script/debrief-42.pl untested

filter: Statementstmt 0.0% 0 / 197 Branchbran 0.0% 0 / 56 Conditioncond 0.0% 0 / 62 MC/DCmcdc 0.0% 0 / 35 Subroutinesub 0.0% 0 / 16 Podpod 0.0% 0 / 2 totaltotal 0.0% 0 / 368 CC 3 SCAR 10.6
CC
3
Cov
0%
BEGIN
6.0
compute_percentage
6.0
BEGIN
2.0
CRAP 2.9
1 #!/usr/bin/env perl
2 # PODNAME: debrief-42.pl
3 # ABSTRACT: Summarize the results of a 42 metagenomic run
4 # CONTRIBUTOR: Mick VAN VLIERBERGHE <mvanvlierberghe@doct.ulg.ac.be>
5  
6 use autodie;
7 use Modern::Perl '2011';
8  
9 use Getopt::Euclid qw(:vars);
10  
11 use File::Basename;
12 use File::Find::Rule;
13 use List::AllUtils qw(uniq);
14 use Path::Class qw(dir file);
15 use Tie::IxHash;
16  
17 use Bio::MUST::Core;
18 use aliased 'Bio::MUST::Core::SeqId';
19 use aliased 'Bio::MUST::Core::IdList';
20 use aliased 'Bio::MUST::Core::Taxonomy';
21 use aliased 'Bio::MUST::Apps::Debrief42::TaxReport';
22  
23  
24 ## no critic (RequireLocalizedPunctuationVars)
25 BEGIN{
26 $ENV{Smart_Comments} = $ARGV_verbosity
27 ? join q{ }, map { '#' x (2 + $_) } 1..$ARGV_verbosity
28 : q{}
29 ;
30 }
31 use Smart::Comments -ENV;
32  
33  
34 die <<'EOT' if !$ARGV_taxdir && ($ARGV_seq_labeling || $ARGV_contam_labeling);
35 Missing required arguments:
36 --taxdir=<dir>
37 EOT
38  
39 # optionally build taxonomy object
40 my $tax;
41 if ($ARGV_taxdir) {
42 $tax = Taxonomy->new_from_cache( tax_dir => $ARGV_taxdir );
43 }
44  
45 # optionally build classifier objects
46 my ($seq_classifier, $contam_classifier);
47 if ($ARGV_seq_labeling) {
48 my $seq_labeling_list = IdList->load($ARGV_seq_labeling);
49 $seq_classifier = $tax->tax_labeler_from_list($seq_labeling_list);
50 }
51 if ($ARGV_contam_labeling) {
52 my $contam_labeling_list = IdList->load($ARGV_contam_labeling);
53 $contam_classifier = $tax->tax_labeler_from_list($contam_labeling_list);
54 }
55  
56 ### Processing: $ARGV_indir
57 my @tax_reports = File::Find::Rule
58 ->file()
59 ->name( qr{ $ARGV_in_strip \.tax-report\z}xmsi )
60 ->maxdepth(1)
61 ->in($ARGV_indir)
62 ;
63 my $ali_test = @tax_reports;
64 #### @tax_reports
65  
66 # TODO: rewrap long lines (>80 chars)
67 my %data_for;
68  
69 ### Processing tax-reports...
70 for my $report (@tax_reports) {
71 ### Processing: $report
72  
73 # Loading file as a TaxReport object
74 my $tax_report = TaxReport->new( file => file($report) );
75 #### $tax_report
76  
77 # Deriving MSA/OG basename
78 $report =~ s/$ARGV_in_strip//xms;
79 my ($ali) = fileparse($report, qr{\.[^.]*}xms);
80 #### $ali
81  
82 NEWSEQ:
83 while ( my $new_seq = $tax_report->next_seq ) {
84  
85 next NEWSEQ unless $new_seq->lca || $new_seq->lineage;
86 my $full_id = $new_seq->seq_id;
87 $full_id =~ s/(GC[AF])\s/$1_/xms;
88 my $seq_id = SeqId->new(full_id => $full_id);
89  
90 # TODO: check that literal strings are still up-to-date with current tax
91 my ($contam, $unclass, $unknown, $self) = (0, 0, 0, 0);
92 $unknown = 1 if map { $new_seq->lca =~ m/$_/xmsg } 'cellular\sorganisms', 'Eukaryota', 'Bacteria', 'Archaea' , 'unclassified sequences';
93 $unclass = 1 if $new_seq->lineage eq 'unclassified sequences';
94 #### $new_seq
95 #### $full_id
96 #### $seq_id
97 #### $unknown
98 #### $unclass
99  
100 unless ( $unknown or $unclass ) {
101  
102 if ($seq_classifier) {
103 #### new_seq: $full_id
104 #### affiliation: $new_seq->lineage
105 ($self, $contam) = $tax->eq_tax($full_id, $new_seq->lineage, $seq_classifier) ? (1, 0) : (0, 1);
106 #### $self
107 #### $contam
108 }
109 else {
110 $contam = 1 if defined $seq_id->tag && $seq_id->tag eq 'c';
111 $unclass = 1 if defined $seq_id->tag && $seq_id->tag eq 'u';
112 $self = 1 unless $contam || $unclass;
113 #### tag#: $seq_id->tag
114 }
115 }
116  
117 my $bank = $seq_id->full_org || $seq_id->taxon_id;
118 #### $bank
119  
120 # Fill up data hash...
121 $data_for{$bank}{seq}{phyla}{$new_seq->lca}++ if $contam;
122 $data_for{$bank}{lineages}{$new_seq->lineage}++ if $contam;
123  
124 $data_for{$bank}{seq}{c}++ if $contam;
125 $data_for{$bank}{seq}{nc}++ if $self;
126 $data_for{$bank}{seq}{uc}++ if $unclass;
127 $data_for{$bank}{seq}{uk}++ if $unknown;
128 $data_for{$bank}{seq}{total}++;
129 #### get: $data_for{$bank}{seq}
130  
131 $data_for{$bank}{ali}{c}{$ali} = 1 if $contam;
132 $data_for{$bank}{ali}{uc}{$ali} = 1 if $unclass;
133 $data_for{$bank}{ali}{uk}{$ali} = 1 if $unknown;
134 $data_for{$bank}{ali}{nc}{$ali} = 1 if $self;
135 $data_for{$bank}{ali}{total}{$ali} = 1;
136 }
137 }
138 #### %data_for
139  
140 # analyse contaminating phyla and try to remap them to a higher level...
141 # ... in order to alleviate contam reports
142 my %contam_data_for;
143 my @banks = keys %data_for;
144 PROCESS_DATA:
145 for my $bank ( @banks ) {
146  
147 my %lineages_for;
148 my $contam_lineage_count = $data_for{$bank}{lineages};
149  
150 for my $lineage ( keys %$contam_lineage_count ) {
151  
152 my $contam_taxon = $contam_classifier ? $contam_classifier->classify($lineage) : (split /;\s/xms, $lineage)[-1];
153 $contam_taxon ||= 'other taxa';
154 push @{ $lineages_for{$contam_taxon} }, $lineage for 1..$contam_lineage_count->{$lineage};
155 #### $lineage
156 #### classification: $contam_taxon
157 }
158 #### Considering: $bank
159 #### $contam_lineage_count
160 #### %lineages_for
161  
162 for my $contam_taxon ( keys %lineages_for ) {
163 my $seq_count = eval { scalar @{ $lineages_for{$contam_taxon} } } || 0;
164 $contam_data_for{$bank}{$contam_taxon} = $seq_count;
165 }
166 }
167 #### %contam_data_for
168  
169 # TODO: consolidate KRONA output with Physeter's subs
170  
171 my $dir = '.'; # default to working dir
172 if ($ARGV_outdir) {
173 $dir = dir($ARGV_outdir)->relative;
174 $dir->mkpath();
175 }
176  
177 ### Writing output TSV files for KRONA...
178 for my $bank (@banks) {
179 ( my $serie = $bank ) =~ s/\s/_/xmsg;
180 open my $lineout, '>', file($dir, $serie . $ARGV_in_strip . $ARGV_out_suffix . '.tsv');
181 say {$lineout} join "\t", $serie, 'domain;phylum;class;order;family;genus;species';
182 say {$lineout} join "\t", ( $data_for{$bank}{lineages}{$_}, $_ ) for keys %{ $data_for{$bank}{lineages} };
183 say {$lineout} join "\t", $data_for{$bank}{seq}{nc} || 0, 'SELF';
184 say {$lineout} join "\t", $data_for{$bank}{seq}{uk} || 0, 'unknown';
185 say {$lineout} join "\t", $data_for{$bank}{seq}{uc} || 0, 'unclassified';
186 }
187  
188 my @elected_lineages = uniq( sort map { keys %{ $contam_data_for{$_} } } @banks );
189 my @phyla = map { (split /;\s/xms)[-1] } @elected_lineages;
190 my @mod_phyla = map { $_ =~ s/\s/_/xmsg; $_ } @phyla;
191 #### @elected_lineages
192 #### @phyla
193  
194 ### Writing summary output files...
195 STATS_FILES:
196  
197 open my $out, '>', file($dir, 'per-genome' . $ARGV_in_strip . $ARGV_out_suffix . '.stats');
198 open my $out_sum, '>', file($dir, 'per-genome' . $ARGV_in_strip . $ARGV_out_suffix . '.stats-sum');
199  
200 say {$out} join "\t", 'bank', 'tested_genes', 'added_ali', 'clean_ali',
201 'contam_ali', 'completeness', 'added_seq', 'clean_seq', 'contam_seq',
202 'unclass_contam_seq', 'unknown_seq', @mod_phyla, 'foreign_phyla',
203 'contam_perc', 'class_contam_perc', 'unclass_contam_perc', 'unknown_perc'
204 ;
205  
206 say {$out_sum} join "\t", 'bank', 'tested_genes', 'added_ali', 'clean_ali',
207 'contam_ali', 'completeness', 'added_seq', 'clean_seq', 'contam_seq',
208 'unclass_contam_seq', 'unknown_seq', 'foreign_phyla', 'contam_perc',
209 'class_contam_perc', 'unclass_contam_perc', 'unknown_perc';
210  
211 my @ali_totals;
212 tie my %line_for, 'Tie::IxHash';
213 for my $bank (@banks) {
214  
215 # ali stats
216 my $ali_c = scalar keys %{ $data_for{$bank}{ali}{c} } // 0;
217 my $ali_nc = scalar keys %{ $data_for{$bank}{ali}{nc} } // 0;
218 my $ali_total = scalar keys %{ $data_for{$bank}{ali}{total} } // 0;
219 my $completeness = eval { ( $ali_nc / $ali_test ) * 100 };
220  
221 # seq stats
222 my $seq_c = $data_for{$bank}{seq}{c} // 0;
223 my $seq_nc = $data_for{$bank}{seq}{nc} // 0;
224 my $seq_total = $data_for{$bank}{seq}{total} // 0;
225  
226 # taxonomic data
227 my @phyla_data = map { $contam_data_for{$bank}{$_} // 0 } @elected_lineages;
228 my $foreign_phyla = scalar keys %{ $contam_data_for{$bank} } // 0;
229 my $unclassified = $data_for{$bank}{seq}{uc} // 0;
230 my $unknown = $data_for{$bank}{seq}{uk} // 0;
231  
232 my $all_c_p = eval { ( ($seq_c + $unclassified) * 100)/$seq_total } // 0;
233 my $contam_p = eval { ( $seq_c * 100)/$seq_total } // 0;
234 my $unclas_p = eval { ( $unclassified * 100)/$seq_total } // 0;
235 my $unknwn_p = eval { ( $unknown * 100)/$seq_total } // 0;
236  
237 $bank =~ s/\s/_/xms;
238  
239 $line_for{$bank} = [
240 $bank, $ali_test, $ali_total, $ali_nc, $ali_c, $completeness,
241 $seq_total, $seq_nc, $seq_c, $unclassified, $unknown,
242 @phyla_data, $foreign_phyla >= 0 ? $foreign_phyla : 0,
243 $all_c_p, $contam_p, $unclas_p, $unknwn_p
244 ];
245  
246 push @ali_totals, $ali_total;
247 }
248 #### %line_for
249  
250 my @sort_all_banks = sort { @{ $line_for{$b} }[-3] <=> @{ $line_for{$a} }[-3] } keys %line_for;
251 #### @sort_all_banks
252  
253 # Write file contents
254 say {$out} join "\t", @{ $line_for{$_} } for @sort_all_banks;
255 say {$out_sum} join "\t", @{ $line_for{$_} }[0..8,-7..-1] for @sort_all_banks;
256  
257 ### Done!
258  
259  
260 ##################################### SUBS #####################################
261  
262 sub compute_percentage {
263 my $array = shift;
264 my $total = shift;
265  
266 my @results;
267 my $percentage;
268 CALC:
269 for my $value (@$array) {
270 if ($value == 0) {
271 $percentage = 0;
272 push @results, '-/-';
273 next CALC;
274 }
275 else {
276 $percentage = $value / $total;
277 # $percentage = $value * 100 / $total;
278 }
279 $percentage = sprintf("%.2f", $percentage);
280 push @results, $value . '/' . $percentage;
281 }
282 return \@results;
283 }
284  
285 # for testing:
286 # perl -Ilib bin/debrief-42.pl --indir=xtest/tax_reports/ \
287 # --in-strip=-42-camera-megan99-tf --taxdir=../Bio-MUST-Core/test/taxdump \
288 # --seq_labeling=xtest/seq-labels.idl --contam_labeling=xtest/contam-labels.idl \
289 # --outdir=dbout
290  
291 __END__