Egaz/Command/blastmatch.pm

filter: Statementstmt 25.1% 41 / 163 Branchbran 8.8% 3 / 34 Conditioncond n/a MC/DCmcdc n/a Subroutinesub 80.0% 12 / 15 Podpod 85.7% 6 / 7 totaltotal 28.3% 62 / 219 CC 33 SCAR 52.8
CC
33
Cov
22%
execute
420.0
__ANON__
90.0
string_to_set
6.0
CRAP 197.0
« Egaz/Command/exactmatch.pm ↑ Summary Egaz/Command/lastz.pm »
1 package App::Egaz::Command::blastmatch;
2 19 ▶use strict;
Subroutine BEGIN: called CC=1 SCAR=0.0
3 19 ▶use warnings;
Subroutine BEGIN: called CC=1 SCAR=0.0
4 19 ▶use autodie;
Subroutine BEGIN: called CC=1 SCAR=0.0
5  
6 19 ▶use MCE;
Subroutine BEGIN: called CC=1 SCAR=0.0
7 19 ▶use MCE::Flow;
Subroutine BEGIN: called CC=1 SCAR=0.0
8  
9 19 ▶use App::Egaz -command;
Subroutine BEGIN: called CC=1 SCAR=0.0
10 19 ▶use App::Egaz::Common;
Subroutine BEGIN: called CC=1 SCAR=0.0
11  
12 2 ▶sub abstract {
Subroutine abstract: called CC=1 SCAR=0.0
Pod documented
13 2 return 'matched positions by blastn in genome sequences';
14 }
15  
16 3 ▶sub opt_spec {
Subroutine opt_spec: called CC=1 SCAR=0.0
Pod documented
17 return (
18 3 [ "outfile|o=s", "Output filename. [stdout] for screen", { default => "stdout" }, ],
19 [ "perchr", "one (fake) runlist per chromosome", ],
20 [ "coverage|c=f", "coverage of identical matches", { default => 0.9 }, ],
21 [ "batch=i", "batch size of blast records", { default => 500000 }, ],
22 [ "parallel|p=i", "number of threads", { default => 2 }, ],
23 [ "verbose|v", "verbose mode", ],
24 { show_defaults => 1, }
25 );
26 }
27  
28 3 ▶sub usage_desc {
Subroutine usage_desc: called CC=1 SCAR=0.0
Pod documented
29 3 return "egaz blastmatch [options] <infile>";
30 }
31  
32 1 ▶sub description {
Subroutine description: called CC=1 SCAR=0.0
Pod documented
33 1 my $desc;
34 1 $desc .= ucfirst(abstract) . ".\n";
35 1 $desc .= <<'MARKDOWN';
36  
37 * <infile> is reports produced by `egaz blastn`
38 * <infile> can't be stdin
39 * --perchr produces fake runlists, there might be overlaps, e.g. I:17221-25234,21428-25459
40  
41 MARKDOWN
42  
43 1 return $desc;
44 }
45  
46 2 ▶sub validate_args {
Subroutine validate_args: called CC=5 SCAR=16.1
Pod documented
47 2 my ( $self, $opt, $args ) = @_;
48  
49 2 ▶ if ( @{$args} != 1 ) {
Branch: 1 branch
BranchTrueFalseTotal
if (@{$args;} != 1) 1 1 2
50 1 my $message = "This command need one input file.\n\tIt found";
51 1 $message .= sprintf " [%s]", $_ for @{$args};
52 1 $message .= ".\n";
53 1 $self->usage_error($message);
54 }
55 1 for ( @{$args} ) {
56 1 ▶ if ( !Path::Tiny::path($_)->is_file ) {
Branch: 1 branch
BranchTrueFalseTotal
unless (Path::Tiny::path($_)->is_file) 1 0 1
57 1 $self->usage_error("The input file [$_] doesn't exist.");
58 }
59 }
60  
61 }
62  
63 0 ▶sub execute {
Subroutine execute: not called CC=20 SCAR=60.4
Pod documented
64 0 my ( $self, $opt, $args ) = @_;
65  
66 #----------------------------#
67 # Parse reports
68 #----------------------------#
69 0 print STDERR "Parse reports\n";
70 0 ▶ my $worker = sub {
Subroutine __ANON__: not called CC=9 SCAR=45.0
71 0 my ( $self, $chunk_ref, $chunk_id ) = @_;
72  
73 0 my $wid = MCE->wid;
74 0 ▶ print STDERR "* Process task [$chunk_id] by worker #$wid\n" if $opt->{verbose};
Branch: 1 branch
BranchTrueFalseTotal
if $opt->{"verbose"} 0 0 0
75  
76 0 my @lines = @{$chunk_ref};
77 0 my %heads;
78 0 for my $line (@lines) {
79 0 ▶ next if $line =~ /^#/;
Branch: 1 branch
BranchTrueFalseTotal
if $line =~ /^#/ 0 0 0
80 0 chomp $line;
81  
82 # qseqid sseqid qstart qend sstart send qlen slen nident
83 0 my @fields = grep {defined} split /\s+/, $line;
84 0 ▶ if ( @fields != 9 ) {
Branch: 1 branch
BranchTrueFalseTotal
if (@fields != 9) 0 0 0
85 0 print "Fields error: $line\n";
86 0 next;
87 }
88  
89 0 my $query_name = $fields[0];
90 0 my $hit_name = $fields[1];
91  
92 0 my $query_length = $fields[6];
93 0 my $identical_match = $fields[8];
94  
95 0 my ( $h_start, $h_end ) = ( $fields[4], $fields[5] );
96 0 ▶ if ( $h_start > $h_end ) {
Branch: 1 branch
BranchTrueFalseTotal
if ($h_start > $h_end) 0 0 0
97 0 ( $h_start, $h_end ) = ( $h_end, $h_start );
98 }
99  
100 0 my $query_coverage = $identical_match / $query_length;
101  
102 0 ▶ if ( $query_coverage >= $opt->{coverage} ) {
Branch: 1 branch
BranchTrueFalseTotal
if ($query_coverage >= $opt->{"coverage"}) 0 0 0
103 0 my $head = "$hit_name(+):$h_start-$h_end";
104 0 $heads{$head}++;
105 }
106 }
107  
108 0 ▶ printf STDERR " " x 4 . "Gather %d pieces\n", scalar keys %heads if $opt->{verbose};
Branch: 1 branch
BranchTrueFalseTotal
if $opt->{"verbose"} 0 0 0
109 printf STDERR " " x 4 . "Last query name is %s\n", ( split /\s+/, $lines[-1] )[0]
110 0 ▶ if $opt->{verbose};
Branch: 1 branch
BranchTrueFalseTotal
if $opt->{"verbose"} 0 0 0
111 0 MCE->gather(%heads);
112 0 };
113  
114 MCE::Flow::init {
115 chunk_size => $opt->{batch},
116 max_workers => $opt->{parallel},
117 0 };
118 0 my %ranges = mce_flow_f $worker, $args->[0];
119 0 MCE::Flow::finish;
120  
121 #----------------------------#
122 # Remove ranges fully contained by others
123 #----------------------------#
124 0 print STDERR "Convert nodes to sets\n";
125 0 my $nodes_of_chr = {};
126 0 my %set_of;
127 0 for my $node ( keys %ranges ) {
128 0 my ( $chr, $set, undef ) = string_to_set($node);
129 0 ▶ if ( !exists $nodes_of_chr->{$chr} ) {
Branch: 1 branch
BranchTrueFalseTotal
unless (exists $nodes_of_chr->{$chr}) 0 0 0
130 0 $nodes_of_chr->{$chr} = [];
131 }
132 0 push @{ $nodes_of_chr->{$chr} }, $node;
133 0 $set_of{$node} = $set;
134 }
135  
136 0 print STDERR "Sort by positions at chromosomes\n";
137  
138 # end positions
139 0 for my $chr ( sort keys %{$nodes_of_chr} ) {
140 0 my @temps = map { $_->[0] }
141 0 sort { $a->[1] <=> $b->[1] }
142 0 map { /\:(?:\d+)\-(\d+)/; [ $_, $1 ] } @{ $nodes_of_chr->{$chr} };
143 0 $nodes_of_chr->{$chr} = \@temps;
144 }
145  
146 # start positions
147 0 for my $chr ( sort keys %{$nodes_of_chr} ) {
148 0 my @temps = map { $_->[0] }
149 0 sort { $a->[1] <=> $b->[1] }
150 0 map { /\:(\d+)/; [ $_, $1 ] } @{ $nodes_of_chr->{$chr} };
151 0 $nodes_of_chr->{$chr} = \@temps;
152 }
153  
154 0 print STDERR "Overlaps between ranges\n";
155 0 my %to_remove;
156 0 for my $chr ( sort keys %{$nodes_of_chr} ) {
157 0 my @nodes = @{ $nodes_of_chr->{$chr} };
158  
159 0 my $vicinity = 10;
160 0 for my $idx ( 0 .. $#nodes - $vicinity ) {
161  
162 0 for my $i ( 0 .. $vicinity - 1 ) {
163 0 for my $j ( $i .. $vicinity - 1 ) {
164 0 my $node_i = $nodes[ $idx + $i ];
165 0 my $set_i = $set_of{$node_i};
166  
167 0 my $node_j = $nodes[ $idx + $j ];
168 0 my $set_j = $set_of{$node_j};
169  
170 0 ▶ if ( $set_i->larger_than($set_j) ) {
Branch: 2 branches
BranchTrueFalseTotal
if ($set_i->larger_than($set_j)) { } 0 0 0
elsif ($set_j->larger_than($set_i)) { } 0 0 0
171 0 $to_remove{$node_j}++;
172 }
173 elsif ( $set_j->larger_than($set_i) ) {
174 0 $to_remove{$node_i}++;
175 }
176  
177 }
178 }
179 }
180 }
181  
182 0 printf STDERR " " x 4 . "Remove [%d] nested nodes\n", scalar keys %to_remove;
183 0 my @sorted;
184 0 for my $chr ( sort keys %{$nodes_of_chr} ) {
185 0 for my $node ( @{ $nodes_of_chr->{$chr} } ) {
186 0 ▶ if ( !exists $to_remove{$node} ) {
Branch: 1 branch
BranchTrueFalseTotal
unless (exists $to_remove{$node}) 0 0 0
187 0 push @sorted, $node;
188 }
189 }
190 }
191  
192 #----------------------------#
193 # Write outputs
194 #----------------------------#
195 0 my $out_fh;
196 0 ▶ if ( lc( $opt->{outfile} ) eq "stdout" ) {
Branch: 1 branch
BranchTrueFalseTotal
if (lc $opt->{'outfile'} eq 'stdout') { } 0 0 0
197 0 $out_fh = *STDOUT{IO};
198 }
199 else {
200 0 open $out_fh, ">", $opt->{outfile};
201 }
202  
203 0 ▶ if ( $opt->{perchr} ) {
Branch: 1 branch
BranchTrueFalseTotal
if ($opt->{'perchr'}) { } 0 0 0
204 0 my $runlist_of;
205 0 for my $node (@sorted) {
206 0 my ( $chr, $set, undef ) = string_to_set($node);
207 0 ▶ if ( !exists $runlist_of->{$chr} ) {
Branch: 1 branch
BranchTrueFalseTotal
unless (exists $runlist_of->{$chr}) 0 0 0
208 0 $runlist_of->{$chr} = [];
209 }
210 0 push @{ $runlist_of->{$chr} }, $set->runlist;
211 }
212  
213 0 for my $chr ( sort keys %{$runlist_of} ) {
214 0 my $range = "$chr:" . join( ",", @{ $runlist_of->{$chr} } );
215 0 print {$out_fh} $range . "\n";
216 }
217 }
218 else {
219 0 for my $node (@sorted) {
220 0 my ( $chr, $set, undef ) = string_to_set($node);
221 0 my $range = "$chr:" . $set->runlist;
222 0 print {$out_fh} $range . "\n";
223 }
224 }
225  
226 0 close $out_fh;
227 }
228  
229 0 ▶sub string_to_set {
Subroutine string_to_set: not called CC=2 SCAR=17.9
Pod undocumented
230 0 my $node = shift;
231  
232 0 my ( $chr_part, $runlist ) = split /:/, $node;
233  
234 0 my $head_qr = qr{
235 (?:(?P<name>[\w_]+)\.)?
236 (?P<chr_name>[\w-]+)
237 (?:\((?P<chr_strand>.+)\))?
238 }xi;
239 0 $chr_part =~ $head_qr;
240  
241 0 my $chr = $+{chr_name};
242 0 my $strand = "+";
243 0 ▶ if ( defined $+{chr_strand} ) {
Branch: 1 branch
BranchTrueFalseTotal
if (defined $+{"chr_strand"}) 0 0 0
244 0 $strand = $+{chr_strand};
245 }
246 0 my $set = AlignDB::IntSpan->new($runlist);
247  
248 0 return ( $chr, $set, $strand );
249 }
250  
251 19 1;
« Egaz/Command/exactmatch.pm ↑ Summary Egaz/Command/lastz.pm »