
|
182
|
+
options_assert(@options, ':coverage > 0')
|
183
|
+
options_assert(@options, ':coverage <= 1')
|
184
|
+
end
|
185
|
+
|
186
|
+
# Set default options.
|
187
|
+
def defaults
|
188
|
+
@options[:prefix] ||= 'taxonomy'
|
189
|
+
@options[:kmer_size] ||= 8
|
190
|
+
@options[:step_size] ||= 1
|
191
|
+
@options[:hits_max] ||= 50
|
192
|
+
@options[:consensus] ||= 0.51
|
193
|
+
@options[:coverage] ||= 0.9
|
194
|
+
@options[:best_only] = true if @options[:best_only].nil?
|
195
|
+
end
|
196
|
+
|
197
|
+
# Execute classfication of a sequence containing record.
|
198
|
+
#
|
199
|
+
# @param record [Hash] BioDSL record.
|
200
|
+
# @param i [Fixnum] Record number,
|
201
|
+
# @param search [BioDSL::Taxonomy::Search] Search object.
|
202
|
+
def classify_seq(record, i, search)
|
203
|
+
@status[:sequences_in] += 1
|
204
|
+
@status[:sequences_out] += 1
|
205
|
+
@status[:residues_in] += record[:SEQ].length
|
206
|
+
@status[:residues_out] += record[:SEQ].length
|
207
|
+
seq_name = record[:SEQ_NAME] || i.to_s
|
208
|
+
|
209
|
+
result = search.execute(BioDSL::Seq.new(seq_name: seq_name,
|
210
|
+
seq: record[:SEQ]))
|
211
|
+
|
212
|
+
record[:TAXONOMY] = result.taxonomy
|
213
|
+
record[:TAXONOMY_HITS] = result.hits
|
214
|
+
record[:RECORD_TYPE] = 'taxonomy'
|
215
|
+
end
|
216
|
+
end
|
217
|
+
end
|
@@ -0,0 +1,226 @@
|
|
1
|
+
# >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>><<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< #
|
2
|
+
# #
|
3
|
+
# Copyright (C) 2007-2015 Martin Asser Hansen (mail@maasha.dk). #
|
4
|
+
# #
|
5
|
+
# This program is free software; you can redistribute it and/or #
|
6
|
+
# modify it under the terms of the GNU General Public License #
|
7
|
+
# as published by the Free Software Foundation; either version 2 #
|
8
|
+
# of the License, or (at your option) any later version. #
|
9
|
+
# #
|
10
|
+
# This program is distributed in the hope that it will be useful, #
|
11
|
+
# but WITHOUT ANY WARRANTY; without even the implied warranty of #
|
12
|
+
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the #
|
13
|
+
# GNU General Public License for more details. #
|
14
|
+
# #
|
15
|
+
# You should have received a copy of the GNU General Public License #
|
16
|
+
# along with this program; if not, write to the Free Software #
|
17
|
+
# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, #
|
18
|
+
# USA. #
|
19
|
+
# #
|
20
|
+
# http://www.gnu.org/copyleft/gpl.html #
|
21
|
+
# #
|
22
|
+
# >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>><<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< #
|
23
|
+
# #
|
24
|
+
# This software is part of the BioDSL framework (www.BioDSL.org). #
|
25
|
+
# #
|
26
|
+
# >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>><<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< #
|
27
|
+
|
28
|
+
module BioDSL
|
29
|
+
# == Run classify_seq_mothur on sequences in the stream.
|
30
|
+
#
|
31
|
+
# This is a wrapper for the +mothur+ command +classify.seqs()+. Basically,
|
32
|
+
# it classifies sequences in the stream given a database file and a taxonomy
|
33
|
+
# file which can be downloaded here:
|
34
|
+
#
|
35
|
+
# http://www.mothur.org/w/images/5/59/Trainset9_032012.pds.zip
|
36
|
+
#
|
37
|
+
# Please refer to the manual:
|
38
|
+
#
|
39
|
+
# http://www.mothur.org/wiki/Classify.seqs
|
40
|
+
#
|
41
|
+
# Mothur must be installed for +classify_seq_mothurs+ to work. Read more here:
|
42
|
+
#
|
43
|
+
# http://www.mothur.org/
|
44
|
+
#
|
45
|
+
# == Usage
|
46
|
+
#
|
47
|
+
# classify_seq_mothur(<database: <file>>, <taxonomy: <file>>
|
48
|
+
# [, confidence: <uint>[, cpus: <uint>]])
|
49
|
+
#
|
50
|
+
# === Options
|
51
|
+
#
|
52
|
+
# * database: <file> - Database to search.
|
53
|
+
# * taxonomy: <file> - Taxonomy file for mapping names.
|
54
|
+
# * confidence: <uint> - Confidence threshold (defualt=80).
|
55
|
+
# * cpus: <uint> - Number of CPU cores to use (default=1).
|
56
|
+
#
|
57
|
+
# == Examples
|
58
|
+
#
|
59
|
+
# To classify a bunch of OTU sequences in the file +otus.fna+ we do:
|
60
|
+
#
|
61
|
+
# database = "trainset9_032012.pds.fasta"
|
62
|
+
# taxonomy = "trainset9_032012.pds.tax"
|
63
|
+
#
|
64
|
+
# BP.new.
|
65
|
+
# read_fasta(input: "otus.fna").
|
66
|
+
# classify_seq_mothur(database: database, taxonomy: taxonomy).
|
67
|
+
# grab(exact: true, keys: :RECORD_TYPE, select: "taxonomy").
|
68
|
+
# write_table(output: "classified.tab", header: true, force: true,
|
69
|
+
# skip: [:RECORD_TYPE]).
|
70
|
+
# run
|
71
|
+
class ClassifySeqMothur
|
72
|
+
require 'English'
|
73
|
+
require 'BioDSL/helpers/aux_helper'
|
74
|
+
|
75
|
+
include AuxHelper
|
76
|
+
|
77
|
+
STATS = %i(records_in records_out sequences_in sequences_out
|
78
|
+
residues_in residues_out)
|
79
|
+
|
80
|
+
# Constructor for ClassifySeqMothur.
|
81
|
+
#
|
82
|
+
# @param options [Hash] Options hash.
|
83
|
+
# @option options [String] :database Path to database file.
|
84
|
+
# @option options [String] :taxonomy Path to taxonomy file.
|
85
|
+
# @option options [Integer] :confidence Confidence cutoff.
|
86
|
+
# @option options [Integer] :cpus Number of CPUs to use.
|
87
|
+
#
|
88
|
+
# @return [ClassifySeqMothur] Instance of class.
|
89
|
+
def initialize(options)
|
90
|
+
@options = options
|
91
|
+
|
92
|
+
aux_exist('mothur')
|
93
|
+
check_options
|
94
|
+
defaults
|
95
|
+
end
|
96
|
+
|
97
|
+
# Command lambda for ClassifySeqMothur.
|
98
|
+
#
|
99
|
+
# @return [Proc] Lambda for the command.
|
100
|
+
def lmb
|
101
|
+
lambda do |input, output, status|
|
102
|
+
status_init(status, STATS)
|
103
|
+
|
104
|
+
TmpDir.create('input.fasta') do |tmp_in, tmp_dir|
|
105
|
+
process_input(input, output, tmp_in)
|
106
|
+
run_mothur(tmp_dir, tmp_in)
|
107
|
+
tmp_out = Dir.glob("#{tmp_dir}/input.*.taxonomy").first
|
108
|
+
process_output(output, tmp_out)
|
109
|
+
end
|
110
|
+
end
|
111
|
+
end
|
112
|
+
|
113
|
+
private
|
114
|
+
|
115
|
+
# Check options.
|
116
|
+
def check_options
|
117
|
+
options_allowed(@options, :database, :taxonomy, :confidence, :cpus)
|
118
|
+
options_required(@options, :database, :taxonomy)
|
119
|
+
options_files_exist(@options, :database, :taxonomy)
|
120
|
+
options_assert(@options, ':confidence > 0')
|
121
|
+
options_assert(@options, ':confidence <= 100')
|
122
|
+
options_assert(@options, ':cpus >= 1')
|
123
|
+
options_assert(@options, ":cpus <= #{BioDSL::Config::CORES_MAX}")
|
124
|
+
|
125
|
+
defaults
|
126
|
+
end
|
127
|
+
|
128
|
+
# Set default options.
|
129
|
+
def defaults
|
130
|
+
@options[:confidence] ||= 80
|
131
|
+
@options[:cpus] ||= 1
|
132
|
+
end
|
133
|
+
|
134
|
+
# Process input data and save sequences to a temporary file for
|
135
|
+
# classifcation.
|
136
|
+
#
|
137
|
+
# @param input [Enumerator] Input stream.
|
138
|
+
# @param output [Enumerator::Yielder] Output stream.
|
139
|
+
# @param tmp_in [String] Path to temporary FASTA file.
|
140
|
+
def process_input(input, output, tmp_in)
|
141
|
+
BioDSL::Fasta.open(tmp_in, 'w') do |ios|
|
142
|
+
input.each_with_index do |record, i|
|
143
|
+
@status[:records_in] += 1
|
144
|
+
|
145
|
+
if record[:SEQ]
|
146
|
+
@status[:sequences_in] += 1
|
147
|
+
@status[:sequences_out] += 1
|
148
|
+
@status[:residues_in] += record[:SEQ].length
|
149
|
+
@status[:records_out] += record[:SEQ].length
|
150
|
+
seq_name = record[:SEQ_NAME] || i.to_s
|
151
|
+
|
152
|
+
entry = BioDSL::Seq.new(seq_name: seq_name, seq: record[:SEQ])
|
153
|
+
|
154
|
+
ios.puts entry.to_fasta
|
155
|
+
end
|
156
|
+
|
157
|
+
output << record
|
158
|
+
@status[:records_out] += 1
|
159
|
+
end
|
160
|
+
end
|
161
|
+
end
|
162
|
+
|
163
|
+
# Run Mothur using a system call.
|
164
|
+
#
|
165
|
+
# @param tmp_dir [String] Path to temporary dir.
|
166
|
+
# @param tmp_in [String] Path to input file.
|
167
|
+
#
|
168
|
+
# @raise [RunTimeError] If system call fails.
|
169
|
+
def run_mothur(tmp_dir, tmp_in)
|
170
|
+
cmd = <<-CMD.gsub(/^\s+\|/, '').delete("\n")
|
171
|
+
|mothur "#set.dir(input=#{tmp_dir});
|
172
|
+
|set.dir(output=#{tmp_dir});
|
173
|
+
|classify.seqs(fasta=#{tmp_in},
|
174
|
+
|reference=#{@options[:database]},
|
175
|
+
|taxonomy=#{@options[:taxonomy]},
|
176
|
+
|method=wang,
|
177
|
+
|processors=#{@options[:cpus]})"
|
178
|
+
CMD
|
179
|
+
|
180
|
+
BioDSL.verbose ? system(cmd) : system("#{cmd} > /dev/null 2>&1")
|
181
|
+
|
182
|
+
fail 'Mothur failed' unless $CHILD_STATUS.success?
|
183
|
+
end
|
184
|
+
|
185
|
+
# Parse mothur classfication output and emit to stream.
|
186
|
+
#
|
187
|
+
# @param output [Enumerator::Yielder] Output stream.
|
188
|
+
# @param tmp_out [String] Path to file with classfication result.
|
189
|
+
def process_output(output, tmp_out)
|
190
|
+
BioDSL::CSV.open(tmp_out) do |ios|
|
191
|
+
ios.each_hash do |new_record|
|
192
|
+
new_record[:SEQ_NAME] = new_record[:V0]
|
193
|
+
new_record[:TAXONOMY] = new_record[:V1]
|
194
|
+
new_record[:TAXONOMY].tr!('"', '')
|
195
|
+
new_record.delete(:V0)
|
196
|
+
new_record.delete(:V1)
|
197
|
+
new_record[:TAXONOMY] = confidence_filter(new_record)
|
198
|
+
new_record[:RECORD_TYPE] = 'taxonomy'
|
199
|
+
output << new_record
|
200
|
+
@status[:records_out] += 1
|
201
|
+
end
|
202
|
+
end
|
203
|
+
end
|
204
|
+
|
205
|
+
# Filter taxonomic leveles based on the confidence.
|
206
|
+
#
|
207
|
+
# @param record [Hash] BioDSL record with taxonomy.
|
208
|
+
#
|
209
|
+
# @return [String] Return taxonomic string.
|
210
|
+
def confidence_filter(record)
|
211
|
+
new_levels = []
|
212
|
+
|
213
|
+
record[:TAXONOMY].split(';').each do |level|
|
214
|
+
next unless level =~ /^([^(]+)\((\d+)\)$/
|
215
|
+
name = Regexp.last_match(1)
|
216
|
+
confidence = Regexp.last_match(2).to_i
|
217
|
+
|
218
|
+
if confidence >= @options[:confidence]
|
219
|
+
new_levels << "#{name}(#{confidence})"
|
220
|
+
end
|
221
|
+
end
|
222
|
+
|
223
|
+
new_levels.empty? ? 'Unclassified' : new_levels.join(';')
|
224
|
+
end
|
225
|
+
end
|
226
|
+
end
|