ProtoGene.pl 57.2 KB
Newer Older
Sebastien Moretti's avatar
Sebastien Moretti committed
1
2
3
4
5
6
7
8
9
10
11
12
#!/usr/bin/env perl

########################################################
#                                                      #
# ProtoGene : Bona-Fide Back Translation of Protein    #
#             Multiple Sequence Alignments             #
#                                                      #
########################################################

use strict;
use warnings;
use diagnostics;
Sebastien Moretti's avatar
Sebastien Moretti committed
13
14
use Carp;

Sebastien Moretti's avatar
Sebastien Moretti committed
15
use File::Which qw(which);                 # Locate external executable programs in the PATH
16
17
use Time::localtime;                       # Use localtime+PID for a pseudo-uniq temp file name
use Getopt::Long;                          # Options specifications
Sebastien Moretti's avatar
Sebastien Moretti committed
18
use File::Copy qw(move);                   # Avoid external 'mv' command usage
19
use LWP::Simple;                           # To test gigablaster availability
Sebastien Moretti's avatar
Sebastien Moretti committed
20
21
22

use Mail::Send;                            # Send warnings and errors files by e-mail ==> only if the $userEMail variable is defined

23
use lib '/mnt/common/share/ProtoGene/';    # Local path for ProtoGene's own perl modules
24
use lib '/mnt/local/lib/tcoffee_perl/';
Sebastien Moretti's avatar
Sebastien Moretti committed
25
use Exonerate;                             # Exonerate runner, parser, ...
26
use Views;                                 # Non-text outputs, e.g. HTML/CSS
27
#use CheckOutput;                         # Check output for cds consistancy with query
Sebastien Moretti's avatar
Sebastien Moretti committed
28
29
30



31
################## CONFIGURATION ##################
Sebastien Moretti's avatar
Sebastien Moretti committed
32
$ENV{'PATH'}        .= ':/mnt/local/bin/:./';                               # Additional path for executables
33
34
35
36
37

my $cachePath        = '/scratch/cluster/monthly/t_coffee/ProtoGene_Cache'; # Cache directory
my $cacheStorageTime = 15;                                                  # Do not update sequences younger than X days

my $userEMail        = 'moretti.sebastien@gmail.com';                       # To receive e-mails with encountered problems; leave blank to inactive
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54

### BLAST parameters ###
my $blast_param = { 'evalue'    => 0.05,
                    'matrix'    => 'BLOSUM62',
                    'filter'    => 'Off',
                    'species'   => 'All_organisms',

                    'db1'       => 'refseq_protein', # NCBI RefSeq protein
                    'identity1' => 100,
                    'cover1'    => 100,
                    'db2'       => 'nr', # NCBI nr protein
                    'identity2' =>  95,
                    'cover2'    =>  95,

                    'nbesthits' => 10,
                    'core'      => 1,
                  };
55
###################################################
Sebastien Moretti's avatar
Sebastien Moretti committed
56
57


Sebastien Moretti's avatar
Sebastien Moretti committed
58
my $VERSION  = '4.1.0';
Sebastien Moretti's avatar
Sebastien Moretti committed
59

60
my $webblast_exe  = '/mnt/common/share/ProtoGene/webblast.pl';
61
my $blast_exe     = 'blastall';            # Or wu-blastall for Wu-BLAST; for local blast usage
Sebastien Moretti's avatar
Sebastien Moretti committed
62
my $exonerate_exe = 'exonerate-1.0';       # Exonerate 1.0 because current parser only works with this version
63
64
65
66
67


################## Option management
my ($msa, $revtrans, $pep, $hideBOJ, $run_name, $template, $lim, $cache) = ('', 0, 0, 0, '', '', 0, 'update');
my ($debug, $tmp)                                                        = (0, 0);
68
my ($db, $species, $local, $giga)                                        = ($blast_param->{'db1'}, $blast_param->{'species'}, 0, 0);
Sebastien Moretti's avatar
Sebastien Moretti committed
69
70
71
72
73
74
75
76
77
78
79
my %opts = ('msa|in=s'        => \$msa,        # Input sequences
            'revtrans'        => \$revtrans,   # Use to reverse-translate sequences with no match
            'pep'             => \$pep,        # Add the original peptide query beneath the related CDS seq
            'hideBOJ'         => \$hideBOJ,    # Hide BOJ output
            'run_name=s'      => \$run_name,   # Use another name, instead of input seq name, for result files
            'template=s'      => \$template,   # Use a template file
            'lim=i'           => \$lim,        # Limit number of input query sequences
            'cache=s'         => \$cache,      # Cache behavior

            'orgm|species=s'  => \$species,    # Organism(s) to blast against
            'db|database=s'   => \$db,         # Database to blast against
80
            'local'           => \$local,      # Use to specify a local db query, defined in $local_db
Sebastien Moretti's avatar
Sebastien Moretti committed
81
82
83
84
85
86
87
88
89
90
            'giga'            => \$giga,       # Use GigaBlaster server

            'version|V'       => sub { print "\n\tPROTOGENE : Bona-Fide Back Translation of Protein Multiple Sequence Alignments\n\tversion   : $VERSION\n\n";
                                       exit 0;
                                 },            # Print version information
            'help|H|?'        => sub { system('perldoc', "$0")==0 || print {*STDERR} "\n\tCannot open the documentation with 'perldoc'\n\n";
                                       exit 0;
                                     },        # Print full help message
            'debug'           => \$debug,      # Verbose output
            'tmp'             => \$tmp,        # To keep traces of fake intermediate files like fake xml from NCBI, fake aln, ...
Sebastien Moretti's avatar
Sebastien Moretti committed
91
           );
Sebastien Moretti's avatar
Sebastien Moretti committed
92
93
my $test_option_values = Getopt::Long::GetOptions(%opts);

94
95

################## Short help message
96
if ( !$test_option_values || ($msa eq '' && $cache ne 'empty' && $cache ne 'old') ){
97
98
99
100
101
102
103
104
    print {*STDERR} "\n\tCannot open the MSA file in FASTA format
\tTry:  $0 --msa=path_of_the_fasta_msa_file [Options]

\tOptions: --orgm=All_organisms, Bacteria, Viruses, Vertebrata,
\t                Eukaryota, Mammalia, Primates, Homo_sapiens,
\t                Gallus_gallus, Bos_taurus, Escherichia_coli,
\t                Arabidopsis_thaliana, Mus_musculus,
\t                Drosophila_Melanogaster, ...
105
\t                default is '$blast_param->{'species'}'
106
\t         --db=nr, pdb, swissprot, refseq_protein
107
\t                default is '$blast_param->{'db1'}'
108
109
110
111
\t         --local to execute a local BLAST query with
\t                --db=path_for_a_local_db_blast_formated\n
\t         --template to provide your own nucleotidic sequences
\t                following the cds file format
112
\t         --revtrans reverse-translates sequences with no
113
114
\t                blast hit, in IUB (IUPAC) depiction code
\t                They are removed from the alignement by default
115
\t         --pep  adds the original peptide query beneath the
116
117
118
119
\t                back-translated sequence
\t         --cache=none, update, use, 'own_PATH_directory', old, empty
\t                to select the cache mode
\t                default is 'update'\n
120
121
122
\t         --debug   prints extra information when running
\t         --version prints version information
\t         --help    prints a full help message\n\n";
123
    exit(1);
Sebastien Moretti's avatar
Sebastien Moretti committed
124
}
125
126


127
128
################## Check cache directory accessibility
checkCacheAccessibility($cachePath);
129

130
131
# Cache management
$cache = cacheManagement($cache, $cachePath, 0) || '.';
132

133
134
# Remove old files from the cache directory everytime ProtoGene is running
cacheManagement('old', $cachePath, 1);
135
136


137
138

################## Check external software in the PATH
139
checkExternalSoftware($webblast_exe, $exonerate_exe);
140
141
if ( $local ){
    checkExternalSoftware($blast_exe);
Sebastien Moretti's avatar
Sebastien Moretti committed
142
143
144
145
}



146
147
148
################## Check input files
# Open and check template file
my ($template_NT, $template_AA, $template_SEQ);
149
#TODO: template is fasta only now
150
151
($template_NT, $template_AA, $template_SEQ) = checkTemplate($template) if ( $template );

Sebastien Moretti's avatar
Sebastien Moretti committed
152

Sebastien Moretti's avatar
Sebastien Moretti committed
153
154
155
156
157
# Open and Check the input fasta file
my $originalMSA    = $msa;
my $converted      = 0; # Accept only fasta format
#TODO: Do a better file converter and identifier is msa is not fasta
#($msa, $converted) = isFastaFile($msa, $converted);
Sebastien Moretti's avatar
Sebastien Moretti committed
158

159
my $fasta_checker = -1;
160
my (@original_names, @original_seq); #Use 2 lists in parallel : original_names, seq
Sebastien Moretti's avatar
Sebastien Moretti committed
161
162
open( my $MSA, '<', "$msa" ) or die "\n\tCannot open your MSA file\n\n";
READ_MSA:
163
while(<$MSA>){
Sebastien Moretti's avatar
Sebastien Moretti committed
164
165
    next READ_MSA if ( $_ =~ /^#/ );

166
    if ( $_ =~ /^>/ ){
167
168
        $fasta_checker++;
        my $name = $_;
Sebastien Moretti's avatar
Sebastien Moretti committed
169
170
171
        $name    =~ s{\r\n}{}g; #Remove return lines from windows OS '^M'
        $name    =~ s{\r}{\n}g;
        $name    =~ s{^>[ \t]+}{>};
172
        chomp($name);
Sebastien Moretti's avatar
Sebastien Moretti committed
173
174

        push @original_names, $name;
Sebastien Moretti's avatar
Sebastien Moretti committed
175
    }
Sebastien Moretti's avatar
Sebastien Moretti committed
176
177
178
179
180
181
182
183
184
185
    elsif ( $_ =~ /[A-Z\-\.]/i ){
        my $sequence = $_;
        $sequence    =~ s{\r\n}{}g;
        $sequence    =~ s{\r}{\n}g;
        $sequence    =~ s{\.}{-}g; #for msa with '.' as gap
        #TODO because exonerate cannot work with selenocystein
        $sequence    =~ s{[BJOUZ]}{X}ig; #U here for selenocystein
        $sequence    =~ s{[^A-Za-z\-\*\n]}{}g; #Remove all the non-gap or non-alphabetic characters from the seq
        chomp($sequence);

Sebastien Moretti's avatar
Sebastien Moretti committed
186
        #fasta sequence on 1 line
187
        if ( !exists($original_seq[$fasta_checker]) ){
Sebastien Moretti's avatar
Sebastien Moretti committed
188
            push @original_seq, $sequence;
189
190
        }
        else {
Sebastien Moretti's avatar
Sebastien Moretti committed
191
            $original_seq[$fasta_checker] .= $sequence;
192
        }
Sebastien Moretti's avatar
Sebastien Moretti committed
193
194
    }
}
195
close $MSA;
Sebastien Moretti's avatar
Sebastien Moretti committed
196
197

#Are there FASTA sequences ?
198
if ( $fasta_checker == -1 ){
Sebastien Moretti's avatar
Sebastien Moretti committed
199
    failure();
Sebastien Moretti's avatar
Sebastien Moretti committed
200
201
    print {*STDERR} "\tThe MSA file does NOT seem to be a protein FASTA format
\tSee an example of FASTA format <a href='http://en.wikipedia.org/wiki/FASTA_format' target='_blank'>here</a>\n\n";
202
    exit(1);
Sebastien Moretti's avatar
Sebastien Moretti committed
203
}
204
elsif ( $lim >0 && $fasta_checker > $lim ){
Sebastien Moretti's avatar
Sebastien Moretti committed
205
    failure();
206
    print {*STDERR} "\tThe FASTA file is too large, try with less than $lim sequences\n\tor split your file\n\n";
207
208
    exit(1);
}
209
elsif ( exists( $original_seq[0] ) && $original_seq[0] =~ /[acgtu]/i ){
210
    my $first_seq = $original_seq[0];
Sebastien Moretti's avatar
Sebastien Moretti committed
211
212
213
214
215
216
    #Check if sequences are amino acids and not nucleotides ONLY on the 1st sequence
    my $a         = ($first_seq =~ s{[aA]}{}g)               || 0;
    my $c         = ($first_seq =~ s{[cC]}{}g)               || 0;
    my $g         = ($first_seq =~ s{[gG]}{}g)               || 0;
    my $t         = ($first_seq =~ s{[tTuU]}{}g)             || 0;
    my $non       = ($first_seq =~ s{[^aAcCgGtTuUXxNn-]}{}g) || 0;
217
    if ( ($a+$c+$g+$t) >= (($a+$c+$g+$t+$non)*80/100) ){
Sebastien Moretti's avatar
Sebastien Moretti committed
218
        failure();
Sebastien Moretti's avatar
Sebastien Moretti committed
219
220
        print {*STDERR} "\tYour sequences look already to be nucleotides
\tThe goal of this program is to turn AMINO ACID alignments into CDS nucleotide alignments\n\n";
221
222
        exit(1);
    }
Sebastien Moretti's avatar
Sebastien Moretti committed
223
224
225
226
}
undef $fasta_checker;


Sebastien Moretti's avatar
Sebastien Moretti committed
227
# Check GigaBlaster status if used
228
if ( $giga==1 ){
229
    $local = 0;
Sebastien Moretti's avatar
Sebastien Moretti committed
230
    $local = 2  if ( head('http://www.igs.cnrs-mrs.fr/public_g/remoteblast.cgi') );
231
232
}

Sebastien Moretti's avatar
Sebastien Moretti committed
233

Sebastien Moretti's avatar
Sebastien Moretti committed
234
# Temporary file extension
235
my $date = sprintf("%d-%02d-%02d_%02dh%02d", localtime->year() + 1900, localtime->mon() + 1, localtime->mday(), localtime->hour, localtime->min);
Sebastien Moretti's avatar
Sebastien Moretti committed
236
237


238

Sebastien Moretti's avatar
Sebastien Moretti committed
239
#Start main program with version # of programs and list original queries
240
print "\n\t   Protogene\t$VERSION\t$date\n\n";
241

242
open(my $EXONERATEISHERE, "$exonerate_exe --version |");
243
my $IsExonerateHere = <$EXONERATEISHERE>;
244
close $EXONERATEISHERE;
Sebastien Moretti's avatar
Sebastien Moretti committed
245
246
print "\t   $IsExonerateHere\n";
undef $IsExonerateHere;
247
undef $VERSION;
248

Sebastien Moretti's avatar
Sebastien Moretti committed
249
for(my $m=0; $m<=$#original_names; $m++){
250
    printf("\t>%-3s  is  %s\n", $m, $original_names[$m]);
Sebastien Moretti's avatar
Sebastien Moretti committed
251
}
Sebastien Moretti's avatar
Sebastien Moretti committed
252
$date .= "_$$"; #Add PID to be more uniq
Sebastien Moretti's avatar
Sebastien Moretti committed
253
254
255
256



# run_name implementation for web server
257
$originalMSA = $run_name if ( $run_name ne '' );
Sebastien Moretti's avatar
Sebastien Moretti committed
258

259
260
unlink("$originalMSA.cds", "$originalMSA.cdsP", "$originalMSA.cdsP.html",
       "$originalMSA.out", "$originalMSA.boj",  "$originalMSA.log");
Sebastien Moretti's avatar
Sebastien Moretti committed
261
262


263
# Build the sequences, from the alignment, to perform blast search
264
EACH_SEQ:
265
266
267
268
269
270
for(my $r=0; $r<=$#original_names; $r++){
    my $fasta_header = $original_names[$r];
    my $short_name   = $original_names[$r];
    $fasta_header    =~ s{^>}{};               # Full FASTA header
    $short_name      =~ s{^([^\s\t]+).*$}{$1}; # Short name without description

Sebastien Moretti's avatar
Sebastien Moretti committed
271

Sebastien Moretti's avatar
Sebastien Moretti committed
272
    my @equivalent_blast_hits; #Multiple BLAST hits with the same highest e-value/score
273
274

    if ( exists($template_NT->{$short_name}) && $template_NT->{$short_name} ne '' ){
Sebastien Moretti's avatar
Sebastien Moretti committed
275
276
277
        # First try nucleotide template
        push @equivalent_blast_hits, 'My_own_seq';
        print "\n\n$r done\n\nNucleotide template for $r:\t$template_NT->{$short_name}\n\n*******************************************************************\n\n";
Sebastien Moretti's avatar
Sebastien Moretti committed
278
    }
279
    elsif ( exists($template_AA->{$short_name}) && $template_AA->{$short_name} ne '' ){
Sebastien Moretti's avatar
Sebastien Moretti committed
280
281
282
        # Second try amino acid/BLASTp template
        push @equivalent_blast_hits, $template_AA->{$short_name};
        print "\n\n$r done\n\nBlastP template for $r:\t$template_AA->{$short_name}\n\n*******************************************************************\n\n";
Sebastien Moretti's avatar
Sebastien Moretti committed
283
284
    }
    else{
285
    # else send them to webblast
Sebastien Moretti's avatar
Sebastien Moretti committed
286
287
288
289
290
291
292
        # Prepare sequence $r for BLAST
        open(my $READY2BLAST, '>', "$cache/$date.seq2blast.fas.$r");
        my $noGap = $original_seq[$r];
        $noGap    =~ s{\-}{}g; #remove gaps
        print {$READY2BLAST} ">$r\n$noGap\n";
        close $READY2BLAST;

293
294
295
296
297
        if ($r==0){
            # Show BLAST parameters only for the first BLAST run
            display_blast_param();
        }
        run_BLAST("$cache/$date.seq2blast.fas.$r", $local, $blast_param->{'db1'}, $blast_exe, "$cache/$date.blastp.$r");
Sebastien Moretti's avatar
Sebastien Moretti committed
298

299
300
301

        # Get the BLASTp hit(s)  acc number
        open(my $BLASTP, '<', "$cache/$date.blastp.$r");
302
        while(<$BLASTP>){
303
            push @equivalent_blast_hits, $1  if ( $_ =~ /^>\s*\d+@?\w?\w?\w?__([^\s\|\.]+).*$/ ); #Warning: double '_'
304
305
        }
        close $BLASTP;
306
        unlink("$cache/$date.blastp.$r", "$cache/$date.seq2blast.fas.$r") if ( $tmp == 0 || exists($equivalent_blast_hits[0]) );
Sebastien Moretti's avatar
Sebastien Moretti committed
307
        if ( !exists($equivalent_blast_hits[0]) ){
308
            print "\n$r ...\tNo blast result found above thresholds for $fasta_header\n\n";
309
            buildFailureOutputFiles($r, 'No_BLASTp_Result', 'Not_searched');
310
311
312
313
            undef $original_seq[$r];
            undef $original_names[$r];
            next EACH_SEQ;
        }
Sebastien Moretti's avatar
Sebastien Moretti committed
314
315
316
317
    }



318
    #When multiple equivalent blast hits  ==> use, and add, every hits
319
    my ($resultPOS, $resultBOJ) = ('', '');
320
    my @failureStatus = (''); # To always have $failureStatus[0] defined
321
    my @intronStatus  = ('');
322
    HIT_LINK:
Sebastien Moretti's avatar
Sebastien Moretti committed
323
324
    for(my $qq=0; $qq<=$#equivalent_blast_hits; $qq++){
        print "\n$fasta_header  -->  [$equivalent_blast_hits[$qq]]\n";
325
326
327
328


        my @nt_GIs;
        my $intronStep = 0;
329
330
331
        # Get Nt sequence from template
        if ( exists($template_NT->{$short_name}) && $template_NT->{$short_name} ne '' ){
            if ( $template_NT->{$short_name} !~ /^My_Seq$/i ){
332
                download_seq($cache, '', '', $template_NT->{$short_name});
333
334
335
336
337
338
339
340
                @nt_GIs = $template_NT->{$short_name};
            }
            else {
                open(my $MYSEQ, '>', "$cache/My_Seq-$$.fas");
                print {$MYSEQ} ">My_Seq-$$ for $short_name\n".$template_SEQ->{$short_name}."\n";
                close $MYSEQ;
                @nt_GIs = "My_Seq-$$";
            }
341
        }
342
        # Get Nt sequence through prot acc -> prot gi -> nt linked gi -> nt acc
343
        else{
Sebastien Moretti's avatar
Sebastien Moretti committed
344
            # Prot ACC -> PUID
Sebastien Moretti's avatar
Sebastien Moretti committed
345
            my $protGI = blastPAcc2PGI($equivalent_blast_hits[$qq]);
346
            if ( $protGI eq '' ){
Sebastien Moretti's avatar
Sebastien Moretti committed
347
348
                print "\tNo protein (prot GI) link found for $equivalent_blast_hits[$qq] in $fasta_header\n\n";
                @failureStatus = ('PUI_unavailable', $equivalent_blast_hits[$qq]);
349
                next HIT_LINK;
350
            }
Sebastien Moretti's avatar
Sebastien Moretti committed
351

Sebastien Moretti's avatar
Sebastien Moretti committed
352
            # PUID -> nt GIs & GeneID
Sebastien Moretti's avatar
Sebastien Moretti committed
353
            my ($ntGIs, $geneID) = protGI2NTGIs($protGI, $equivalent_blast_hits[$qq]);
354
            if ( $ntGIs eq '' && $geneID eq '' ){
Sebastien Moretti's avatar
Sebastien Moretti committed
355
356
                print "\tNo nucleotide (nt GIs) link found for $equivalent_blast_hits[$qq] in $fasta_header\n\n";
                @failureStatus = ('No_nt_link', $equivalent_blast_hits[$qq]);
357
                next HIT_LINK;
358
359
360
361
            }
            print " linked with [$ntGIs $geneID]\n";


Sebastien Moretti's avatar
Sebastien Moretti committed
362
            # Get transcript and contig seq
Sebastien Moretti's avatar
Sebastien Moretti committed
363
            if ( $ntGIs ne '' ){
Sebastien Moretti's avatar
Sebastien Moretti committed
364
                @nt_GIs = split(/,/, $ntGIs);
Sebastien Moretti's avatar
Sebastien Moretti committed
365
366
                downloadSeqFromGIs($cache, '', '', @nt_GIs);
            }
367

Sebastien Moretti's avatar
Sebastien Moretti committed
368
            # GeneID -> Chr
369
370
371
372
373
374
375
            if ( $geneID ne '' ){
                my @geneIDs = split(/,/, $geneID);
                while(<@geneIDs>){
                    my $geneID = $_;
                    my ($chr, $amont, $aval) = geneID2Chr($geneID, $equivalent_blast_hits[$qq]);
                    print "\n\tNo gene locus found for $equivalent_blast_hits[$qq] with $geneID in $fasta_header\n\n" if ($chr eq '' and $geneID ne '');

Sebastien Moretti's avatar
Sebastien Moretti committed
376
                    # Get Chr seq
377
378
                    if ( $chr ne '' ){
                        download_seq($cache, $amont, $aval, $chr);
Sebastien Moretti's avatar
Sebastien Moretti committed
379
                        $intronStep = 1;
380

Sebastien Moretti's avatar
Sebastien Moretti committed
381
                        unshift(@nt_GIs, "$chr:$amont-$aval");
382
                    }
383
384
385
386
387
                }
            }
        }


Sebastien Moretti's avatar
Sebastien Moretti committed
388
389
        # Align Nt seq with our protein query seq
        my ($POS, $BOJ) = runExonerate($cache, $date, $r, $original_seq[$r], $fasta_header, @nt_GIs);
390
        next HIT_LINK if ( !defined $POS );
391

Sebastien Moretti's avatar
Sebastien Moretti committed
392
        my ($POSresult, $bestNucleotide) = prepareResults4CDS($POS, $equivalent_blast_hits[$qq], $original_seq[$r], $fasta_header);
393
394
        $resultPOS .= $POSresult if ( $bestNucleotide ne '' && $POSresult ne '' && $resultPOS !~ /_G_$bestNucleotide/ );

395
396
        @intronStatus = ($POS->{0}, $equivalent_blast_hits[$qq], '')              if ( $intronStep==1 && exists($POS->{0}) && $POS->{0} =~ /\:/ );
        @intronStatus = ($POS->{0}, $equivalent_blast_hits[$qq], $bestNucleotide) if ( $intronStep==0 && $bestNucleotide =~ /^NA[CTWZGS]_/ );
397
398
399

        my ($intronLess, $nameLess) = ('', '');
        $intronLess = $POS->{0}       if ( %$BOJ eq 0 && exists($POS->{0}) && $POS->{0} ne '' );
400
        $nameLess   = $bestNucleotide if ( %$BOJ eq 0 && $intronStep==0 && $bestNucleotide =~ /^NA[CTWZGS]_/ );
Sebastien Moretti's avatar
Sebastien Moretti committed
401
        my ($BOJresult, $bestGenomic) = prepareResults4BOJ($BOJ, $equivalent_blast_hits[$qq], $original_seq[$r], $fasta_header, $intronLess, $nameLess);
402
        $resultBOJ .= $BOJresult      if ( $bestGenomic ne '' && $BOJresult ne '' && $resultBOJ !~ /_G_$bestGenomic[ \-]/ );
Sebastien Moretti's avatar
Sebastien Moretti committed
403
404
405
    }


406
    if ( $resultPOS eq '' ){
407
        if ( $failureStatus[0] ne '' ){
408
            buildFailureOutputFiles($r, $failureStatus[1],         $failureStatus[0]);
409
410
        }
        else {
411
            buildFailureOutputFiles($r, $equivalent_blast_hits[0], 'Alignment_failure');
412
        }
413
        next EACH_SEQ;
Sebastien Moretti's avatar
Sebastien Moretti committed
414
    }
415
416
    elsif ( $resultBOJ eq '' ){
        if ( $failureStatus[0] eq 'PUI_unavailable' || $failureStatus[0] eq 'No_nt_link' ){
417
            buildFailureBOJOutputFile($failureStatus[1], $failureStatus[0], $fasta_header, $original_seq[$r]);
418
419
        }
        elsif ( $failureStatus[0] eq '' && $intronStatus[0] ne '' ){
420
            buildIntronlessBOJOutputFile($intronStatus[1], $intronStatus[0], $intronStatus[2], $fasta_header, $original_seq[$r]);
421
422
        }
        else{
Sebastien Moretti's avatar
Sebastien Moretti committed
423
            buildFailureBOJOutputFile($equivalent_blast_hits[0], 'Unavailable', $fasta_header, $original_seq[$r]);
424
        }
425
        createCDSOutputFile($resultPOS, $original_seq[$r], $fasta_header);
Sebastien Moretti's avatar
Sebastien Moretti committed
426
427
    }
    else {
428
        createCDSOutputFile($resultPOS, $original_seq[$r], $fasta_header);
429
        createBOJOutputFile($resultBOJ);
Sebastien Moretti's avatar
Sebastien Moretti committed
430
431
432
433
434
435
436
437
    }


    undef $original_seq[$r];
    undef $original_names[$r];
}


Sebastien Moretti's avatar
Sebastien Moretti committed
438
checkAndCleanStderrFiles("$cache/$date.ExonerateError") if ( -e "$cache/$date.ExonerateError" );
Sebastien Moretti's avatar
Sebastien Moretti committed
439
checkOutputFiles();
Sebastien Moretti's avatar
Sebastien Moretti committed
440
unlink("$msa") if ( $converted==1 );
441
unlink 'blast_result.txt';
Sebastien Moretti's avatar
Sebastien Moretti committed
442
my @tmpFiles = glob($cache."/$date*");
Sebastien Moretti's avatar
Sebastien Moretti committed
443
if ( exists($tmpFiles[0]) ){
Sebastien Moretti's avatar
Sebastien Moretti committed
444
445
    mkdir("$cache/$date.TMP");
    move("$_", "$cache/$date.TMP") while(<@tmpFiles>);
Sebastien Moretti's avatar
Sebastien Moretti committed
446
447
448
449
450
451
452
}


#Ouput Checker


#STDOUT Log output for our server
Sebastien Moretti's avatar
Sebastien Moretti committed
453
if ( -s "$originalMSA.cds" || -s "$originalMSA.out" ){
Sebastien Moretti's avatar
Sebastien Moretti committed
454
    print "\n\nTERMINATION STATUS: SUCCESS\n";
455
    print 'OUTPUT RESULTS', "\n";
Sebastien Moretti's avatar
Sebastien Moretti committed
456
    print "    #### File Type=        MSA Format= fasta_CDS Name= $originalMSA.cds\n" 
457
        if ( -s "$originalMSA.cds" );
Sebastien Moretti's avatar
Sebastien Moretti committed
458
    print "    #### File Type=        MSA Format= fasta_CDS+Query Name= $originalMSA.cdsP\n" 
459
        if ( -s "$originalMSA.cdsP" );
Sebastien Moretti's avatar
Sebastien Moretti committed
460
    print "    #### File Type=        MSA Format= colored_fasta_CDS Name= $originalMSA.cdsP.html\n" 
461
        if ( Views::Html("$originalMSA.cdsP") && -s "$originalMSA.cdsP.html" );
Sebastien Moretti's avatar
Sebastien Moretti committed
462
    print "    #### File Type=        MSA Format= Rejected_seq Name= $originalMSA.out\n" 
463
        if ( -s "$originalMSA.out" );
Sebastien Moretti's avatar
Sebastien Moretti committed
464
    print "    #### File Type=        MSA Format= fasta_Exon-Boundaries Name= $originalMSA.boj\n" 
465
        if ( -s "$originalMSA.boj" && -s "$originalMSA.cds" && $hideBOJ==0 );
Sebastien Moretti's avatar
Sebastien Moretti committed
466
467

    unlink "$cache/web_tempo.result";
Sebastien Moretti's avatar
Sebastien Moretti committed
468
469
}
else{
Sebastien Moretti's avatar
Sebastien Moretti committed
470
    failure();
471
    exit(1);
Sebastien Moretti's avatar
Sebastien Moretti committed
472
473
}

474
exit(0);
Sebastien Moretti's avatar
Sebastien Moretti committed
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


=head1 NAME

ProtoGene.pl - Converts a peptidic alignment to a nucleotidic alignment through database search

=head1 SYNOPSIS

B<ProtoGene.pl> --msa=path_of_the_fasta_msa_file [options]

with these options available:

=over 4

=item I<--orgm>=targeted species in the database search

=item All_organisms [default], Bacteria, Viruses, Vertebrata, Eukaryota, Mammalia, Primates, Homo_sapiens, Gallus_gallus, Bos_taurus, Escherichia_coli, Arabidopsis_thaliana, Mus_musculus, Drosophila_Melanogaster, ...

=item I<--db>=targeted database in the database search

=item nr, pdb, swissprot, refseq_protein [default]

=item I<--local> to execute a local BLAST query with  --db=path_for_a_local_db_blast_formated

=item I<--template> to provide your own nucleotidic sequences following the cds file format

=item I<--revtrans> to reverse-translate sequences with no blast hit, in IUB (IUPAC) depiction code

=item They are removed from the alignement by default

=item I<--pep> to add the original peptide query beneath the back-translated sequence

=item I<--cache>=none, update, use, own_path_directory, old, empty

=item none: no cache usage, none temporary files are stored

=item update: use cache but update old files [default]

=item use: force use cache, whatever the age of files

=item own_path_directory: use my own directory, and its files

=item old: remove the old files in the cache directory

519
=item empty: empty completely the cache directory
Sebastien Moretti's avatar
Sebastien Moretti committed
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540

=item I<--version> to print version information

=item I<--help> to print this help message

=back

=head1 DESCRIPTION

PROTOGENE : Bona-Fide Back Translation of Protein Multiple Sequence Alignments
It converts a peptidic alignment to a nucleotidic alignment through database search.
Blast queries allow to get nucleotidic sequences from where peptidic sequences come from.
Exonerate aligns nucleotidic and peptidic sequences together.
PROTOGENE re-builds the original alignment with nucleotidic information it has gotten.

=head1 REQUIREMENT

=over 4

=item B<Perl 5.6 or better> is required !

541
=item Standard Perl modules B<lib>, B<strict>,  B<warnings>,  B<diagnostics>, B<Carp> are required
Sebastien Moretti's avatar
Sebastien Moretti committed
542

543
=item as well as some other current ones : B<Getopt::Long>,  B<File::Which>,  B<File::Copy>, B<Mail::Send>, B<Time::localtime>, B<LWP::Simple>
Sebastien Moretti's avatar
Sebastien Moretti committed
544
545
546
547
548
549
550
551
552
553
554
555
556

=item -

=item I<exonerate> from http://www.ebi.ac.uk/~guy/exonerate/

=item I<blast> from http://www.ncbi.nlm.nih.gov/BLAST/download.shtml or http://blast.wustl.edu/

=back

=head1 VERSION

=over 8

Sebastien Moretti's avatar
Sebastien Moretti committed
557
=item version 4.1.0
Sebastien Moretti's avatar
Sebastien Moretti committed
558

Sebastien Moretti's avatar
Sebastien Moretti committed
559
=item on Mar 15th, 2012
Sebastien Moretti's avatar
Sebastien Moretti committed
560
561
562
563
564
565
566
567
568

=back

=head1 AUTHORS

=over 8

=item Sebastien MORETTI

Sebastien Moretti's avatar
Sebastien Moretti committed
569
=item moretti.sebastien [AT] gmail.com
Sebastien Moretti's avatar
Sebastien Moretti committed
570

571
=item Vital-IT computing center
Sebastien Moretti's avatar
Sebastien Moretti committed
572

573
=item Swiss Institute of Bioinformatics
Sebastien Moretti's avatar
Sebastien Moretti committed
574

575
=item Lausanne, Switzerland
Sebastien Moretti's avatar
Sebastien Moretti committed
576

577
=item http://www.vital-it.ch/
Sebastien Moretti's avatar
Sebastien Moretti committed
578
579
580
581
582
583
584
585

=back

=cut


######################  Management  ######################

586
# Failure declaration
Sebastien Moretti's avatar
Sebastien Moretti committed
587
sub failure {
588
589
590
    print {*STDERR} "\n\nFATAL\n\nTERMINATION STATUS: FAILURE\n\n";
    return;
}
Sebastien Moretti's avatar
Sebastien Moretti committed
591

Sebastien Moretti's avatar
Sebastien Moretti committed
592
# Check external programs presence in the PATH
593
sub checkExternalSoftware {
Sebastien Moretti's avatar
Sebastien Moretti committed
594
    for my $exe ( @_ ){
595
        if ( ! which($exe) && !-x $exe ){
Sebastien Moretti's avatar
Sebastien Moretti committed
596
597
598
599
            failure();
            print {*STDERR} "\t'$exe' program is not reachable\n\tIt could not be in your PATH or not installed\n\n";
            exit(1);
        }
Sebastien Moretti's avatar
Sebastien Moretti committed
600
    }
Sebastien Moretti's avatar
Sebastien Moretti committed
601
    return;
Sebastien Moretti's avatar
Sebastien Moretti committed
602
603
}

604
605
# Check cache directory accessibility: Test and change rights, if required, of the cache directory
sub checkCacheAccessibility {
606
    my ($cacheDir) = @_;
Sebastien Moretti's avatar
Sebastien Moretti committed
607

608
    if ( -d "$cacheDir/" && -w "$cacheDir/" && -x "$cacheDir/" && -r "$cacheDir/" ){
609
        # Good
Sebastien Moretti's avatar
Sebastien Moretti committed
610
    }
611
612
613
    elsif ( -d "$cacheDir/" ){
        if ( -o "$cacheDir/" ){
            chmod 0754, "$cacheDir/";
614
        }
615
        else{
616
617
618
            failure();
            print {*STDERR} "\tPermissions do NOT seem to be valid for the '$cacheDir' cache directory\n\n";
            exit(2);
619
        }
Sebastien Moretti's avatar
Sebastien Moretti committed
620
621
    }
    else{
622
        mkdir "$cacheDir/"; # FIXME: mkdir is not recursive !
623
        if ( -o "$cacheDir/" ){
624
            chmod 0754, "$cacheDir/"; #0754 is in octal, so 754 or '0754' are not good !
625
626
        }
        else{
627
628
629
            failure();
            print {*STDERR} "\tPermissions do NOT seem to be valid for the '$cacheDir' cache directory\n\n";
            exit(2);
630
        }
Sebastien Moretti's avatar
Sebastien Moretti committed
631
    }
632

633
    return;
Sebastien Moretti's avatar
Sebastien Moretti committed
634
635
}

636
# Cache Behavior Management
Sebastien Moretti's avatar
Sebastien Moretti committed
637
sub cacheManagement{
638
    my ($cache, $cacheDir, $inScript) = @_;
639

640
    if ( $cache eq 'empty' ){
641
        unlink <$cacheDir/*.fas>;
Sebastien Moretti's avatar
Sebastien Moretti committed
642
        unlink <$cacheDir/*.ExonerateError>;
643
644
        unlink("$cacheDir/webblast.log", "$cacheDir/web_tempo.result", "$cacheDir/debug.tempo");
        print "\n\tcache directory [$cacheDir] is empty\n\n";
645
        exit 0;
Sebastien Moretti's avatar
Sebastien Moretti committed
646
    }
647
    elsif ( $cache eq 'old' ){
648
649
650
651
652
        unlink    grep { -M $_ > $cacheStorageTime }    glob($cacheDir.'/*.fas'); # Grep list only with $cacheStorageTime days old files

        if ( $inScript==0 ){
            print "\n\tcache directory [$cacheDir] is empty of its old files\n\n";
            exit 0;
653
        }
Sebastien Moretti's avatar
Sebastien Moretti committed
654
    }
655
    elsif ( $cache eq 'update' || $cache eq 'use' ){
656
657
        # DEFAULT: Limit cached files to less than $cacheStorageTime days old ones for 'update'
        $cache = $cacheDir;
Sebastien Moretti's avatar
Sebastien Moretti committed
658
    }
659
    elsif ( $cache eq 'none' ){
660
        # Do not use cache
Sebastien Moretti's avatar
Sebastien Moretti committed
661
    }
662
663
664
665
    elsif ( -d $cache ){
        die "\n\tProtoGene cannot access to your directory\n\n"  if ( ! -x $cache );
        die "\n\tProtoGene cannot write into your directory\n\n" if ( ! -w $cache );
        die "\n\tProtoGene cannot read into your directory\n\n"  if ( ! -r $cache );
Sebastien Moretti's avatar
Sebastien Moretti committed
666
667
    }
    else {
668
669
670
        failure();
        print {*STDERR} "\n\tWrong cache argument, or your cache folder '$cacheDir' doesn't seem to exist\n\n";
        exit(3);
Sebastien Moretti's avatar
Sebastien Moretti committed
671
672
673
674
675
676
677
678
679
680
    }

    return($cache);
}

##########################################################


######################    Fasta    ######################

Sebastien Moretti's avatar
Sebastien Moretti committed
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
# Check input aln format
sub isFastaFile{
    my ($msa, $converted) = @_;

    if ( -s "$msa" ){ # Check is the file exists and is not empty
        # Is FASTA format ?
        my $firstLine = `head $msa`;
        return($msa, 0) if ( $firstLine =~ /^>/ );

        # Convert no fasta format to fasta with readseq
        my $reformat = which('readseq') || '';
        if ( $reformat ne '' ){
            system("$reformat -a -f8 $msa > $msa.007")==0 or do {failure(); print {*STDERR} "\n\tEnable to convert your file to fasta format\n\n";};
            $msa      .= '.007';
            $converted = 1;
696
        }
Sebastien Moretti's avatar
Sebastien Moretti committed
697
698
    }
    else {
Sebastien Moretti's avatar
Sebastien Moretti committed
699
        failure();
700
701
        print {*STDERR} "\tThe file \'$msa\' doesn't not seem to be reachable\n\n";
        exit(1);
Sebastien Moretti's avatar
Sebastien Moretti committed
702
703
    }

Sebastien Moretti's avatar
Sebastien Moretti committed
704
    return($msa, $converted);
Sebastien Moretti's avatar
Sebastien Moretti committed
705
706
707
708
}

#Prepare Fasta Header of the query name to incorporate our annotations
sub fastaHeaders4ProtoGene{
709
    my ($QueryName) = @_;
Sebastien Moretti's avatar
Sebastien Moretti committed
710

711
    $QueryName =~ s{  +}{ }g;
712
    my ($acc, $desc) = ($QueryName, $QueryName);
713
714
    $acc  =~ s{^ *([^ ]*) .*$}{$1};
    $acc  =~ s{\s+$}{}g;
715
716
    $acc  =  $acc.'_G_@@';

717
718
719
    $desc =~ s{^ *[^ ]* *(.*)$}{$1};
    $desc =~ s{^[\| \.]+}{};
    $desc =~ s{[\|\. ]+$}{};
720
721

    return($acc.$desc);
Sebastien Moretti's avatar
Sebastien Moretti committed
722
723
724
725
}

#Prepare Fasta Header to put it on the fasta header of the result outputs
sub getFastaHeaderAnnot{
726
    my ($bestTarget) = @_;
Sebastien Moretti's avatar
Sebastien Moretti committed
727
728

    #Get annotation from nucleotide file header
Sebastien Moretti's avatar
Sebastien Moretti committed
729
730
    open(my $BEST, '<', "$cache/$bestTarget.fas");
    my $annot = <$BEST> || '';
731
    close $BEST;
Sebastien Moretti's avatar
Sebastien Moretti committed
732

Sebastien Moretti's avatar
Sebastien Moretti committed
733
    chomp($annot);
734
735
736
737
    $annot =~ s{^ *[^ ]* }{};
    $annot =~ s{[\. ]*$}{};
    $annot =~ s{^ *}{};
    $annot =~ s{  +}{ }g;
Sebastien Moretti's avatar
Sebastien Moretti committed
738
739
740
741
742
743
744
745
746
747
748

    return($annot);
}

#########################################################


###################### Translation ######################

#Direct Reverse translation
sub reverse_trad{
749
    my ($aa) = @_;
Sebastien Moretti's avatar
Sebastien Moretti committed
750

751
752
753
754
755
756
    my %reverse_code = ('A' => 'GCN', 'F' => 'TTy', 'K' => 'AAr', 'P' => 'CCN', 'T' => 'ACN',
                        'C' => 'TGy', 'G' => 'GGN', 'L' => 'yTN', 'Q' => 'CAr', 'V' => 'GTN',
                        'D' => 'GAy', 'H' => 'CAy', 'M' => 'ATG', 'R' => 'mGN', 'W' => 'TGG',
                        'E' => 'GAr', 'I' => 'ATh', 'N' => 'AAy', 'S' => 'wsN', 'Y' => 'TAy',
                        'X' => 'NNN', '-' => '---', '*' => 'Trr',
                       ); # '*' is for stop codon
Sebastien Moretti's avatar
Sebastien Moretti committed
757

758
    my $triplet = 'nnn';
759
    while( my ($x, $y) = each(%reverse_code) ){
760
761
762
763
        if ( uc($aa) eq $x ){
            $triplet = $y;
            last;
        }
Sebastien Moretti's avatar
Sebastien Moretti committed
764
765
766
767
768
769
770
771
772
    }

    return($triplet);
}

#########################################################


####################### webblast #######################
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
sub display_blast_param {
    print "
             Program :                   BLASTp
             Evalue threshold :          $blast_param->{'evalue'}
             Matrix :                    $blast_param->{'matrix'}
             Filter :                    $blast_param->{'filter'}
             Equivalent best hits used : $blast_param->{'nbesthits'}

             Database:                   NCBI $blast_param->{'db1'}
             Blast_identity_threshold :  $blast_param->{'identity1'}
             Cover threshold :           $blast_param->{'cover1'}
          if no match
             Database:                   NCBI $blast_param->{'db2'}
             Blast_identity_threshold :   $blast_param->{'identity2'}
             Cover threshold :            $blast_param->{'cover2'}
\n";
    return;
}

sub run_BLAST{
    my ($infile, $whatBlast, $db, $blast_exe, $outfile) = @_;

    my $blast_at = $whatBlast==2  ? '-gigablast=yes'
                 : $whatBlast==1  ? "-blast_dir=$blast_exe"
                 :                  '';

    # BLAST run 1
    system("$webblast_exe -program=blastp -database=$db $blast_at -infile=$infile -matrix=$blast_param->{'matrix'} -evalue=$blast_param->{'evalue'} -method=geneid -filter=$blast_param->{'filter'} -organism=$species -identity=$blast_param->{'identity1'} -cover=$blast_param->{'cover1'} -hits=$blast_param->{'nbesthits'} -quiet=on -outfile=$outfile");


    if ( -z "$outfile" ){
    # BLAST run 2 if no hit with 100% identity and coverage: refseq100 -> nr95  OR  nr100 -> nr95
Sebastien Moretti's avatar
Sebastien Moretti committed
805
        print "run BLAST.. against $blast_param->{'db2'} because there was no acceptable hit against $blast_param->{'db1'}\n"
806
            if ( $whatBlast != 1 && $db ne 'nr' );
Sebastien Moretti's avatar
Sebastien Moretti committed
807
        print "run BLAST.. again, with decreased thresholds, because there was no acceptable hit against $db\n"
808
            if ( $whatBlast == 1 || $db eq 'nr' );
809
810

        system("$webblast_exe -program=blastp -database=$blast_param->{'db2'} $blast_at -infile=$infile -matrix=$blast_param->{'matrix'} -evalue=$blast_param->{'evalue'} -method=geneid -filter=$blast_param->{'filter'} -organism=$species -identity=$blast_param->{'identity2'} -cover=$blast_param->{'cover2'} -hits=$blast_param->{'nbesthits'} -quiet=on -outfile=$outfile");
Sebastien Moretti's avatar
Sebastien Moretti committed
811
812
    }

813
814
815
816
817
818
819
    move('webblast.log',     "$cache"); #Move temporary webblast.pl files
    move('web_tempo.result', "$cache");
    move('debug.tempo',      "$cache");

    unlink("$cache/webblast.log")     if ( -z "$cache/webblast.log" );
    unlink("$cache/web_tempo.result") if ( -z "$cache/web_tempo.result" );
    unlink("$cache/debug.tempo")      if ( -z "$cache/debug.tempo" );
Sebastien Moretti's avatar
Sebastien Moretti committed
820

821
    return;
Sebastien Moretti's avatar
Sebastien Moretti committed
822
823
824
825
826
827
828
}


########################################################


##################### NCBI requests #####################
829
830
831
sub fetch {
    my ($url) = @_;

832
    XML:
833
834
    for (my $tries=0; $tries <20; $tries++ ){
        my $content = get($url);
835
        print {*STDERR} "[$content]\n\n" if ($debug);
836
837
838
839
        if ( defined $content ){
            next XML if ( $content =~ /<ERROR>/i );
            return $content;
        }
840
    }
Sebastien Moretti's avatar
Sebastien Moretti committed
841
    print {*STDERR} "\n\tERROR: Problem with NCBI eutils, please try again later\n";
842
    print {*STDOUT} "\n\tERROR: Problem with NCBI eutils, please try again later\n";
Sebastien Moretti's avatar
Sebastien Moretti committed
843
    exit(4);
844
}
845

846
847
848
sub fetch_fasta {
    my ($url, $outfile) = @_;

849
    return 1 if ( -s "$outfile" && $cache eq 'update' && -M "$outfile" <= $cacheStorageTime );
850

851
852
853
854
855
856
857
858
859
860
861
    GET_FASTA:
    for (my $tries=0; $tries <20; $tries++ ){
        my $HTTP_status = getstore($url, $outfile);
        if ( is_error($HTTP_status) ){
            next GET_FASTA;
        }
        else {
            open(my $FASTA, '<', "$outfile");
            my $counter = 0;
            my $lines   = 0;
            while(<$FASTA>){
862
#                $lines++              if ( $_ !~ /^>/ ); #Allow download of entries with only fasta header and no sequence associated
863
864
865
866
867
                $counter = $counter+2 if ( $counter==1 && $_ !~ /^\w/ && $lines==1 && $_ !~ /^>/ );
                $counter++            if ( $_ =~ /^>/ );
                $counter = $counter+2 if ( $_ !~ /^>/ && ($_ =~ /Error:/ || $_ =~ /[<>]/) );
            }
            close $FASTA;
868

869
870
871
            return 1 if ( $counter == 1 );
        }
    }
872
    print {*STDERR} "[[$url]]\n" if ( $debug );
873
    unlink("$outfile");
Sebastien Moretti's avatar
Sebastien Moretti committed
874
    print {*STDERR} "\n\tERROR: Problem with NCBI eutils, please try again later.\n";
875
    print {*STDOUT} "\n\tERROR: Problem with NCBI eutils, please try again later.\n";
Sebastien Moretti's avatar
Sebastien Moretti committed
876
    exit(5);
877
878
}

Sebastien Moretti's avatar
Sebastien Moretti committed
879
880
881

#Prot ACC -> PUID
sub blastPAcc2PGI{
882
    my ($blastHit) = @_;
Sebastien Moretti's avatar
Sebastien Moretti committed
883

884
    my $protGI = '';
885
    #FIXME: should be ${blastHit}[pacc] but something is broken at NCBI
886
    my $content = fetch("http://eutils.ncbi.nlm.nih.gov/entrez/eutils/esearch.fcgi?db=protein&term=$blastHit&retmode=xml&tool=ProtoGene&email=smoretti\@unil.ch");
887
888
    if ( $content =~ /<Id>(\d+)<\/Id>/ ){
        $protGI = $1;
Sebastien Moretti's avatar
Sebastien Moretti committed
889
890
891
892
893
894
895
    }

    return($protGI);
}

#PUID -> nt GIs
sub protGI2NTGIs{
896
    my ($protGI, $blastHit) = @_;
Sebastien Moretti's avatar
Sebastien Moretti committed
897

898
899
    my $ntGIs  = '';
    my $geneID = '';
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
    my $content = fetch("http://eutils.ncbi.nlm.nih.gov/entrez/eutils/elink.fcgi?dbfrom=protein&db=nuccore,nucleotide,gene&id=$protGI&retmode=xml&tool=ProtoGene&email=smoretti\@unil.ch");

    my @xml  = split("\n", $content);
    my $flag = 0;
    for my $line (@xml){
        if (    $line =~ /<LinkName>protein_nuc[a-z]+<\/LinkName>/ ){ #for nuccleotide and nuccore
            $flag = 1;
        }
        elsif ( $line =~ /<LinkName>protein_gene<\/LinkName>/ ){
            $flag  = 2;
        }
        elsif ( $flag==1 && $line =~ /^.*<Id>(\d+)<\/Id>.*$/ ){
            my $match = $1;
            $ntGIs  .= "$match,$match,";
        }
        elsif ( $flag==2 && $line =~ /^.*<Id>(\d+)<\/Id>.*$/ ){
            my $match = $1;
            $geneID .= "$match,$match,";
918
        }
Sebastien Moretti's avatar
Sebastien Moretti committed
919
920
    }

921
    #Remove redundancy if any
922
    chop $ntGIs; #Remove last ','
923
924
925
926
927
928
    my %hash_NT = split(',', $ntGIs);
    $ntGIs      = join(',',  keys(%hash_NT) );
    chop $geneID;
    my %hash_GE = split(',', $geneID);
    $geneID     = join(',',  keys(%hash_GE) );

929
    return($ntGIs, $geneID);
Sebastien Moretti's avatar
Sebastien Moretti committed
930
931
932
}

sub geneID2Chr{
933
    my ($geneID, $blastHit) = @_;
Sebastien Moretti's avatar
Sebastien Moretti committed
934

935
936
    my $chr   = '';
    my ($amont, $aval) = ('', '');
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
#   my $content = fetch("http://eutils.ncbi.nlm.nih.gov/entrez/eutils/efetch.fcgi?db=gene&id=$geneID&retmode=xml&tool=ProtoGene&email=smoretti\@unil.ch");
    my $content = fetch("http://eutils.ncbi.nlm.nih.gov/entrez/eutils/esummary.fcgi?db=gene&id=$geneID&retmode=xml&tool=ProtoGene&email=smoretti\@unil.ch");

    my @xml  = split("\n", $content);
    my $flag = 0;
    GENE_INFO:
    for my $line (@xml){
        if ( $flag==0    && $line =~ /<Item Name="GenomicInfoType" Type="Structure">/ ){
            $flag = 1;
        }
        elsif ( $flag==1 && $line =~ /<Item Name="ChrAccVer" Type="String">([^<]+?)\.?\d*<\/Item>/ ){
            $chr   = $1;
            $flag  = 3;
        }
        elsif ( $flag==3 && $line =~ /<Item Name="ChrStart" Type="Integer">(\d+)<\/Item>/ ){
            $amont = $1;
            $flag  = 4;
        }
        elsif ( $flag==4 && $line =~ /<Item Name="ChrStop" Type="Integer">(\d+)<\/Item>/ ){
            $aval  = $1;
            last GENE_INFO;
958
        }
Sebastien Moretti's avatar
Sebastien Moretti committed
959
    }
960

961
962
    if ( $amont eq '' || $aval eq '' ){
        $amont = '';
963
        $aval  = '';
Sebastien Moretti's avatar
Sebastien Moretti committed
964
    }
965
966
    elsif ( $amont ne '' && $aval ne '' ){
        if ( $amont > $aval ){
967
968
            $amont = $amont + 5000;
            $aval  = $aval  - 5000;
969
970
        }
        else{
971
972
            $amont = $amont - 5000;
            $aval  = $aval  + 5000;
973
        }
Sebastien Moretti's avatar
Sebastien Moretti committed
974
    }
975
    $amont = 1 if ( $amont ne '' && $amont <= 0 );
976
    $aval  = 1 if ( $aval  ne '' && $aval  <= 0 );
Sebastien Moretti's avatar
Sebastien Moretti committed
977

978
    return($chr, $amont, $aval);
Sebastien Moretti's avatar
Sebastien Moretti committed
979
980
981
}

sub downloadSeqFromGIs{
982
    my ($cache, $amont, $aval, @acc) = @_;
Sebastien Moretti's avatar
Sebastien Moretti committed
983

984
    GET_SEQ:
Sebastien Moretti's avatar
Sebastien Moretti committed
985
    for(my $a=0; $a<=$#acc; $a++){
986
        if ( $amont =~ /^\d+$/ && $aval =~ /^\d+$/ ){
987
            fetch_fasta("http://eutils.ncbi.nlm.nih.gov/entrez/eutils/efetch.fcgi?db=nucleotide&id=$acc[$a]&rettype=fasta&retmode=text&from=$amont&to=$aval&tool=ProtoGene&email=smoretti\@unil.ch", "$cache/$acc[$a]-$amont.fas") ;
988
989
        }
        else{
990
            fetch_fasta("http://eutils.ncbi.nlm.nih.gov/entrez/eutils/efetch.fcgi?db=nucleotide&id=$acc[$a]&rettype=fasta&retmode=text&tool=ProtoGene&email=smoretti\@unil.ch'", "$cache/$acc[$a].fas");
991
        }
Sebastien Moretti's avatar
Sebastien Moretti committed
992
    }
993
994

    return;
Sebastien Moretti's avatar
Sebastien Moretti committed
995
996
}

Sebastien Moretti's avatar
Sebastien Moretti committed
997
sub download_seq{
Sebastien Moretti's avatar