7
|
1 #!/usr/bin/perl
|
|
2 #V1.0.1 added log, option parameters
|
|
3 use strict;
|
|
4 use warnings;
|
|
5 use Getopt::Long;
|
|
6
|
|
7 my $read1_file;
|
|
8 my $read2_file;
|
|
9 my $log_file;
|
|
10 my $output1_file;
|
|
11 my $output2_file;
|
|
12
|
|
13 my $TYPE="sanger";
|
|
14 my $MIN_LENGTH=30;
|
|
15 my $MIN_QUALITY=30;
|
|
16
|
|
17 my $VERBOSE = "OFF";
|
|
18
|
|
19 GetOptions (
|
|
20 "read1_file=s" => \$read1_file,
|
|
21 "read2_file=s" => \$read2_file,
|
|
22 "log_file=s" => \$log_file,
|
|
23 "output1_file=s" => \$output1_file,
|
|
24 "output2_file=s" => \$output2_file,
|
|
25 "type=s" => \$TYPE,
|
|
26 "min_length=i" => \$MIN_LENGTH,
|
|
27 "min_quality=i" => \$MIN_QUALITY,
|
|
28 "verbose=s" => \$VERBOSE
|
|
29 ) or die("Error in command line arguments\n");
|
|
30
|
|
31
|
|
32 my $nb_read1=0;
|
|
33 my $nb_base_read1=0;
|
|
34 my $nb_read2=0;
|
|
35 my $nb_base_read2=0;
|
|
36
|
|
37 my $nb_read1_t=0;
|
|
38 my $nb_base_read1_t=0;
|
|
39 my $nb_read2_t=0;
|
|
40 my $nb_base_read2_t=0;
|
|
41
|
|
42 my $nb_base_current_t=0;
|
|
43
|
|
44
|
|
45 open(READ1, $read1_file) or die ("Can't open $read1_file\n");
|
|
46 open(READ2, $read2_file) or die ("Can't open $read2_file\n");
|
|
47 open(OUT1, ">$output1_file") or die ("Can't open $output1_file\n");
|
|
48 open(OUT2, ">$output2_file") or die ("Can't open $output2_file\n");
|
|
49
|
|
50 my $error1=0;
|
|
51 my $error2=0;
|
|
52 my $error3=0;
|
|
53 my $error4=0;
|
|
54 my $error5=0;
|
|
55 my $error6=0;
|
|
56 my $error7=0;
|
|
57 my $error8=0;
|
|
58 my $error9=0;
|
|
59 my $error10=0;
|
|
60
|
|
61 while (my $ligne1_r1 =<READ1>){
|
|
62 my $ligne2_r1 =<READ1>;
|
|
63 my $ligne3_r1 =<READ1>;
|
|
64 my $ligne4_r1 =<READ1>;
|
|
65 my $ligne1_r2 =<READ2>;
|
|
66 my $ligne2_r2 =<READ2>;
|
|
67 my $ligne3_r2 =<READ2>;
|
|
68 my $ligne4_r2 =<READ2>;
|
|
69
|
|
70 $nb_read1++;
|
|
71 $nb_read2++;
|
|
72
|
|
73 #@ 1 sec
|
|
74 if ((!$ligne1_r1)||(!$ligne2_r1)||(!$ligne3_r1)||(!$ligne4_r1)||(!$ligne1_r2)||(!$ligne2_r2)||(!$ligne3_r2)||(!$ligne4_r2)){
|
|
75 if ($VERBOSE eq "ON"){
|
|
76 print "Error in file format";
|
|
77 if ($ligne1_r1){print $ligne1_r1;}
|
|
78 if ($ligne2_r1){print $ligne2_r1;}
|
|
79 if ($ligne3_r1){print $ligne3_r1;}
|
|
80 if ($ligne4_r1){print $ligne4_r1;}
|
|
81 if ($ligne1_r2){print $ligne1_r2;}
|
|
82 if ($ligne2_r2){print $ligne2_r2;}
|
|
83 if ($ligne3_r2){print $ligne3_r2;}
|
|
84 if ($ligne4_r2){print $ligne4_r2;}
|
|
85 print "\n";
|
|
86 }
|
|
87 $error1++;
|
|
88 }
|
|
89 elsif(($ligne1_r1 !~/^\@/)||($ligne1_r2 !~/^\@/)||($ligne3_r1 !~/^\+/)||($ligne3_r2 !~/^\+/)){
|
|
90 if ($VERBOSE eq "ON"){
|
|
91 print "Error in header : format\n";
|
|
92 print $ligne1_r1;
|
|
93 print $ligne2_r1;
|
|
94 print $ligne3_r1;
|
|
95 print $ligne4_r1;
|
|
96 print $ligne1_r2;
|
|
97 print $ligne2_r2;
|
|
98 print $ligne3_r2;
|
|
99 print $ligne4_r2;
|
|
100 print "\n";
|
|
101 }
|
|
102 $error2++;
|
|
103 }
|
|
104 #@ 1 - 2 sec
|
|
105 else {
|
|
106
|
|
107 my $length_seq1 = length($ligne2_r1);
|
|
108 my $length_qual1 =length($ligne4_r1);
|
|
109 my $seq1;
|
|
110 my $qual1;
|
|
111
|
|
112 my $length_seq2 = length($ligne2_r2);
|
|
113 my $length_qual2 =length($ligne4_r2);
|
|
114 my $seq2;
|
|
115 my $qual2;
|
|
116 my $header1="";
|
|
117 my $header2="";
|
|
118 my $repheader1="";
|
|
119 my $repheader2="";
|
|
120
|
|
121
|
|
122 if ($ligne1_r1 =~/^\@(.*?)\#/){
|
|
123 $header1 = $1;
|
|
124 }
|
|
125
|
|
126 if ($ligne3_r1 =~/^\+(.*?)\#/){
|
|
127 $repheader1 = $1;
|
|
128 }
|
|
129
|
|
130 if ($ligne1_r2 =~/^\@(.*?)\#/){
|
|
131 $header2 = $1;
|
|
132 }
|
|
133
|
|
134 if ($ligne3_r2 =~/^\+(.*?)\#/){
|
|
135 $repheader2 = $1;
|
|
136 }
|
|
137 #@ 2 sec
|
|
138
|
|
139 ### Verification de la coherence sequence /qualité @ 1 sec
|
|
140 if (($TYPE eq "illumina")&&((!$header1)||(!$header2)||(!$repheader1)||(!$repheader2))){
|
|
141 if ($VERBOSE eq "ON"){
|
|
142 print "Error in header : empty\n";
|
|
143 print $ligne1_r1;
|
|
144 print $ligne2_r1;
|
|
145 print $ligne3_r1;
|
|
146 print $ligne4_r1;
|
|
147 print $ligne1_r2;
|
|
148 print $ligne2_r2;
|
|
149 print $ligne3_r2;
|
|
150 print $ligne4_r2;
|
|
151 print "\n";
|
|
152 }
|
|
153 $error3++;
|
|
154 }
|
|
155 elsif (($TYPE eq "sanger")&&((!$header1)||(!$header2))){
|
|
156 if ($VERBOSE eq "ON"){
|
|
157 print "Error in header refgsd : empty\n";
|
|
158 print $ligne1_r1;
|
|
159 print $ligne2_r1;
|
|
160 print $ligne3_r1;
|
|
161 print $ligne4_r1;
|
|
162 print $ligne1_r2;
|
|
163 print $ligne2_r2;
|
|
164 print $ligne3_r2;
|
|
165 print $ligne4_r2;
|
|
166 print "\n";
|
|
167 }
|
|
168 $error3++;
|
|
169 }
|
|
170 elsif (($TYPE eq "illumina")&&(($header1 ne $repheader1)||($header2 ne $repheader2)||($header1 ne $header2))){
|
|
171 if ($VERBOSE eq "ON"){
|
|
172 print "Error in header : different\n";
|
|
173 print $ligne1_r1;
|
|
174 print $ligne2_r1;
|
|
175 print $ligne3_r1;
|
|
176 print $ligne4_r1;
|
|
177 print $ligne1_r2;
|
|
178 print $ligne2_r2;
|
|
179 print $ligne3_r2;
|
|
180 print $ligne4_r2;
|
|
181 print "\n";
|
|
182 }
|
|
183 $error4++;
|
|
184 }
|
|
185 elsif (($TYPE eq "sanger")&&($header1 ne $header2)){
|
|
186 if ($VERBOSE eq "ON"){
|
|
187 print "Error in header : different\n";
|
|
188 print $ligne1_r1;
|
|
189 print $ligne2_r1;
|
|
190 print $ligne3_r1;
|
|
191 print $ligne4_r1;
|
|
192 print $ligne1_r2;
|
|
193 print $ligne2_r2;
|
|
194 print $ligne3_r2;
|
|
195 print $ligne4_r2;
|
|
196 print "\n";
|
|
197 }
|
|
198 $error4++;
|
|
199 }
|
|
200 elsif (($length_seq1 != $length_qual1)||($length_seq2 != $length_qual2)){
|
|
201 if ($VERBOSE eq "ON"){
|
|
202 print "Error in seq/qual length\n";
|
|
203 print $ligne1_r1;
|
|
204 print $ligne2_r1;
|
|
205 print $ligne3_r1;
|
|
206 print $ligne4_r1;
|
|
207 print $ligne1_r2;
|
|
208 print $ligne2_r2;
|
|
209 print $ligne3_r2;
|
|
210 print $ligne4_r2;
|
|
211 print "\n";
|
|
212 }
|
|
213 $error5++;
|
|
214 }
|
|
215 #@ 1 - 2 sec
|
|
216 else {
|
|
217 ### Parsing sequence & qualité
|
|
218 if ($ligne2_r1 =~ /^([ATGCNX]+)\s*$/i){
|
|
219 $seq1 = $1;
|
|
220 $nb_base_read1 += length($seq1);
|
|
221 }
|
|
222 if ($ligne2_r2 =~ /^([ATGCNX]+)\s*$/i){
|
|
223 $seq2 = $1;
|
|
224 $nb_base_read2 += length($seq2);
|
|
225 }
|
|
226 if ($ligne4_r1 =~ /^(.*)\s*$/i){
|
|
227 $qual1 = $1;
|
|
228 }
|
|
229 if ($ligne4_r2 =~ /^(.*)\s*$/i){
|
|
230 $qual2 = $1;
|
|
231 }
|
|
232 #@ 2 sec
|
|
233 ### Verification du parsing et de la coherence sequence /qualité (n°2)
|
|
234 if ((!$seq1)||(!$seq2)||(!$qual1)||(!$qual2)){
|
|
235 if ($VERBOSE eq "ON"){
|
|
236 print "Error parsing seq / quality \n";
|
|
237 print $ligne1_r1;
|
|
238 print $ligne2_r1;
|
|
239 print $ligne3_r1;
|
|
240 print $ligne4_r1;
|
|
241 print $ligne1_r2;
|
|
242 print $ligne2_r2;
|
|
243 print $ligne3_r2;
|
|
244 print $ligne4_r2;
|
|
245 print "\n";
|
|
246 }
|
|
247 $error6++;
|
|
248 }
|
|
249 elsif ((length($seq1) != length($qual1))||(length($seq2) != length($qual2))){
|
|
250 if ($VERBOSE eq "ON"){
|
|
251 print "Error in seq/qual length after parsing\n";
|
|
252 print $ligne1_r1;
|
|
253 print $ligne2_r1;
|
|
254 print $ligne3_r1;
|
|
255 print $ligne4_r1;
|
|
256 print $ligne1_r2;
|
|
257 print $ligne2_r2;
|
|
258 print $ligne3_r2;
|
|
259 print $ligne4_r2;
|
|
260 print "\n";
|
|
261 }
|
|
262 $error7++;
|
|
263 }
|
|
264 #@ <1 sec
|
|
265 else {
|
|
266 my $fastq_lines_r1="";
|
|
267 my $fastq_lines_r2="";
|
|
268 my $nb_base_current_read1_t = 0;
|
|
269 my $nb_base_current_read2_t = 0;
|
|
270
|
|
271 $fastq_lines_r1 = &grooming_and_trimming($ligne1_r1,$seq1,$qual1);
|
|
272 $nb_base_current_read1_t = $nb_base_current_t;
|
|
273 if ($fastq_lines_r1){
|
|
274 $fastq_lines_r2 = &grooming_and_trimming($ligne1_r2,$seq2,$qual2);
|
|
275 $nb_base_current_read2_t = $nb_base_current_t;
|
|
276 }
|
|
277 if ($fastq_lines_r2){
|
|
278 print OUT1 $fastq_lines_r1;
|
|
279 print OUT2 $fastq_lines_r2;
|
|
280
|
|
281 $nb_read1_t++;
|
|
282 $nb_read2_t++;
|
|
283 $nb_base_read1_t += $nb_base_current_read1_t;
|
|
284 $nb_base_read2_t += $nb_base_current_read2_t;
|
|
285
|
|
286
|
|
287 }
|
|
288 }
|
|
289 }
|
|
290
|
|
291
|
|
292 #@ 7 sec
|
|
293 }
|
|
294 }
|
|
295
|
|
296 close (READ1);
|
|
297 close (READ2);
|
|
298 close (OUT1);
|
|
299 close (OUT2);
|
|
300
|
|
301 open (LF,">$log_file") or die("Can't open $log_file\n");
|
|
302 print LF "\n####\t Fastq preparation \n";
|
|
303 print LF "## Before preparation\n";
|
|
304 print LF "#Read1 :\t$nb_read1\t#Base :\t$nb_base_read1\n";
|
|
305 print LF "#Read2 :\t$nb_read2\t#Base :\t$nb_base_read2\n";
|
|
306 print LF "## After preparation\n";
|
|
307 print LF "#Read1 :\t$nb_read1_t\t#Base :\t$nb_base_read1_t\n";
|
|
308 print LF "#Read2 :\t$nb_read2_t\t#Base :\t$nb_base_read2_t\n";
|
|
309 close (LF);
|
|
310
|
|
311
|
|
312 sub grooming_and_trimming{
|
|
313 my $header = shift;
|
|
314 my $seq = shift;
|
|
315 my $quality = shift;
|
|
316 my $quality_converted="";
|
|
317 my $quality_ori=$quality;
|
|
318
|
|
319 my $lengthseq = length($seq);
|
|
320 my $startTrim = 0;
|
|
321 my $stopTrim = length($quality)-1;
|
|
322 my $startnoN = $startTrim;
|
|
323 my $stopnoN = $stopTrim;
|
|
324
|
|
325
|
|
326 my $chercheN = $seq;
|
|
327 my @bad_position_N;
|
|
328 my @bad_position_Q;
|
|
329 my $current_index = index($chercheN,"N");
|
|
330 my $abs_index = $current_index;
|
|
331 while ($current_index >=0){
|
|
332 push (@bad_position_N,$abs_index);
|
|
333
|
|
334 if ($current_index<length($seq)){
|
|
335 $chercheN = substr($chercheN,$current_index+1);
|
|
336 $current_index = index($chercheN,"N");
|
|
337 $abs_index = $current_index + $bad_position_N[$#bad_position_N]+1;
|
|
338 }
|
|
339 else {
|
|
340 last;
|
|
341 }
|
|
342 }
|
|
343
|
|
344 my @q = split(//,$quality);
|
|
345 for (my $i=0;$i<=$#q;$i++){
|
|
346 my $chr = $q[$i];
|
|
347 my $num = ord($q[$i]);
|
|
348 if ($TYPE eq "illumina"){
|
|
349 $num = $num - 31; # 31 comme la difference entre la plage sanger (33-> 93 / 0->60) et illumina (64->104 / 0->40)
|
|
350 $quality_converted .= chr($num);
|
|
351 }
|
|
352
|
|
353 if ($num < $MIN_QUALITY + 33){ #33 comme le départ de la plage sanger
|
|
354 push(@bad_position_Q,$i);
|
|
355 }
|
|
356 }
|
|
357 if ($quality_converted){$quality = $quality_converted;}
|
|
358
|
|
359 my @bad_position = (@bad_position_N, @bad_position_Q);
|
|
360
|
|
361 if ($#bad_position>=0){
|
|
362 @bad_position = sort {$a <=> $b} @bad_position;
|
|
363 my %coord=%{&extract_longer_string_coordinates_from_bad_position(0,$stopTrim,\@bad_position)};
|
|
364 $startTrim = $coord{"start"};
|
|
365 $stopTrim = $coord{"stop"};
|
|
366 #print "$startTrim .. $stopTrim\n";
|
|
367
|
|
368 }
|
|
369 my $lengthTrim = $stopTrim - $startTrim +1;
|
|
370
|
|
371 #if ($stats_length{$lengthTrim}){
|
|
372 # $stats_length{$lengthTrim} = 1;
|
|
373 #}
|
|
374 #else {
|
|
375 # $stats_length{$lengthTrim}++;
|
|
376 #}
|
|
377 my $fastq_lines="";
|
|
378
|
|
379 # if ($header =~ /GA8\-EAS671_0005\:3\:1\:1043\:4432/){
|
|
380 # print "HEAD:\t$header";
|
|
381 # print "SEQ:\n$seq\n";
|
|
382 # print "$quality_ori\n";
|
|
383 # print "$quality\n";
|
|
384 # for (my $i=0;$i<=$#bad_position;$i++){
|
|
385 # print $bad_position[$i]."(".$q[$bad_position[$i]]." : ".ord($q[$bad_position[$i]]).")"."\t";
|
|
386 # }
|
|
387 # print "\n";
|
|
388 # print "$startTrim .. $stopTrim / $lengthTrim \n";
|
|
389 # print $fastq_lines;
|
|
390 # print "\n";
|
|
391 # }
|
|
392
|
|
393 #for (my $i=$startTrim;$i<=$stopTrim;$i++){
|
|
394 # if ($stats_quality{ord($q{$i])}){
|
|
395 # $stats_quality{ord($q{$i])}=1;
|
|
396 # }
|
|
397 # else {
|
|
398 # $stats_quality{ord($q{$i])}++;
|
|
399 # }
|
|
400 #}
|
|
401
|
|
402 if ($lengthTrim >= $MIN_LENGTH){
|
|
403 $fastq_lines .= $header;
|
|
404 my $new_seq = substr($seq,$startTrim,$lengthTrim);
|
|
405 # $nb_base_current_t = length($new_seq);
|
|
406 $fastq_lines .= $new_seq."\n";
|
|
407 $fastq_lines .= "+\n";
|
|
408 my $new_q = substr($quality,$startTrim,$lengthTrim);
|
|
409 $fastq_lines .= $new_q."\n";
|
|
410 return $fastq_lines;
|
|
411
|
|
412 }
|
|
413 else {
|
|
414 #print "Insufficient length after trimming\n";
|
|
415 return "";
|
|
416 }
|
|
417 }
|
|
418
|
|
419 sub extract_longer_string_coordinates_from_bad_position{
|
|
420 my $start=shift;
|
|
421 my $stop =shift;
|
|
422 my $refbad = shift;
|
|
423 my @bad_position = @$refbad;
|
|
424 my %coord;
|
|
425
|
|
426 my $current_start = $start;
|
|
427 my $current_stop = $bad_position[0]-1;
|
|
428 if ($current_stop < $start){$current_stop = $start;}
|
|
429
|
|
430
|
|
431 #debut -> premier N
|
|
432 my $current_length = $current_stop - $current_start +1;
|
|
433 my $test_length;
|
|
434
|
|
435 #entre les N
|
|
436 for (my $i=1;$i<=$#bad_position;$i++){
|
|
437 $test_length = $bad_position[$i]+1-$bad_position[$i-1]-1;
|
|
438 if ( $test_length > $current_length){
|
|
439 $current_start = $bad_position[$i-1]+1;
|
|
440 $current_stop = $bad_position[$i]-1;
|
|
441 $current_length = $current_stop - $current_start +1;
|
|
442 }
|
|
443 }
|
|
444
|
|
445 #dernier N -> fin
|
|
446 $test_length = $stop-$bad_position[$#bad_position]+1;
|
|
447 if ( $test_length > $current_length){
|
|
448 $current_start = $bad_position[$#bad_position]+1;
|
|
449 if ($current_start > $stop){$current_start=$stop;}
|
|
450 $current_stop = $stop;
|
|
451 }
|
|
452 $coord{"start"}=$current_start;
|
|
453 $coord{"stop"}= $current_stop;
|
|
454 $coord{"lenght"}=$current_stop-$current_start+1;
|
|
455
|
|
456 return \%coord;
|
|
457 }
|