Mercurial > repos > brasset_jensen > srnapipe

comparison bin/sRNAPipe.pl @ 1:1df6aaac800e draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Deleted selected files
author brasset_jensen
date Wed, 13 Dec 2017 10:40:50 -0500
parents
children b80aa719069f
comparison
equal deleted inserted replaced
0:e4e71401c577 1:1df6aaac800e
1 #!/usr/bin/perl
2 use strict;
3 use warnings;
4 use Getopt::Long;
5 use Parallel::ForkManager;
6 use File::Basename;
7 use File::Copy::Recursive qw( dircopy );
8 use POSIX;
9 use FindBin;
10 use lib $FindBin::Bin;
11 use resize qw ( size_distribution );
12 use subgroups qw (subgroups );
13 use ppp qw ( ping_pong_partners );
14 use Rcall qw (pie_chart bg_to_png );
15 use align qw ( to_build get_unique sam_count sam_count_mis sam_sorted_bam rpms_rpkm BWA_call get_fastq_seq extract_sam sam_to_bam_bg );
16 use html qw ( main_page details_pages menu_page ppp_page );
17 use File::Copy;
18
19 my ( @fastq, @fastq_n, $dir, $min, $max, $mis, $misTE, $help, $Pcheck, $mapnumf, $html_out);
20 my ( $ref, $tRNAs, $rRNAs, $snRNAs, $miRNAs, $transcripts, $TE );
21 my ( $si_min, $si_max, $pi_min, $pi_max );
22 my ( $build_index, $build_tRNAs, $build_rRNAs, $build_snRNAs, $build_miRNAs, $build_transcripts, $build_TE );
23 my $max_procs = 8;
24
25 ( $build_index, $build_tRNAs, $build_rRNAs, $build_snRNAs, $build_miRNAs, $build_transcripts, $build_TE ) = (0,0,0,0,0,0,0);
26 ( $min, $max, $mis, $misTE, $si_min, $si_max, $pi_min, $pi_max, $dir ) = ( 18, 29, 0, 3, 21, 21, 23, 29 );
27 $Pcheck ='true';
28
29 GetOptions (
30 "fastq=s" => \@fastq,
31 "fastq_n=s" => \@fastq_n,
32 "dir=s" => \$dir,
33 "min:i" => \$min,
34 "max:i" => \$max,
35 "si_min:i" => \$si_min,
36 "si_max:i" => \$si_max,
37 "pi_min:i" => \$pi_min,
38 "pi_max:i" => \$pi_max,
39 "mis:i" => \$mis,
40 "misTE:i" => \$misTE,
41 "html:s" => \$html_out,
42 "PPPon:s" => \$Pcheck,
43 "help" => \$help,
44 "ref:s" => \$ref,
45 "tRNAs:s" => \$tRNAs,
46 "rRNAs:s" => \$rRNAs,
47 "snRNAs:s" => \$snRNAs,
48 "miRNAs:s" => \$miRNAs,
49 "transcripts:s" => \$transcripts,
50 "TE:s" => \$TE,
51 "build_index" => \$build_index,
52 "build_tRNAs" => \$build_tRNAs,
53 "build_snRNAs" => \$build_snRNAs,
54 "build_miRNAs" => \$build_miRNAs,
55 "build_transcripts" => \$build_transcripts,
56 "build_rRNAs" => \$build_rRNAs,
57 "build_TE" => \$build_TE
58 );
59
60 my $fq_collection = 'fastq_dir/';
61 mkdir $dir; mkdir $fq_collection;
62 $dir = $dir.'/' unless $dir =~ /\/$/;
63 mkdir $dir.'/css';mkdir $dir.'/js';
64 dircopy( $FindBin::Bin.'/css', $dir.'/css' );
65 dircopy( $FindBin::Bin.'/js', $dir.'/js' );
66
67 my $file = $dir.'report.txt';
68 open my $report, '>', $file or die "Cannot open $file $!\n";
69
70 my @toBuild = ( [$build_index, \$ref], [$build_tRNAs, \$tRNAs], [$build_rRNAs, \$rRNAs], [$build_snRNAs, \$snRNAs], [$build_miRNAs, \$miRNAs], [$build_transcripts, \$transcripts], [$build_TE, \$TE] );
71 to_build ( \@toBuild, $report, $dir );
72
73 my $proc_child = ceil($max_procs / scalar(@fastq));
74 my $proc_grand_child = ceil($proc_child/4);
75 my $pm = Parallel::ForkManager->new($max_procs);
76 my $pm2 = Parallel::ForkManager->new($proc_grand_child);
77
78 $pm->run_on_finish( sub {
79 my ($pid, $exit_code, $ident) = @_;
80 print $report "Fastq fork $ident just finished ".
81 "with PID $pid and exit code: $exit_code\n";
82 die "Something went wrong!\n" if $exit_code != 0;
83 });
84 $pm->run_on_start( sub {
85 my ($pid,$ident)=@_;
86 print $report "Fastq fork : $ident started, pid: $pid\n";
87 });
88 $pm2->run_on_finish( sub {
89 my ($pid, $exit_code, $ident) = @_;
90 print $report "** Subgroup fork $ident just finished ".
91 "with PID $pid and exit code: $exit_code\n";
92 die "Something went wrong!\n" if $exit_code != 0;
93 });
94 $pm2->run_on_start( sub {
95 my ($pid,$ident)=@_;
96 print $report "** Subgroup fork $ident started, pid: $pid\n";
97 });
98
99
100 foreach my $child ( 0 .. $#fastq )
101 {
102 my @suffix = ('.fastq', '.fastq.gz,', '.fq', '.fq.gz', 'ref', '.dat', '.fa','.fas','.fasta', '.txt');
103 my ( $name, $path, $suffix ) = fileparse( $fastq[$child], @suffix );
104 my ( $ref_name, $ref_path, $ref_suffix ) = fileparse( $ref, @suffix );
105 my ( $TE_name, $TE_path, $TE_suffix ) = fileparse( $TE, @suffix );
106 my ( $ex_name, $ex_path, $ex_suffix ) = fileparse( $transcripts, @suffix );
107
108 $pm->start($fastq[$child]) and next;
109
110 my $dir_fq = $dir.$fastq_n[$child].'/';
111 mkdir $dir_fq;
112
113 my $gen_dir = $dir_fq.'genome/';
114 mkdir $gen_dir;
115
116 my $size_dir = $dir_fq.'size/';
117 mkdir $size_dir;
118
119 my $fastq_resized = $dir_fq.$name.'_'.$min.'-'.$max.'.fastq';
120 size_distribution ( $fastq[$child], $fastq_resized, $size_dir, $min, $max );
121
122 my $sam_genome = $gen_dir.$fastq_n[$child].'_'.$min.'-'.$max.'.sam';
123 my $sam_genome_unique = $gen_dir.$fastq_n[$child].'_'.$min.'-'.$max.'_unique.sam';
124 my $fastq_prefix = $gen_dir.$fastq_n[$child].'_'.$min.'-'.$max;
125
126 BWA_call ( $ref, $fastq_resized, $sam_genome, $mis, $proc_child, $report );
127 my ( $fai_ref_hashP, $ma, $ma_uni ) = get_unique ( $sam_genome, $sam_genome_unique, $gen_dir, $fq_collection.$fastq_n[$child], 1, $report );
128
129 die "No Reads mapped on the genome reference!\n" if $ma == 0;
130 my $scale = 1000000 / $ma;
131 sam_to_bam_bg ( $sam_genome_unique, $scale, $proc_child );
132 sam_to_bam_bg ( $sam_genome, $scale, $proc_child );
133
134 my $Gviz_dir = $gen_dir.'Gviz/';
135 my $fai_file = $gen_dir.'fai';
136 mkdir $Gviz_dir;
137 my $Gviz_dir_rand = $Gviz_dir.'rand/';
138 mkdir $Gviz_dir_rand;
139 my $Gviz_dir_uni = $Gviz_dir.'unique/';
140 mkdir $Gviz_dir_uni;
141
142 open my $gfai, '>', $fai_file;
143 foreach my $k ( sort keys %{$fai_ref_hashP} )
144 {
145 print $gfai "$k\t$fai_ref_hashP->{$k}\n";
146 }
147 close $gfai;
148 bg_to_png ( $fai_file, $fastq_prefix.'_unique_plus.bedgraph', $fastq_prefix.'_unique_minus.bedgraph', $Gviz_dir_uni, 'Mb' );
149 bg_to_png ( $fai_file, $fastq_prefix.'_plus.bedgraph', $fastq_prefix.'_minus.bedgraph', $Gviz_dir_rand, 'Mb' );
150
151 my $group_dir = $dir_fq.'subgroups/';
152 my $fastq_uni = $fq_collection.$fastq_n[$child].'_unique_mappers.fastq';
153 my $fastq_all = $fq_collection.$fastq_n[$child].'_all_mappers.fastq';
154 my ($bo, $mi, $pi) = subgroups ( $fastq_all, $group_dir, $mis, $misTE, $proc_child, $tRNAs, $rRNAs, $snRNAs, $miRNAs, $transcripts, $TE, $si_min, $si_max, $pi_min, $pi_max, $report);
155
156 pie_chart($group_dir);
157
158 open (my $dupnum, $gen_dir.'dup_mapnum.txt') || die "cannot open dup_mapnum.txt $!";
159 my %dupnum_genome;
160 my $header = <$dupnum>;
161 while (<$dupnum>)
162 {
163 chomp $_;
164 my @dupline = split /\t/, $_;
165 $dupnum_genome{$dupline[0]} = [$dupline[1], $dupline[2]];
166 }
167 close $dupnum;
168
169 my $mi_sam = $group_dir.'miRNAs.sam';
170 mkdir $group_dir.'miRNAs/';
171 my $mi_count_file = $group_dir.'miRNAs/miRNAs_reads_counts.txt';
172 my ( $mi_count, $mi_ref_size ) = sam_count ( $mi_sam );
173
174 rpms_rpkm( $mi_count, $mi_ref_size, $ma, $mi_count_file, $pi, $mi, $bo );
175
176 my ( $sam_transcripts, $sam_TEs ) = ( $group_dir.'transcripts.sam', $group_dir.'TEs.sam' );
177 my @types = ($group_dir.'bonafide_reads.fastq', $group_dir.'miRNAs.fastq', $group_dir.'siRNAs.fastq', $group_dir.'piRNAs.fastq' );
178 my @types_names = ('bonafide_reads', 'miRNAs', 'siRNAs', 'piRNAs');
179 foreach my $grand_child ( 0 .. $#types )
180 {
181 my $type_dir = $group_dir.$types_names[$grand_child].'/';
182 my $type_prefix = $types_names[$grand_child].'-';
183 mkdir $type_dir;
184 $pm2->start($types[$grand_child]) and next;
185 my ( $type_sam_genome, $type_sam_TEs, $type_sam_transcripts ) = ( $type_dir.$type_prefix.'genome.sam', $type_dir.$type_prefix.'TEs.sam', $type_dir.$type_prefix.'transcripts.sam' );
186 my ( $type_sam_uni_genome, $type_sam_uni_TEs, $type_sam_uni_transcripts ) = ( $type_dir.$type_prefix.'genome_unique.sam', $type_dir.$type_prefix.'TEs_unique.sam', $type_dir.$type_prefix.'transcripts_unique.sam' );
187 my ( $type_uni_genome_fastq, $type_uni_TEs_fastq, $type_uni_transcripts_fastq ) = ( $fq_collection.$fastq_n[$child].'-'.$type_prefix.'genome_uni.fastq', $fq_collection.$fastq_n[$child].'-'.$type_prefix.'TEs_uni.fastq', $fq_collection.$fastq_n[$child].'-'.$type_prefix.'transcripts_uni.fastq');
188 my ( $type_genome_fastq, $type_TEs_fastq, $type_transcripts_fastq ) = ( $fq_collection.$fastq_n[$child].'-'.$type_prefix.'genome.fastq', $fq_collection.$fastq_n[$child].'-'.$type_prefix.'TEs.fastq', $fq_collection.$fastq_n[$child].'-'.$type_prefix.'transcripts.fastq');
189 my $type_sequence_hashP = get_fastq_seq ( $types[$grand_child] );
190
191 if ( $grand_child == 1 )
192 {
193 BWA_call ( $TE, $types[$grand_child], $type_sam_TEs, $misTE, $proc_child, $report );
194 BWA_call ( $transcripts, $types[$grand_child], $type_sam_transcripts, $mis, $proc_child, $report );
195 BWA_call ( $ref, $types[$grand_child], $type_sam_genome, $mis, $proc_child, $report );
196 extract_sam ( undef, $type_sam_TEs, $type_sam_TEs, $type_sam_uni_TEs, $type_uni_TEs_fastq, $type_uni_TEs_fastq );
197 extract_sam ( undef, $type_sam_transcripts, $type_sam_transcripts, $type_sam_uni_transcripts, $type_transcripts_fastq, $type_uni_transcripts_fastq );
198 extract_sam ( undef, $type_sam_genome, $type_sam_genome, $type_sam_uni_genome, $type_genome_fastq, $type_uni_genome_fastq );
199 }
200 else
201 {
202 extract_sam ( $type_sequence_hashP, $sam_TEs, $type_sam_TEs, $type_sam_uni_TEs, $type_TEs_fastq, $type_uni_TEs_fastq );
203 extract_sam ( $type_sequence_hashP, $sam_transcripts, $type_sam_transcripts, $type_sam_uni_transcripts, $type_transcripts_fastq, $type_uni_transcripts_fastq );
204 extract_sam ( $type_sequence_hashP, $sam_genome, $type_sam_genome, $type_sam_uni_genome, $type_genome_fastq, $type_uni_genome_fastq );
205 }
206
207 my $ex_count_file = $type_dir.$type_prefix.'transcripts_reads_counts.txt';
208 my ( $ex_count, $ex_ref_size ) = sam_count ( $type_sam_transcripts );
209 rpms_rpkm( $ex_count, $ex_ref_size, $ma, $ex_count_file, $pi, $mi, $bo );
210
211 my ( $TEs_count, $TEs_ref_size, $TEs_count_NoM, $TEs_count_M ) = sam_count_mis ( $type_sam_TEs );
212 my $TEs_count_file = $type_dir.$type_prefix.'TEs_reads_counts.txt';
213 my $TEs_count_file_M = $type_dir.$type_prefix.'TEs_reads_counts_mismatches.txt';
214 my $TEs_count_file_noM = $type_dir.$type_prefix.'TEs_reads_counts_nomismatches.txt';
215 rpms_rpkm( $TEs_count, $TEs_ref_size, $ma, $TEs_count_file, $pi, $mi, $bo );
216 rpms_rpkm( $TEs_count_NoM, $TEs_ref_size, $ma, $TEs_count_file_noM, $pi, $mi, $bo );
217 rpms_rpkm( $TEs_count_M, $TEs_ref_size, $ma, $TEs_count_file_M, $pi, $mi, $bo );
218
219 sam_to_bam_bg ( $type_sam_TEs, $scale, $grand_child );
220 sam_sorted_bam ( $type_sam_transcripts, $grand_child ); sam_sorted_bam ( $type_sam_uni_transcripts, $grand_child );
221 sam_sorted_bam ( $type_sam_uni_TEs, $grand_child );
222
223 my $Gviz_TEs = $type_dir.'Gviz_TEs/';
224 mkdir $Gviz_TEs;
225 bg_to_png ( $group_dir.'TEs.fai', $type_dir.$type_prefix.'TEs_plus.bedgraph', $type_dir.$type_prefix.'TEs_minus.bedgraph', $Gviz_TEs, 'Kb' );
226
227 my $Gviz_genome= $type_dir.'Gviz_genome/';
228 my $Gviz_genome_rand = $Gviz_genome.'rand/';
229 my $Gviz_genome_uni = $Gviz_genome.'unique/';
230 mkdir $Gviz_genome; mkdir $Gviz_genome_uni; mkdir $Gviz_genome_rand;
231
232 sam_to_bam_bg ( $type_sam_genome, $scale, $grand_child );
233 sam_to_bam_bg ( $type_sam_uni_genome, $scale, $grand_child );
234
235 bg_to_png ( $fai_file, $type_dir.$type_prefix.'genome_unique_plus.bedgraph', $type_dir.$type_prefix.'genome_unique_minus.bedgraph', $Gviz_genome_uni, 'Mb' );
236 bg_to_png ( $fai_file, $type_dir.$type_prefix.'genome_plus.bedgraph', $type_dir.$type_prefix.'genome_minus.bedgraph', $Gviz_genome_rand, 'Mb' );
237
238 #HTML Details
239 my $prefix_details_pages = $dir.$fastq_n[$child].'-'.$types_names[$grand_child];
240 details_pages ( $type_dir, $prefix_details_pages, \@fastq_n, $fastq_n[$child], $misTE, $dir, $Pcheck );
241
242 $pm2->finish();
243 }
244 $pm2->wait_all_children;
245
246 if ( $Pcheck eq 'true' )
247 {
248 my $ppp = $group_dir.'PPPartners/'; mkdir $ppp;
249 print $report "ping_pong_partners $group_dir/piRNAs/TEs.sam $ppp\n";
250 ping_pong_partners ( $group_dir.'TEs.fai', $group_dir.'piRNAs/piRNAs-TEs_sorted.bam', $ppp, $pi_min );
251 my $ppp_page = $dir.$fastq_n[$child].'-piRNAs-PPP.html';
252 ppp_page ( $group_dir, $ppp_page, \@fastq_n, $fastq_n[$child], $ppp, $dir );
253 }
254
255 #HTML Main Webpage
256 my $index_page = $dir.$fastq_n[$child].'.html';
257 main_page ( $gen_dir, $index_page, \@fastq_n, $fastq_n[$child], $ma, $ma_uni, $dir );
258 copy ($index_page, $html_out) if $child == 0;
259 #HTML Menu
260 my $menu_page = $dir.$fastq_n[$child].'-sub.html';
261 menu_page ( $group_dir, $menu_page, \@fastq_n, $fastq_n[$child], $min, $max, $si_min, $si_max, $pi_min, $pi_max, $dir );
262 unlink glob "$group_dir*.sam"; unlink glob "$group_dir*.fastq";
263 $pm->finish(); # pass an exit code to finish
264 }
265 $pm->wait_all_children;
266 unlink glob $dir."dataset_*symlink.fa*";
267 print $report "Job done!\n";
268 close $report;