File: GFF.pm

package info (click to toggle)
bioperl 1.7.2-3
  • links: PTS, VCS
  • area: main
  • in suites: buster
  • size: 49,564 kB
  • sloc: perl: 170,474; xml: 22,869; lisp: 2,034; sh: 1,990; makefile: 22
file content (1311 lines) | stat: -rw-r--r-- 38,178 bytes parent folder | download
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
#
# BioPerl module for Bio::Tools::GFF
#
# Please direct questions and support issues to <bioperl-l@bioperl.org>
#
# Cared for by the Bioperl core team
#
# Copyright Matthew Pocock
#
# You may distribute this module under the same terms as perl itself

# POD documentation - main docs before the code

=head1 NAME

Bio::Tools::GFF - A Bio::SeqAnalysisParserI compliant GFF format parser

=head1 SYNOPSIS

    use Bio::Tools::GFF;

    # specify input via -fh or -file
    my $gffio = Bio::Tools::GFF->new(-fh => \*STDIN, -gff_version => 2);
    my $feature;
    # loop over the input stream
    while($feature = $gffio->next_feature()) {
        # do something with feature
    }
    $gffio->close();

    # you can also obtain a GFF parser as a SeqAnalasisParserI in
    # HT analysis pipelines (see Bio::SeqAnalysisParserI and
    # Bio::Factory::SeqAnalysisParserFactory)
    my $factory = Bio::Factory::SeqAnalysisParserFactory->new();
    my $parser = $factory->get_parser(-input => \*STDIN, -method => "gff");
    while($feature = $parser->next_feature()) {
        # do something with feature
    }

=head1 DESCRIPTION

This class provides a simple GFF parser and writer. In the sense of a
SeqAnalysisParser, it parses an input file or stream into SeqFeatureI
objects, but is not in any way specific to a particular analysis
program and the output that program produces.

That is, if you can get your analysis program spit out GFF, here is
your result parser.

=head1 GFF3 AND SEQUENCE DATA

GFF3 supports sequence data; see

http://www.sequenceontology.org/gff3.shtml

There are a number of ways to deal with this -

If you call

  $gffio->ignore_sequence(1)

prior to parsing the sequence data is ignored; this is useful if you
just want the features. It avoids the memory overhead in building and
caching sequences

Alternatively, you can call either

  $gffio->get_seqs()

Or

  $gffio->seq_id_by_h()

At the B<end> of parsing to get either a list or hashref of Bio::Seq
objects (see the documentation for each of these methods)

Note that these objects will not have the features attached - you have
to do this yourself, OR call

  $gffio->features_attached_to_seqs(1)

PRIOR to parsing; this will ensure that the Seqs have the features
attached; ie you will then be able to call

  $seq->get_SeqFeatures();

And use Bio::SeqIO methods

Note that auto-attaching the features to seqs will incur a higher
memory overhead as the features must be cached until the sequence data
is found

=head1 TODO

Make a Bio::SeqIO class specifically for GFF3 with sequence data

=head1 FEEDBACK

=head2 Mailing Lists

User feedback is an integral part of the evolution of this and other
Bioperl modules. Send your comments and suggestions preferably to one
of the Bioperl mailing lists.  Your participation is much appreciated.

  bioperl-l@bioperl.org                  - General discussion
  http://bioperl.org/wiki/Mailing_lists  - About the mailing lists

=head2 Support

Please direct usage questions or support issues to the mailing list:

I<bioperl-l@bioperl.org>

rather than to the module maintainer directly. Many experienced and
reponsive experts will be able look at the problem and quickly
address it. Please include a thorough description of the problem
with code and data examples if at all possible.

=head2 Reporting Bugs

Report bugs to the Bioperl bug tracking system to help us keep track
the bugs and their resolution.  Bug reports can be submitted the web:

  https://github.com/bioperl/bioperl-live/issues

=head1 AUTHOR - Matthew Pocock

Email mrp-at-sanger.ac.uk

=head1 CONTRIBUTORS

Jason Stajich, jason-at-biperl-dot-org
Chris Mungall, cjm-at-fruitfly-dot-org
Steffen Grossmann [SG], grossman at molgen.mpg.de
Malcolm Cook, mec-at-stowers-institute.org

=head1 APPENDIX

The rest of the documentation details each of the object methods. Internal methods are usually preceded with a _

=cut

# Let the code begin...

package Bio::Tools::GFF;

use vars qw($HAS_HTML_ENTITIES);
use strict;

use Bio::Seq::SeqFactory;
use Bio::LocatableSeq;
use Bio::SeqFeature::Generic;

use base qw(Bio::Root::Root Bio::SeqAnalysisParserI Bio::Root::IO);

my $i = 0;
my %GFF3_ID_Tags = map { $_ => $i++ } qw(ID Parent Target);

# for skipping data that may be represented elsewhere; currently, this is
# only the score
my %SKIPPED_TAGS = map { $_ => 1 } qw(score);


=head2 new

 Title   : new
 Usage   : my $parser = Bio::Tools::GFF->new(-gff_version => 2,
                                             -file        => "filename.gff");
           or
           my $writer = Bio::Tools::GFF->new(-gff_version => 3,
                                             -file        => ">filename.gff3");
 Function: Creates a new instance. Recognized named parameters are -file, -fh,
           and -gff_version.
 Returns : a new object
 Args    : named parameters
           -gff_version => [1,2,3]

=cut

{   # make a class variable such that we can generate unique ID's over
    # a session, no matter how many instances of GFF.pm we make
    # since these have to be unique within the scope of a GFF file.

    my $gff3_featureID = 0;

    sub _incrementGFF3ID {
        my ($self) = @_;
        return ++ $gff3_featureID;
    }
}


sub new {
    my ($class, @args) = @_;
    my $self = $class->SUPER::new(@args);

    my ($gff_version, $noparse) = $self->_rearrange([qw(GFF_VERSION NOPARSE)],@args);

    # initialize IO
    $self->_initialize_io(@args);
    $self->_parse_header() unless $noparse;

    $gff_version ||= 2;
    if( ! $self->gff_version($gff_version) )  {
        $self->throw("Can't build a GFF object with the unknown version ".
            $gff_version);
    }
    $self->{'_first'} = 1;
    return $self;
}


=head2 _parse_header

 Title   : _parse_header
 Usage   : $gffio->_parse_header()
 Function: used to turn parse GFF header lines.  currently
           produces Bio::LocatableSeq objects from ##sequence-region
           lines
 Returns : 1 on success
 Args    : none


=cut

sub _parse_header{
    my ($self) = @_;

    my @unhandled;
    local $^W = 0; # hide warnings when we try and parse from a file opened
                   # for writing - there isn't really a better way to do
                   # AFAIK - cannot detech if a FH is read or write.
    while(my $line = $self->_readline()){
        my $handled = 0;
        next if /^\s+$/;
        if($line =~ /^\#\#sequence-region\s+(\S+)\s+(\S+)\s+(\S+)\s*/){
            my($seqid,$start,$end) = ($1,$2,$3);
            push @{ $self->{'segments'} }, Bio::LocatableSeq->new(
                -id    => unescape($seqid),
                -start => $start,
                -end   => $end,
                -length => ($end - $start + 1),  ## make the length explicit
           );
           $handled = 1;
        } elsif($line =~ /^(\#\#feature-ontology)/) {
            #to be implemented
            $self->warn("$1 header tag parsing unimplemented");
        } elsif($line =~ /^(\#\#attribute-ontology)/) {
            #to be implemented
            $self->warn("$1 header tag parsing unimplemented");
        } elsif($line =~ /^(\#\#source-ontology)/) {
            #to be implemented
            $self->warn("$1 header tag parsing unimplemented");
        } elsif($line =~ /^(\#\#\#)/) {
            #to be implemented
            $self->warn("$1 header tag parsing unimplemented");
        } elsif($line =~ /^(\#\#FASTA)/) {
            # initial ##FASTA is optional - artemis does not use it
            $line = $self->_readline();
            if ($line !~ /^\>(\S+)/) {
                $self->throw("##FASTA directive must be followed by fasta header, not: $line");
            }
        }

        if ($line =~ /^\>(.*)/) {
            # seq data can be at header or footer
            my $seq = $self->_parse_sequence($line);
            if ($seq) {
                $self->_seq_by_id_h->{$seq->primary_id} = $seq;
            }
        }


        if(!$handled){
            push @unhandled, $line;
        }

        #looks like the header is over!
        last unless $line =~ /^\#/;
    }

    foreach my $line (@unhandled){
        $self->_pushback($line);
    }

    return 1;
}

sub _parse_sequence {
    my ($self, $line) = @_;

    if ($line =~ /^\>(.*)/) {

        my $seqid = $1;
        $seqid =~ s/\s+$//;
        my $desc = '';
        if ($seqid =~ /(\S+)\s+(.*)/) {
            ($seqid, $desc) = ($1,$2);
        }
        my $res = '';
        while (my $line = $self->_readline) {
            if ($line =~ /^\#/) {
                last;
            }
            if ($line =~ /^\>/) {
                $self->_pushback($line);
                last;
            }
            $line =~ s/\s//g;
            $res .= $line;
        }
        return if $self->ignore_sequence;

        my $seqfactory = Bio::Seq::SeqFactory->new('Bio::Seq');
        my $seq = $seqfactory->create(-seq=>$res,
                                      -id=>$seqid,
                                      -desc=>$desc);
        $seq->accession_number($seqid);
        if ($self->features_attached_to_seqs) {
            my @feats =
              @{$self->_feature_idx_by_seq_id->{$seqid}};
            $seq->add_SeqFeature($_) foreach @feats;
            @{$self->_feature_idx_by_seq_id->{$seqid}} = ();
        }
        return $seq;
    }
    else {
        $self->throw("expected fasta header, not: $line");
    }
}


=head2 next_segment

 Title   : next_segment
 Usage   : my $seq = $gffio->next_segment;
 Function: Returns a Bio::LocatableSeq object corresponding to a
           GFF "##sequence-region" header line.
 Example :
 Returns : A Bio::LocatableSeq object, or undef if
           there are no more sequences.
 Args    : none


=cut

sub next_segment{
    my ($self,@args) = @_;
    return shift @{ $self->{'segments'} } if defined $self->{'segments'};
    return;
}


=head2 next_feature

 Title   : next_feature
 Usage   : $seqfeature = $gffio->next_feature();
 Function: Returns the next feature available in the input file or stream, or
           undef if there are no more features.
 Example :
 Returns : A Bio::SeqFeatureI implementing object, or undef if there are no
           more features.
 Args    : none

=cut

sub next_feature {
    my ($self) = @_;

    my $gff_string;

    # be graceful about empty lines or comments, and make sure we return undef
    # if the input's consumed
    while(($gff_string = $self->_readline()) && defined($gff_string)) {
        if ($gff_string =~ /^\#\#\#/) {
            # all forward refs have been seen; TODO
        }
        next if($gff_string =~ /^\#/ || $gff_string =~ /^\s*$/ ||
                $gff_string =~ m{^//});

        while ($gff_string =~ /^\>(.+)/) {
            # fasta can be in header or footer
            my $seq = $self->_parse_sequence($gff_string);
            if ($seq) {
                $self->_seq_by_id_h->{$seq->primary_id} = $seq;
                $gff_string = $self->_readline;
                last unless $gff_string;
            }
        }
        last;
    }
    return unless $gff_string;

    my $feat = Bio::SeqFeature::Generic->new();
    $self->from_gff_string($feat, $gff_string);

    if ($self->features_attached_to_seqs) {
        push(@{$self->_feature_idx_by_seq_id->{$feat->seq_id}},
             $feat);
    }

    return $feat;
}

sub _feature_idx_by_seq_id {
    my $self = shift;
    $self->{__feature_idx_by_seq_id} = shift if @_;
    $self->{__feature_idx_by_seq_id} = {}
      unless $self->{__feature_idx_by_seq_id};
    return $self->{__feature_idx_by_seq_id};
}


=head2 from_gff_string

 Title   : from_gff_string
 Usage   : $gff->from_gff_string($feature, $gff_string);
 Function: Sets properties of a SeqFeatureI object from a GFF-formatted
           string. Interpretation of the string depends on the version
           that has been specified at initialization.

           This method is used by next_feature(). It actually dispatches to
           one of the version-specific (private) methods.
 Example :
 Returns : void
 Args    : A Bio::SeqFeatureI implementing object to be initialized
           The GFF-formatted string to initialize it from

=cut

sub from_gff_string {
    my ($self, $feat, $gff_string) = @_;

    if($self->gff_version() == 1)  {
        return $self->_from_gff1_string($feat, $gff_string);
    } elsif( $self->gff_version() == 3 ) {
        return $self->_from_gff3_string($feat, $gff_string);
    } else {
        return $self->_from_gff2_string($feat, $gff_string);
    }
}


=head2 _from_gff1_string

 Title   : _from_gff1_string
 Usage   :
 Function:
 Example :
 Returns : void
 Args    : A Bio::SeqFeatureI implementing object to be initialized
           The GFF-formatted string to initialize it from

=cut

sub _from_gff1_string {
    my ($gff, $feat, $string) = @_;
    chomp $string;
    my ($seqname, $source, $primary, $start, $end, $score,
        $strand, $frame, @group) = split(/\t/, $string);

    if ( !defined $frame ) {
        $feat->throw("[$string] does not look like GFF to me");
    }
    $frame = 0 unless( $frame =~ /^\d+$/);
    $feat->seq_id($seqname);
    $feat->source_tag($source);
    $feat->primary_tag($primary);
    $feat->start($start);
    $feat->end($end);
    $feat->frame($frame);
    if ( $score eq '.' ) {
        #$feat->score(undef);
    } else {
        $feat->score($score);
    }
    if ( $strand eq '-' ) { $feat->strand(-1); }
    if ( $strand eq '+' ) { $feat->strand(1); }
    if ( $strand eq '.' ) { $feat->strand(0); }
    foreach my $g ( @group ) {
        if ( $g =~ /(\S+)=(\S+)/ ) {
            my $tag = $1;
            my $value = $2;
            $feat->add_tag_value($1, $2);
        } else {
            $feat->add_tag_value('group', $g);
        }
    }
}


=head2 _from_gff2_string

 Title   : _from_gff2_string
 Usage   :
 Function:
 Example :
 Returns : void
 Args    : A Bio::SeqFeatureI implementing object to be initialized
           The GFF2-formatted string to initialize it from


=cut

sub _from_gff2_string {
    my ($gff, $feat, $string) = @_;
    chomp($string);

    # according to the Sanger website, GFF2 should be single-tab
    # separated elements, and the free-text at the end should contain
    # text-translated tab symbols but no "real" tabs, so splitting on
    # \t is safe, and $attribs gets the entire attributes field to be
    # parsed later

    # sendu: but the tag value pair can (should?) be separated by a tab. The
    # 'no tabs' thing seems to apply only to the free text that is allowed for
    # the value

    my ($seqname, $source, $primary, $start,
        $end, $score, $strand, $frame, @attribs) = split(/\t+/, $string);
    my $attribs = join ' ', @attribs;

    if ( !defined $frame ) {
        $feat->throw("[$string] does not look like GFF2 to me");
    }
    $feat->seq_id($seqname);
    $feat->source_tag($source);
    $feat->primary_tag($primary);
    $feat->start($start);
    $feat->end($end);
    $feat->frame($frame);
    if ( $score eq '.' ) {
        # $feat->score(undef);
    } else {
        $feat->score($score);
    }
    if ( $strand eq '-' ) { $feat->strand(-1); }
    if ( $strand eq '+' ) { $feat->strand(1); }
    if ( $strand eq '.' ) { $feat->strand(0); }


    #  <Begin Inefficient Code from Mark Wilkinson>
    # this routine is necessay to allow the presence of semicolons in
    # quoted text Semicolons are the delimiting character for new
    # tag/value attributes.  it is more or less a "state" machine, with
    # the "quoted" flag going up and down as we pass thorugh quotes to
    # distinguish free-text semicolon and hash symbols from GFF control
    # characters

    my $flag = 0; # this could be changed to a bit and just be twiddled
    my @parsed;

    # run through each character one at a time and check it
    # NOTE: changed to foreach loop which is more efficient in perl
    # --jasons
    for my $a ( split //, $attribs ) {
        # flag up on entering quoted text, down on leaving it
        if( $a eq '"') { $flag = ( $flag == 0 ) ? 1:0 }
        elsif( $a eq ';' && $flag ) { $a = "INSERT_SEMICOLON_HERE"}
        elsif( $a eq '#' && ! $flag ) { last }
        push @parsed, $a;
    }
    $attribs = join "", @parsed; # rejoin into a single string

    # <End Inefficient Code>
    # Please feel free to fix this and make it more "perlish"

    my @key_vals = split /;/, $attribs;   # attributes are semicolon-delimited

    foreach my $pair ( @key_vals ) {
        # replace semicolons that were removed from free-text above.
        $pair =~ s/INSERT_SEMICOLON_HERE/;/g;

        # separate the key from the value
        my ($blank, $key, $values) = split  /^\s*([\w\d]+)\s/, $pair;

        if( defined $values ) {
            my @values;
            # free text is quoted, so match each free-text block
            # and remove it from the $values string
            while ($values =~ s/"(.*?)"//){
                # and push it on to the list of values (tags may have
                # more than one value... and the value may be undef)
                push @values, $1;
            }

            # and what is left over should be space-separated
            # non-free-text values

            my @othervals = split /\s+/, $values;
            foreach my $othervalue(@othervals){
                # get rid of any empty strings which might
                # result from the split
                if (CORE::length($othervalue) > 0) {push @values, $othervalue}
            }

            foreach my $value(@values){
                $feat->add_tag_value($key, $value);
            }
        }
    }
}


sub _from_gff3_string {
    my ($gff, $feat, $string) = @_;
    chomp($string);

    # according to the now nearly final GFF3 spec, columns should
    # be tab separated, allowing unescaped spaces to occur in
    # column 9

    my ($seqname, $source, $primary, $start, $end,
        $score, $strand, $frame, $groups) = split(/\t/, $string);

    if ( ! defined $frame ) {
        $feat->throw("[$string] does not look like GFF3 to me");
    }
    $feat->seq_id($seqname);
    $feat->source_tag($source);
    $feat->primary_tag($primary);
    $feat->start($start);
    $feat->end($end);
    $feat->frame($frame);
    if ( $score eq '.' ) {
        #$feat->score(undef);
    } else {
        $feat->score($score);
    }
    if ( $strand eq '-' ) { $feat->strand(-1); }
    if ( $strand eq '+' ) { $feat->strand(1); }
    if ( $strand eq '.' ) { $feat->strand(0); }
    my @groups = split(/\s*;\s*/, $groups);

    for my $group (@groups) {
        my ($tag,$value) = split /=/,$group;
        $tag             = unescape($tag);
        my @values       = map {unescape($_)} split /,/,$value;
        for my $v ( @values ) {  $feat->add_tag_value($tag,$v); }
    }
}

# taken from Bio::DB::GFF
sub unescape {
  my $v = shift;
  $v =~ s/%([0-9a-fA-F]{2})/chr hex($1)/ge;
  return $v;
}


=head2 write_feature

 Title   : write_feature
 Usage   : $gffio->write_feature($feature);
 Function: Writes the specified SeqFeatureI object in GFF format to the stream
           associated with this instance.
 Returns : none
 Args    : An array of Bio::SeqFeatureI implementing objects to be serialized

=cut

sub write_feature {
    my ($self, @features) = @_;
    return unless @features;
    if( $self->{'_first'} && $self->gff_version() == 3 ) {
        $self->_print("##gff-version 3\n");
    }
    $self->{'_first'} = 0;
    foreach my $feature ( @features ) {
        $self->_print($self->gff_string($feature)."\n");
    }
}


=head2 gff_string

 Title   : gff_string
 Usage   : $gffstr = $gffio->gff_string($feature);
 Function: Obtain the GFF-formatted representation of a SeqFeatureI object.
           The formatting depends on the version specified at initialization.

           This method is used by write_feature(). It actually dispatches to
           one of the version-specific (private) methods.
 Example :
 Returns : A GFF-formatted string representation of the SeqFeature
 Args    : A Bio::SeqFeatureI implementing object to be GFF-stringified

=cut

sub gff_string{
    my ($self, $feature) = @_;

    if($self->gff_version() == 1) {
        return $self->_gff1_string($feature);
    } elsif( $self->gff_version() == 3 ) {
        return $self->_gff3_string($feature);
    } elsif( $self->gff_version() == 2.5 ) {
        return $self->_gff25_string($feature);
    } else {
        return $self->_gff2_string($feature);
    }
}


=head2 _gff1_string

 Title   : _gff1_string
 Usage   : $gffstr = $gffio->_gff1_string
 Function:
 Example :
 Returns : A GFF1-formatted string representation of the SeqFeature
 Args    : A Bio::SeqFeatureI implementing object to be GFF-stringified

=cut

sub _gff1_string{
    my ($gff, $feat) = @_;
    my ($str,$score,$frame,$name,$strand);

    if( $feat->can('score') ) {
        $score = $feat->score();
    }
    $score = '.' unless defined $score;

    if( $feat->can('frame') ) {
        $frame = $feat->frame();
    }
    $frame = '.' unless defined $frame;

    $strand = $feat->strand();
    if(! $strand) {
        $strand = ".";
    } elsif( $strand == 1 ) {
        $strand = '+';
    } elsif ( $feat->strand == -1 ) {
        $strand = '-';
    }

    if( $feat->can('seqname') ) {
        $name = $feat->seq_id();
        $name ||= 'SEQ';
    } else {
        $name = 'SEQ';
    }

    $str = join("\t",
                $name,
                $feat->source_tag,
                $feat->primary_tag,
                $feat->start,
                $feat->end,
                $score,
                $strand,
                $frame);

    foreach my $tag ( $feat->get_all_tags ) {
        next if exists $SKIPPED_TAGS{$tag};
        foreach my $value ( $feat->get_tag_values($tag) ) {
        $str .= " $tag=$value" if $value;
        }
    }

    return $str;
}


=head2 _gff2_string

 Title   : _gff2_string
 Usage   : $gffstr = $gffio->_gff2_string
 Function:
 Example :
 Returns : A GFF2-formatted string representation of the SeqFeature
 Args    : A Bio::SeqFeatureI implementing object to be GFF2-stringified

=cut

sub _gff2_string{
    my ($gff, $origfeat) = @_;
    my $feat;
    if ($origfeat->isa('Bio::SeqFeature::FeaturePair')){
        $feat = $origfeat->feature2;
    } else {
        $feat = $origfeat;
    }
    my ($str1, $str2,$score,$frame,$name,$strand);

    if( $feat->can('score') ) {
        $score = $feat->score();
    }
    $score = '.' unless defined $score;

    if( $feat->can('frame') ) {
        $frame = $feat->frame();
    }
    $frame = '.' unless defined $frame;

    $strand = $feat->strand();
    if(! $strand) {
        $strand = ".";
    } elsif( $strand == 1 ) {
        $strand = '+';
    } elsif ( $feat->strand == -1 ) {
        $strand = '-';
    }

    if( $feat->can('seqname') ) {
        $name = $feat->seq_id();
    }
    $name ||= 'SEQ';

    $str1 = join("\t",
                 $name,
                 $feat->source_tag(),
                 $feat->primary_tag(),
                 $feat->start(),
                 $feat->end(),
                 $score,
                 $strand,
                 $frame);
    # the routine below is the only modification I made to the original
    # ->gff_string routine (above) as on November 17th, 2000, the
    # Sanger webpage describing GFF2 format reads: "From version 2
    # onwards, the attribute field must have a tag value structure
    # following the syntax used within objects in a .ace file,
    # flattened onto one line by semicolon separators. Tags must be
    # standard identifiers ([A-Za-z][A-Za-z0-9_]*).  Free text values
    # must be quoted with double quotes".
    # MW

    my @group;

    foreach my $tag ( $feat->get_all_tags ) {
        next if exists $SKIPPED_TAGS{$tag};
        my @v;
        foreach my $value ( $feat->get_tag_values($tag) ) {
            unless( defined $value && length($value) ) {
                # quote anything other than valid tag/value characters
                $value = '""';
            } elsif ($value =~ /[^A-Za-z0-9_]/){
                # substitute tab and newline chars by their UNIX equivalent
                $value =~ s/\t/\\t/g;
                $value =~ s/\n/\\n/g;
                $value = '"' . $value . '" ';
            }
            push @v, $value;
            # for this tag (allowed in GFF2 and .ace format)
        }
        push @group, "$tag ".join(" ", @v);
    }

    $str2 .= join(' ; ', @group);
    # Add Target information for Feature Pairs
    if( ! $feat->has_tag('Target') && # This is a bad hack IMHO
        ! $feat->has_tag('Group')  &&
        $origfeat->isa('Bio::SeqFeature::FeaturePair') ) {
        $str2 = sprintf("Target %s %d %d", $origfeat->feature1->seq_id,
                       ( $origfeat->feature1->strand < 0 ?
                         ( $origfeat->feature1->end,
                           $origfeat->feature1->start) :
                         ( $origfeat->feature1->start,
                           $origfeat->feature1->end)
                         )) . ($str2?" ; ".$str2:"");  # need to put Target information before other tag/value pairs - mw
    }
    return $str1."\t".$str2;
}


=head2 _gff25_string

 Title   : _gff25_string
 Usage   : $gffstr = $gffio->_gff2_string
 Function: To get a format of GFF that is peculiar to Gbrowse/Bio::DB::GFF
 Example :
 Returns : A GFF2.5-formatted string representation of the SeqFeature
 Args    : A Bio::SeqFeatureI implementing object to be GFF2.5-stringified

=cut

sub _gff25_string {
    my ($gff, $origfeat) = @_;
    my $feat;
    if ($origfeat->isa('Bio::SeqFeature::FeaturePair')){
        $feat = $origfeat->feature2;
    } else {
        $feat = $origfeat;
    }
    my ($str1, $str2,$score,$frame,$name,$strand);

    if( $feat->can('score') ) {
        $score = $feat->score();
    }
    $score = '.' unless defined $score;

    if( $feat->can('frame') ) {
        $frame = $feat->frame();
    }
    $frame = '.' unless defined $frame;

    $strand = $feat->strand();
    if(! $strand) {
        $strand = ".";
    } elsif( $strand == 1 ) {
        $strand = '+';
    } elsif ( $feat->strand == -1 ) {
        $strand = '-';
    }

    if( $feat->can('seqname') ) {
        $name = $feat->seq_id();
        $name ||= 'SEQ';
    } else {
        $name = 'SEQ';
    }
    $str1 = join("\t",
                 $name,
                 $feat->source_tag(),
                 $feat->primary_tag(),
                 $feat->start(),
                 $feat->end(),
                 $score,
                 $strand,
                 $frame);

    my @all_tags = $feat->all_tags;
    my @group; my @firstgroup;
    if (@all_tags) {   # only play this game if it is worth playing...
        foreach my $tag ( @all_tags ) {
            my @v;
            foreach my $value ( $feat->get_tag_values($tag) ) {
            next if exists $SKIPPED_TAGS{$tag};
                unless( defined $value && length($value) ) {
                    $value = '""';
                } elsif ($value =~ /[^A-Za-z0-9_]/){
                    $value =~ s/\t/\\t/g; # substitute tab and newline
                    # characters
                    $value =~ s/\n/\\n/g; # to their UNIX equivalents
                    $value = '"' . $value . '" ';
                } # if the value contains
                  # anything other than valid
                  # tag/value characters, then
                  # quote it
                push @v, $value;
                # for this tag (allowed in GFF2 and .ace format)
            }
            if (($tag eq 'Group') || ($tag eq 'Target')){ # hopefully we won't get both...
                push @firstgroup, "$tag ".join(" ", @v);
            } else {
                push @group, "$tag ".join(" ", @v);
            }
        }
    }
    $str2 = join(' ; ', (@firstgroup, @group));
    # Add Target information for Feature Pairs
    if( ! $feat->has_tag('Target') && # This is a bad hack IMHO
        ! $feat->has_tag('Group') &&
        $origfeat->isa('Bio::SeqFeature::FeaturePair') ) {
        $str2 = sprintf("Target %s ; tstart %d ; tend %d", $origfeat->feature1->seq_id,
                        ( $origfeat->feature1->strand < 0 ?
                          ( $origfeat->feature1->end,
                            $origfeat->feature1->start) :
                          ( $origfeat->feature1->start,
                            $origfeat->feature1->end)
                        )) . ($str2?" ; ".$str2:""); # need to put the target info before other tag/value pairs - mw
    }
    return $str1 . "\t".  $str2;
}


=head2 _gff3_string

  Title   : _gff3_string
  Usage   : $gffstr = $gffio->_gff3_string
  Function:
  Example :
  Returns : A GFF3-formatted string representation of the SeqFeature
  Args    : A Bio::SeqFeatureI implementing object to be GFF3-stringified

=cut

sub _gff3_string {
    my ($gff, $origfeat) = @_;
    my $feat;
    if ($origfeat->isa('Bio::SeqFeature::FeaturePair')){
        $feat = $origfeat->feature2;
    } else {
        $feat = $origfeat;
    }

    my $ID = $gff->_incrementGFF3ID();

    my ($score,$frame,$name,$strand);

    if( $feat->can('score') ) {
        $score = $feat->score();
    }
    $score = '.' unless defined $score;

    if( $feat->can('frame') ) {
        $frame = $feat->frame();
    }
    $frame = '1' unless defined $frame;

    $strand = $feat->strand();

    if(! $strand) {
        $strand = ".";
    } elsif( $strand == 1 ) {
        $strand = '+';
    } elsif ( $feat->strand == -1 ) {
        $strand = '-';
    }

    if( $feat->can('seqname') ) {
        $name = $feat->seq_id();
        $name ||= 'SEQ';
    } else {
        $name = 'SEQ';
    }

    my @groups;

    # force leading ID and Parent tags
    my @all_tags =  grep { ! exists $GFF3_ID_Tags{$_} } $feat->all_tags;
    for my $t ( sort { $GFF3_ID_Tags{$b} <=> $GFF3_ID_Tags{$a} }
                keys %GFF3_ID_Tags ) {
        unshift @all_tags, $t if $feat->has_tag($t);
    }

    for my $tag ( @all_tags ) {
    next if exists $SKIPPED_TAGS{$tag};
        # next if $tag eq 'Target';
        if ($tag eq 'Target' && ! $origfeat->isa('Bio::SeqFeature::FeaturePair')){
            my @values = $feat->get_tag_values($tag);
            if(scalar(@values) > 1){ # How is it possible that Target is has a value list ??
                # simple Target,start,stop
                my ($target_id, $b,$e,$strand) = $feat->get_tag_values($tag);
                next unless(defined($e) && defined($b) && $target_id);
                ($b,$e)= ($e,$b) if(defined $strand && $strand<0);
                #if we have the strand we will print it
                if($strand){ push @groups, sprintf("Target=%s %d %d %s", $target_id,$b,$e,$strand); }
                else{ push @groups, sprintf("Target=%s %d %d", $target_id,$b,$e); }
                next;
            }
        }

        my $valuestr;
        # a string which will hold one or more values
        # for this tag, with quoted free text and
        # space-separated individual values.
        my @v;
        for my $value ( $feat->get_tag_values($tag) ) {
            if(  defined $value && length($value) ) {
                                #$value =~ tr/ /+/;  #spaces are allowed now
                if ( ref $value eq 'Bio::Annotation::Comment') {
                    $value = $value->text;
                }

                if ($value =~ /[^a-zA-Z0-9\,\;\=\.:\%\^\*\$\@\!\+\_\?\-]/) {
                    $value =~ s/\t/\\t/g; # substitute tab and newline
                    # characters
                    $value =~ s/\n/\\n/g; # to their UNIX equivalents

                    # Unescaped quotes are not allowed in GFF3
                    #                    $value = '"' . $value . '"';
                }
                $value =~ s/([\t\n\r%&\=;,])/sprintf("%%%X",ord($1))/ge;
            } else {
                # if it is completely empty, then just make empty double quotes
                $value = '""';
            }
            push @v, $value;
        }
        # can we figure out how to improve this?
        $tag = lcfirst($tag) unless ( $tag =~
            /^(ID|Name|Alias|Parent|Gap|Target|Derives_from|Note|Dbxref|Ontology_term)$/);

        push @groups, "$tag=".join(",",@v);
    }
    # Add Target information for Feature Pairs
    if( $feat->has_tag('Target') &&
        ! $feat->has_tag('Group') &&
        $origfeat->isa('Bio::SeqFeature::FeaturePair') ) {

        my $target_id = $origfeat->feature1->seq_id;
        $target_id =~ s/([\t\n\r%&\=;,])/sprintf("%%%X",ord($1))/ge;

        push @groups, sprintf("Target=%s %d %d",
                              $target_id,
                              ( $origfeat->feature1->strand < 0 ?
                                ( $origfeat->feature1->end,
                                  $origfeat->feature1->start) :
                                ( $origfeat->feature1->start,
                                  $origfeat->feature1->end)
                                ));
    }

    # unshift @groups, "ID=autogenerated$ID" unless ($feat->has_tag('ID'));
    if ( $feat->can('name') && defined($feat->name) ) {
        # such as might be for Bio::DB::SeqFeature
        unshift @groups, 'Name=' . $feat->name;
    }

    my $gff_string = "";
    if ($feat->location->isa("Bio::Location::SplitLocationI")) {
        my @locs = $feat->location->each_Location;
        foreach my $loc (@locs) {
            $gff_string .= join("\t",
                                $name,
                                $feat->source_tag() || '.',
                                $feat->primary_tag(),
                                $loc->start(),
                                $loc->end(),
                                $score,
                                $strand,
                                $frame,
                                join(';', @groups)) . "\n";
        }
        chop $gff_string;
        return $gff_string;
    } else {
        $gff_string = join("\t",
                           $name,
                           $feat->source_tag() || '.',
                           $feat->primary_tag(),
                           $feat->start(),
                           $feat->end(),
                           $score,
                           $strand,
                           $frame,
                           join(';', @groups));
    }
    return $gff_string;
}


=head2 gff_version

  Title   : _gff_version
  Usage   : $gffversion = $gffio->gff_version
  Function:
  Example :
  Returns : The GFF version this parser will accept and emit.
  Args    : none

=cut

sub gff_version {
    my ($self, $value) = @_;
    if(defined $value && grep {$value == $_ } ( 1, 2, 2.5, 3)) {
        $self->{'GFF_VERSION'} = $value;
    }
    return $self->{'GFF_VERSION'};
}


# Make filehandles

=head2 newFh

 Title   : newFh
 Usage   : $fh = Bio::Tools::GFF->newFh(-file=>$filename,-format=>'Format')
 Function: does a new() followed by an fh()
 Example : $fh = Bio::Tools::GFF->newFh(-file=>$filename,-format=>'Format')
           $feature = <$fh>;            # read a feature object
           print $fh $feature;          # write a feature object
 Returns : filehandle tied to the Bio::Tools::GFF class
 Args    :

=cut

sub newFh {
    my $class = shift;
    return unless my $self = $class->new(@_);
    return $self->fh;
}


=head2 fh

 Title   : fh
 Usage   : $obj->fh
 Function:
 Example : $fh = $obj->fh;      # make a tied filehandle
           $feature = <$fh>;    # read a feature object
           print $fh $feature;  # write a feature object
 Returns : filehandle tied to Bio::Tools::GFF class
 Args    : none

=cut


sub fh {
    my $self = shift;
    my $class = ref($self) || $self;
    my $s = Symbol::gensym;
    tie $$s,$class,$self;
    return $s;
}

# This accessor is used for accessing the Bio::Seq objects from a GFF3
# file; if the file you are using has no sequence data you can ignore
# this accessor

# This accessor returns a hash reference containing Bio::Seq objects,
# indexed by Bio::Seq->primary_id

sub _seq_by_id_h {
    my $self = shift;

    return $self->{'_seq_by_id_h'} = shift if @_;
    $self->{'_seq_by_id_h'} = {}
    unless $self->{'_seq_by_id_h'};
    return $self->{'_seq_by_id_h'};
}


=head2 get_seqs

 Title   : get_seqs
 Usage   :
 Function: Returns all Bio::Seq objects populated by GFF3 file
 Example :
 Returns :
 Args    :

=cut

sub get_seqs {
    my ($self,@args) = @_;
    return values %{$self->_seq_by_id_h};
}


=head2 features_attached_to_seqs

 Title   : features_attached_to_seqs
 Usage   : $obj->features_attached_to_seqs(1);
 Function: For use with GFF3 containing sequence only

    Setting this B<before> parsing ensures that all Bio::Seq object
    created will have the appropriate features added to them

    defaults to false (off)

    Note that this mode will incur higher memory usage because features
    will have to be cached until the relevant feature comes along

 Example :
 Returns : value of features_attached_to_seqs (a boolean)
 Args    : on set, new value (a boolean, optional)


=cut

sub features_attached_to_seqs{
    my $self = shift;

    return $self->{'_features_attached_to_seqs'} = shift if @_;
    return $self->{'_features_attached_to_seqs'};
}


=head2 ignore_sequence

 Title   : ignore_sequence
 Usage   : $obj->ignore_sequence(1);
 Function: For use with GFF3 containing sequence only

    Setting this B<before> parsing means that all sequence data will be
    ignored

 Example :
 Returns : value of ignore_sequence (a boolean)
 Args    : on set, new value (a boolean, optional)

=cut

sub ignore_sequence{
    my $self = shift;

    return $self->{'_ignore_sequence'} = shift if @_;
    return $self->{'_ignore_sequence'};
}


sub DESTROY {
    my $self = shift;
    $self->close();
}

sub TIEHANDLE {
    my ($class,$val) = @_;
    return bless {'gffio' => $val}, $class;
}

sub READLINE {
    my $self = shift;
    return $self->{'gffio'}->next_feature() || undef unless wantarray;
    my (@list, $obj);
    push @list, $obj while $obj = $self->{'gffio'}->next_feature();
    return @list;
}

sub PRINT {
    my $self = shift;
    $self->{'gffio'}->write_feature(@_);
}

1;