Parent Directory
| 
Revision Log
Revision 1.30 - (view) (download) (as text)
| 1 : | olson | 1.28 | # |
| 2 : | # Copyright (c) 2003-2006 University of Chicago and Fellowship | ||
| 3 : | # for Interpretations of Genomes. All Rights Reserved. | ||
| 4 : | # | ||
| 5 : | # This file is part of the SEED Toolkit. | ||
| 6 : | # | ||
| 7 : | # The SEED Toolkit is free software. You can redistribute | ||
| 8 : | # it and/or modify it under the terms of the SEED Toolkit | ||
| 9 : | # Public License. | ||
| 10 : | # | ||
| 11 : | # You should have received a copy of the SEED Toolkit Public License | ||
| 12 : | # along with this program; if not write to the University of Chicago | ||
| 13 : | # at info@ci.uchicago.edu or the Fellowship for Interpretation of | ||
| 14 : | # Genomes at veronika@thefig.info or download a copy from | ||
| 15 : | # http://www.theseed.org/LICENSE.TXT. | ||
| 16 : | # | ||
| 17 : | |||
| 18 : | redwards | 1.1 | # -*- perl -*- |
| 19 : | |||
| 20 : | =pod | ||
| 21 : | |||
| 22 : | parrello | 1.13 | =head1 RAE Library |
| 23 : | redwards | 1.1 | |
| 24 : | Some routines and things that Rob uses. Please feel free to use at will and incorporate into | ||
| 25 : | your own code or move them into FIG.pm or elsewhere. | ||
| 26 : | |||
| 27 : | =cut | ||
| 28 : | |||
| 29 : | package raelib; | ||
| 30 : | use strict; | ||
| 31 : | redwards | 1.17 | use Bio::SeqIO; |
| 32 : | use Bio::Seq; | ||
| 33 : | use Bio::SeqFeature::Generic; | ||
| 34 : | redwards | 1.1 | use FIG; |
| 35 : | my $fig=new FIG; | ||
| 36 : | |||
| 37 : | redwards | 1.5 | =head2 new |
| 38 : | |||
| 39 : | Just instantiate the object and return $self | ||
| 40 : | |||
| 41 : | =cut | ||
| 42 : | |||
| 43 : | sub new { | ||
| 44 : | redwards | 1.27 | my ($class)=@_; |
| 45 : | my $self={}; | ||
| 46 : | return bless $self, $class; | ||
| 47 : | redwards | 1.5 | } |
| 48 : | |||
| 49 : | |||
| 50 : | |||
| 51 : | |||
| 52 : | redwards | 1.4 | =head2 features_on_contig |
| 53 : | |||
| 54 : | Returns a reference to an array containing all the features on a contig in a genome. | ||
| 55 : | |||
| 56 : | use: | ||
| 57 : | |||
| 58 : | my $arrayref=$rae->features_on_contig($genome, $contig); | ||
| 59 : | |||
| 60 : | or | ||
| 61 : | |||
| 62 : | foreach my $peg (@{$rae->features_on_contig($genome, $contig)}) { | ||
| 63 : | ... blah blah ... | ||
| 64 : | } | ||
| 65 : | |||
| 66 : | returns undef if contig is not a part of genome or there is nothing to return, otherwise returns a list of pegs | ||
| 67 : | |||
| 68 : | v. experimental and guaranteed not to work! | ||
| 69 : | |||
| 70 : | =cut | ||
| 71 : | |||
| 72 : | sub features_on_contig { | ||
| 73 : | my ($self, $genome, $contig)=@_; | ||
| 74 : | # were this in FIG.pm you'd use this line: | ||
| 75 : | #my $rdbH = $self->db_handle; | ||
| 76 : | |||
| 77 : | my $rdbH = $fig->db_handle; | ||
| 78 : | my $relational_db_response=$rdbH->SQL('SELECT id FROM features WHERE (genome = \'' . $genome . '\' AND location ~* \'' . $contig . '\')'); | ||
| 79 : | # this is complicated. A reference to an array of references to arrays, and we only want the first element. | ||
| 80 : | # simplify. | ||
| 81 : | my @results; | ||
| 82 : | foreach my $res (@$relational_db_response) {push @results, $res->[0]} | ||
| 83 : | return \@results; | ||
| 84 : | } | ||
| 85 : | |||
| 86 : | |||
| 87 : | |||
| 88 : | |||
| 89 : | |||
| 90 : | |||
| 91 : | |||
| 92 : | redwards | 1.7 | =head2 pirsfcorrespondence |
| 93 : | redwards | 1.1 | |
| 94 : | redwards | 1.18 | Generate the pirsf->fig id correspondence. This is only done once and the correspondence file is written. This is so that we can easily go back and forth. |
| 95 : | redwards | 1.1 | |
| 96 : | redwards | 1.18 | The correspondence has PIR ID \t FIG ID\n, and is probably based on ftp://ftp.pir.georgetown.edu/pir_databases/pirsf/data/pirsfinfo.dat |
| 97 : | redwards | 1.1 | |
| 98 : | redwards | 1.18 | This method takes three arguments: |
| 99 : | redwards | 1.9 | from : pirsfinfo.dat file |
| 100 : | to : file to write information to | ||
| 101 : | verbose : report on progress | ||
| 102 : | |||
| 103 : | redwards | 1.18 | Note that if the from filename ends in .gz it assumed to be a gzipped file and will be opened accordingly. |
| 104 : | |||
| 105 : | Returns the number of lines in the pirsinfo file that were read. | ||
| 106 : | redwards | 1.9 | |
| 107 : | redwards | 1.1 | =cut |
| 108 : | |||
| 109 : | redwards | 1.7 | sub pirsfcorrespondence { |
| 110 : | redwards | 1.9 | my ($self, $from, $to, $verbose)=@_; |
| 111 : | redwards | 1.10 | unless (-e $from) { |
| 112 : | print STDERR "File $from does not exist as called in $0\n"; | ||
| 113 : | return 0; | ||
| 114 : | } | ||
| 115 : | redwards | 1.18 | if ($from =~ /\.gz$/) { |
| 116 : | open(IN, "|gunzip -c $from") || die "Can't open $from using a gunzip pipe"; | ||
| 117 : | } | ||
| 118 : | else { | ||
| 119 : | open (IN, $from) || die "Can't open $from"; | ||
| 120 : | } | ||
| 121 : | redwards | 1.8 | open (OUT, ">$to") || die "Can't write to $to"; |
| 122 : | redwards | 1.9 | my $linecount; |
| 123 : | redwards | 1.1 | while (<IN>) { |
| 124 : | redwards | 1.9 | $linecount++; |
| 125 : | redwards | 1.14 | if ($verbose && !($linecount % 10000)) {print STDERR "Parsed $linecount lines\n"} |
| 126 : | redwards | 1.1 | if (/^>/) {print OUT; next} |
| 127 : | chomp; | ||
| 128 : | redwards | 1.14 | foreach my $peg ($self->swiss_pir_ids($_)) { |
| 129 : | redwards | 1.1 | print OUT $_, "\t", $peg, "\n"; |
| 130 : | } | ||
| 131 : | redwards | 1.14 | } |
| 132 : | close IN; | ||
| 133 : | close OUT; | ||
| 134 : | return $linecount; | ||
| 135 : | } | ||
| 136 : | |||
| 137 : | =head2 uniprotcorrespondence | ||
| 138 : | |||
| 139 : | Generate a correspondence table between uniprot knowledge base IDs and FIG ID's. | ||
| 140 : | |||
| 141 : | The uniprot KB file is in the form: UniProtKB_Primary_Accession | UniProtKB_ID | Section | Protein Name | ||
| 142 : | |||
| 143 : | This method takes three arguments: | ||
| 144 : | from : uniprotKB file | ||
| 145 : | to : file to write information to | ||
| 146 : | verbose : report on progress | ||
| 147 : | |||
| 148 : | redwards | 1.18 | Note that if the from filename ends in .gz it assumed to be a gzipped file and will be opened accordingly. |
| 149 : | |||
| 150 : | Returns the number of lines in the uniprotkb file that were read. | ||
| 151 : | redwards | 1.14 | |
| 152 : | =cut | ||
| 153 : | |||
| 154 : | sub uniprotcorrespondence { | ||
| 155 : | my ($self, $from, $to, $verbose)=@_; | ||
| 156 : | unless (-e $from) { | ||
| 157 : | print STDERR "File $from does not exist as called in $0\n"; | ||
| 158 : | return 0; | ||
| 159 : | } | ||
| 160 : | redwards | 1.18 | if ($from =~ /\.gz$/) { |
| 161 : | open(IN, "|gunzip -c $from") || die "Can't open $from using a gunzip pipe"; | ||
| 162 : | } | ||
| 163 : | else { | ||
| 164 : | open (IN, $from) || die "Can't open $from"; | ||
| 165 : | } | ||
| 166 : | redwards | 1.14 | open (OUT, ">$to") || die "Can't write to $to"; |
| 167 : | my $linecount; | ||
| 168 : | while (<IN>) { | ||
| 169 : | chomp; | ||
| 170 : | $linecount++; | ||
| 171 : | if ($verbose && !($linecount % 10000)) {print STDERR "Parsed $linecount lines\n"} | ||
| 172 : | my @line=split /\s+\|\s+/; | ||
| 173 : | redwards | 1.16 | my $added; |
| 174 : | redwards | 1.14 | foreach my $peg ($self->swiss_pir_ids($line[0])) { |
| 175 : | print OUT "$_ | $peg\n"; | ||
| 176 : | redwards | 1.16 | $added=1; |
| 177 : | redwards | 1.12 | } |
| 178 : | redwards | 1.16 | unless ($added) {print OUT "$_\n"} |
| 179 : | redwards | 1.1 | } |
| 180 : | close IN; | ||
| 181 : | close OUT; | ||
| 182 : | redwards | 1.9 | return $linecount; |
| 183 : | redwards | 1.1 | } |
| 184 : | |||
| 185 : | redwards | 1.18 | =head2 prositecorrespondence |
| 186 : | |||
| 187 : | Generate a correspondence table between prosite and seed using sp id's and seed ids. | ||
| 188 : | |||
| 189 : | The SwissProt prosite file is from ftp://ca.expasy.org/databases/prosite/release_with_updates/prosite.dat and is in horrible swiss prot format, so we'll parse out those things that we need and put them in the file | ||
| 190 : | |||
| 191 : | The output file will have the following columns: | ||
| 192 : | |||
| 193 : | prosite family accession number, prosite family name, family type, swiss-prot protein id, fig protein id. | ||
| 194 : | |||
| 195 : | The family type is one of rule, pattern, or matrix. Right now (Prosite Release 19.2 of 24-May-2005) there are 4 rules, 1322 patterns, and 521 matrices. | ||
| 196 : | |||
| 197 : | This method takes three arguments: | ||
| 198 : | from : prosite file | ||
| 199 : | to : file to write information to | ||
| 200 : | verbose : report on progress | ||
| 201 : | |||
| 202 : | Note that if the from filename ends in .gz it assumed to be a gzipped file and will be opened accordingly. | ||
| 203 : | |||
| 204 : | Returns the number of lines in the prosite file that were read. | ||
| 205 : | |||
| 206 : | =cut | ||
| 207 : | |||
| 208 : | sub prositecorrespondence { | ||
| 209 : | my ($self, $from, $to, $verbose)=@_; | ||
| 210 : | unless (-e $from) { | ||
| 211 : | print STDERR "File $from does not exist as called in $0\n"; | ||
| 212 : | return 0; | ||
| 213 : | } | ||
| 214 : | if ($from =~ /\.gz$/) { | ||
| 215 : | open(IN, "|gunzip -c $from") || die "Can't open $from using a gunzip pipe"; | ||
| 216 : | } | ||
| 217 : | else { | ||
| 218 : | open (IN, $from) || die "Can't open $from"; | ||
| 219 : | } | ||
| 220 : | open (OUT, ">$to") || die "Can't write to $to"; | ||
| 221 : | my $linecount; | ||
| 222 : | my ($famac, $famname, $famtype)=('','',''); | ||
| 223 : | while (<IN>) { | ||
| 224 : | chomp; | ||
| 225 : | $linecount++; | ||
| 226 : | if ($verbose && !($linecount % 10000)) {print STDERR "Parsed $linecount lines\n"} | ||
| 227 : | if (m#//#) {($famac, $famname, $famtype)=('','',''); next} | ||
| 228 : | elsif (m/^ID\s*(.*?);\s*(\S+)/) {($famname, $famtype)=($1, $2); next} | ||
| 229 : | redwards | 1.19 | elsif (m/^AC\s*(\S+)/) {$famac=$1; $famac =~ s/\;\s*$//; next} |
| 230 : | redwards | 1.18 | next unless (m/^DR/); # ignore all the other crap in the prosite file for now. Note we might, at some point, want to grab all that, but that is for another time. |
| 231 : | # | ||
| 232 : | # this is the format of the DR lines: | ||
| 233 : | # DR P11460, FATB_VIBAN , T; P40409, FEUA_BACSU , T; P37580, FHUD_BACSU , T; | ||
| 234 : | s/^DR\s*//; | ||
| 235 : | foreach my $piece (split /\s*\;\s*/, $_) { | ||
| 236 : | my ($acc, $nam, $unk)=split /\s*\,\s*/, $piece; | ||
| 237 : | foreach my $fig ($self->swiss_pir_ids($acc)) { | ||
| 238 : | redwards | 1.20 | print OUT join "\t", ($famac, $famname, $famtype, $acc, $fig), "\n"; |
| 239 : | redwards | 1.18 | } |
| 240 : | } | ||
| 241 : | } | ||
| 242 : | } | ||
| 243 : | redwards | 1.14 | |
| 244 : | =head2 swiss_pir_ids() | ||
| 245 : | |||
| 246 : | redwards | 1.18 | SwissProt/PIR have lots of ID's that we want to get, usually in this order - uni --> tr --> sp. This routine will map swissprot/pir ids to fig id's, and return an array of FIG id's that match the ID. |
| 247 : | redwards | 1.14 | |
| 248 : | =cut | ||
| 249 : | |||
| 250 : | sub swiss_pir_ids { | ||
| 251 : | my ($self, $id)=@_; | ||
| 252 : | return () unless ($id); | ||
| 253 : | redwards | 1.18 | $id =~ s/^\s+//; $id =~ s/\s+$//; # trim off the whitespace |
| 254 : | |||
| 255 : | redwards | 1.15 | my @return=($fig->by_alias("uni|$id")); |
| 256 : | redwards | 1.14 | return @return if ($return[0]); |
| 257 : | |||
| 258 : | redwards | 1.15 | @return=($fig->by_alias("tr|$id")); |
| 259 : | redwards | 1.14 | return @return if ($return[0]); |
| 260 : | |||
| 261 : | redwards | 1.15 | @return=($fig->by_alias("sp|$id")); |
| 262 : | redwards | 1.14 | return @return if ($return[0]); |
| 263 : | |||
| 264 : | return (); | ||
| 265 : | } | ||
| 266 : | redwards | 1.1 | |
| 267 : | =head2 ss_by_id | ||
| 268 : | |||
| 269 : | Generate a list of subsystems that a peg occurs in. This is a ; separated list. | ||
| 270 : | This is a wrapper that removes roles and ignores essential things | ||
| 271 : | |||
| 272 : | =cut | ||
| 273 : | |||
| 274 : | sub ss_by_id { | ||
| 275 : | my ($self, $peg)=@_; | ||
| 276 : | my $ssout; | ||
| 277 : | foreach my $ss (sort $fig->subsystems_for_peg($peg)) | ||
| 278 : | { | ||
| 279 : | next if ($$ss[0] =~ /essential/i); # Ignore the Essential B-subtilis subsystems | ||
| 280 : | $ssout.=$$ss[0]."; "; | ||
| 281 : | } | ||
| 282 : | $ssout =~ s/; $//; | ||
| 283 : | return $ssout; | ||
| 284 : | } | ||
| 285 : | |||
| 286 : | redwards | 1.3 | =head2 ss_by_homol |
| 287 : | |||
| 288 : | Generate a list of subsystems that homologs of a peg occur in. This is a ; separated list. | ||
| 289 : | This is also a wrapper around sims and ss, but makes everything unified | ||
| 290 : | |||
| 291 : | =cut | ||
| 292 : | |||
| 293 : | sub ss_by_homol { | ||
| 294 : | my ($self, $peg)=@_; | ||
| 295 : | return unless ($peg); | ||
| 296 : | my ($maxN, $maxP)=(50, 1e-20); | ||
| 297 : | |||
| 298 : | # find the sims | ||
| 299 : | my @sims=$fig->sims($peg, $maxN, $maxP, 'fig'); | ||
| 300 : | |||
| 301 : | # we are only going to keep the best hit for each peg | ||
| 302 : | # in a subsystem | ||
| 303 : | my $best_ss_score; my $best_ss_id; | ||
| 304 : | foreach my $sim (@sims) | ||
| 305 : | { | ||
| 306 : | my $simpeg=$$sim[1]; | ||
| 307 : | my $simscore=$$sim[10]; | ||
| 308 : | my @subsys=$fig->subsystems_for_peg($simpeg); | ||
| 309 : | foreach my $ss (@subsys) | ||
| 310 : | { | ||
| 311 : | if (! defined $best_ss_score->{$$ss[0]}) {$best_ss_score->{$$ss[0]}=$simscore; $best_ss_id->{$$ss[0]}=$simpeg} | ||
| 312 : | elsif ($best_ss_score->{$$ss[0]} > $simscore) | ||
| 313 : | { | ||
| 314 : | $best_ss_score->{$$ss[0]}=$simscore; | ||
| 315 : | $best_ss_id->{$$ss[0]}=$simpeg; | ||
| 316 : | } | ||
| 317 : | } | ||
| 318 : | } | ||
| 319 : | |||
| 320 : | my $ssoutput=join "", (map {"$_ (".$best_ss_id->{$_}."), "} keys %$best_ss_id); | ||
| 321 : | |||
| 322 : | $ssoutput =~ s/, $//; | ||
| 323 : | return $ssoutput; | ||
| 324 : | } | ||
| 325 : | |||
| 326 : | =head2 tagvalue | ||
| 327 : | |||
| 328 : | This will just check for tag value pairs and return either an array of values or a single ; separated list (if called as a scalar) | ||
| 329 : | |||
| 330 : | e.g. $values=raelib->tagvalue($peg, "PIRSF"); print join "\n", @$values; | ||
| 331 : | |||
| 332 : | Returns an empty array if no tag/value appropriate. | ||
| 333 : | |||
| 334 : | Just because I use this a lot I don't want to waste rewriting it. | ||
| 335 : | |||
| 336 : | =cut | ||
| 337 : | |||
| 338 : | sub tagvalue { | ||
| 339 : | my ($self, $peg, $tag)=@_; | ||
| 340 : | my @return; | ||
| 341 : | my @attr=$fig->feature_attributes($peg); | ||
| 342 : | foreach my $attr (@attr) { | ||
| 343 : | redwards | 1.11 | my ($gotpeg, $gottag, $val, $link)=@$attr; |
| 344 : | redwards | 1.3 | push @return, $val if ($gottag eq $tag); |
| 345 : | } | ||
| 346 : | return wantarray ? @return : join "; ", @return; | ||
| 347 : | } | ||
| 348 : | redwards | 1.1 | |
| 349 : | redwards | 1.5 | =head2 locations_on_contig |
| 350 : | |||
| 351 : | Return the locations of a sequence on a contig. | ||
| 352 : | |||
| 353 : | This will look for exact matches to a sequence on a contig, and return a reference to an array that has all the locations. | ||
| 354 : | |||
| 355 : | my $locations=$raelib->locations_on_contig($genome, $contig, 'GATC', undef); | ||
| 356 : | foreach my $bp (@$locations) { ... do something ... } | ||
| 357 : | |||
| 358 : | first argument : genome number | ||
| 359 : | second argument : contig name | ||
| 360 : | third argument : sequence to look for | ||
| 361 : | fourth argument : beginning position to start looking from (can be undef) | ||
| 362 : | fifth argument : end position to stop looking from (can be undef) | ||
| 363 : | sixth argument : check reverse complement (0 or undef will check forward, 1 or true will check rc) | ||
| 364 : | |||
| 365 : | Note, the position is calculated before the sequence is rc'd | ||
| 366 : | |||
| 367 : | =cut | ||
| 368 : | |||
| 369 : | sub locations_on_contig { | ||
| 370 : | my ($self, $genome, $contig, $sequence, $from, $to, $check_reverse)=@_; | ||
| 371 : | my $return=[]; | ||
| 372 : | |||
| 373 : | # get the dna sequence of the contig, and make sure it is uppercase | ||
| 374 : | my $contig_ln=$fig->contig_ln($genome, $contig); | ||
| 375 : | return $return unless ($contig_ln); | ||
| 376 : | unless ($from) {$from=1} | ||
| 377 : | unless ($to) {$to=$contig_ln} | ||
| 378 : | if ($from > $to) {($from, $to)=($to, $from)} | ||
| 379 : | my $dna_seq=$fig->dna_seq($genome, $contig."_".$from."_".$to); | ||
| 380 : | $dna_seq=uc($dna_seq); | ||
| 381 : | |||
| 382 : | # if we want to check the rc, we actually rc the query | ||
| 383 : | $sequence=$fig->reverse_comp($sequence) if ($check_reverse); | ||
| 384 : | $sequence=uc($sequence); | ||
| 385 : | |||
| 386 : | # now find all the matches | ||
| 387 : | my $posn=index($dna_seq, $sequence, 0); | ||
| 388 : | while ($posn > -1) { | ||
| 389 : | push @$return, $posn; | ||
| 390 : | $posn=index($dna_seq, $sequence, $posn+1); | ||
| 391 : | } | ||
| 392 : | return $return; | ||
| 393 : | } | ||
| 394 : | |||
| 395 : | |||
| 396 : | =head2 scrolling_org_list | ||
| 397 : | |||
| 398 : | This is the list from index.cgi, that I call often. It has one minor modification: the value returned is solely the organisms id and does not contain genus_species information. I abstracted this here: 1, so I could call it often, and 2, so I could edit it once. | ||
| 399 : | |||
| 400 : | overbeek | 1.24 | use like this push @$html, $raelib->scrolling_org_list($cgi, $multiple, $default); |
| 401 : | redwards | 1.5 | |
| 402 : | multiple selections will only be set if $multiple is true | ||
| 403 : | |||
| 404 : | overbeek | 1.24 | default will set a default to override (maybe) korgs |
| 405 : | |||
| 406 : | redwards | 1.5 | =cut |
| 407 : | |||
| 408 : | sub scrolling_org_list { | ||
| 409 : | overbeek | 1.24 | my ($self, $cgi, $multiple, $default)=@_; |
| 410 : | redwards | 1.5 | unless ($multiple) {$multiple=0} |
| 411 : | |||
| 412 : | my @display = ( 'All', 'Archaea', 'Bacteria', 'Eucarya', 'Viruses', 'Environmental samples' ); | ||
| 413 : | |||
| 414 : | # | ||
| 415 : | # Canonical names must match the keywords used in the DBMS. They are | ||
| 416 : | # defined in compute_genome_counts.pl | ||
| 417 : | # | ||
| 418 : | my %canonical = ( | ||
| 419 : | 'All' => undef, | ||
| 420 : | 'Archaea' => 'Archaea', | ||
| 421 : | 'Bacteria' => 'Bacteria', | ||
| 422 : | 'Eucarya' => 'Eukaryota', | ||
| 423 : | 'Viruses' => 'Virus', | ||
| 424 : | 'Environmental samples' => 'Environmental Sample' | ||
| 425 : | ); | ||
| 426 : | |||
| 427 : | my $req_dom = $cgi->param( 'domain' ) || 'All'; | ||
| 428 : | my @domains = $cgi->radio_group( -name => 'domain', | ||
| 429 : | -default => $req_dom, | ||
| 430 : | -override => 1, | ||
| 431 : | -values => [ @display ] | ||
| 432 : | ); | ||
| 433 : | |||
| 434 : | my $n_domain = 0; | ||
| 435 : | my %dom_num = map { ( $_, $n_domain++ ) } @display; | ||
| 436 : | my $req_dom_num = $dom_num{ $req_dom } || 0; | ||
| 437 : | |||
| 438 : | # | ||
| 439 : | # Viruses and Environmental samples must have completeness = All (that is | ||
| 440 : | # how they are in the database). Otherwise, default is Only "complete". | ||
| 441 : | # | ||
| 442 : | my $req_comp = ( $req_dom_num > $dom_num{ 'Eucarya' } ) ? 'All' | ||
| 443 : | : $cgi->param( 'complete' ) || 'Only "complete"'; | ||
| 444 : | my @complete = $cgi->radio_group( -name => 'complete', | ||
| 445 : | -default => $req_comp, | ||
| 446 : | -override => 1, | ||
| 447 : | -values => [ 'All', 'Only "complete"' ] | ||
| 448 : | ); | ||
| 449 : | # | ||
| 450 : | # Use $fig->genomes( complete, restricted, domain ) to get org list: | ||
| 451 : | # | ||
| 452 : | my $complete = ( $req_comp =~ /^all$/i ) ? undef : "complete"; | ||
| 453 : | |||
| 454 : | my $orgs; my $label; | ||
| 455 : | @$orgs = $fig->genomes( $complete, undef, $canonical{ $req_dom } ); | ||
| 456 : | |||
| 457 : | foreach (@$orgs) { | ||
| 458 : | my $gs = $fig->genus_species($_); | ||
| 459 : | overbeek | 1.29 | my $gc=$fig->number_of_contigs($_); |
| 460 : | redwards | 1.5 | $label->{$_} = "$gs ($_) [$gc contigs]"; |
| 461 : | } | ||
| 462 : | |||
| 463 : | @$orgs = sort {$label->{$a} cmp $label->{$b}} @$orgs; | ||
| 464 : | |||
| 465 : | my $n_genomes = @$orgs; | ||
| 466 : | |||
| 467 : | return ( "<TABLE>\n", | ||
| 468 : | " <TR>\n", | ||
| 469 : | " <TD>", | ||
| 470 : | redwards | 1.6 | $cgi->scrolling_list( -name => 'korgs', |
| 471 : | -values => $orgs, | ||
| 472 : | -labels => $label, | ||
| 473 : | -size => 10, | ||
| 474 : | -multiple => $multiple, | ||
| 475 : | overbeek | 1.24 | -default => $default, |
| 476 : | redwards | 1.5 | ), $cgi->br, |
| 477 : | "$n_genomes genomes shown ", | ||
| 478 : | $cgi->submit( 'Update List' ), $cgi->reset, $cgi->br, | ||
| 479 : | " </TD>", | ||
| 480 : | " <TD>", | ||
| 481 : | join( "<br>", "<b>Domain(s) to show:</b>", @domains), "<br>\n", | ||
| 482 : | join( "<br>", "<b>Completeness?</b>", @complete), "\n", | ||
| 483 : | "</TD>", | ||
| 484 : | " </TR>\n", | ||
| 485 : | "</TABLE>\n", | ||
| 486 : | ); | ||
| 487 : | } | ||
| 488 : | |||
| 489 : | redwards | 1.21 | |
| 490 : | =head2 scrolling_subsys_list | ||
| 491 : | |||
| 492 : | Create a scrolling list of all subsystems. Just like scrolling_org_list, this will make the list and allow you to select multiples. | ||
| 493 : | |||
| 494 : | use like this | ||
| 495 : | |||
| 496 : | push @$html, $raelib->scrolling_subsys_list($cgi, $multiple); | ||
| 497 : | |||
| 498 : | =cut | ||
| 499 : | |||
| 500 : | sub scrolling_subsys_list { | ||
| 501 : | my ($self, $cgi, $multiple)=@_; | ||
| 502 : | $multiple=0 unless (defined $multiple); | ||
| 503 : | redwards | 1.22 | my @ss=sort {uc($a) cmp uc($b)} $fig->all_subsystems(); |
| 504 : | redwards | 1.21 | my $label; |
| 505 : | # generate labels for the list | ||
| 506 : | foreach my $s (@ss) {my $k=$s; $k =~ s/\_/ /g; $k =~ s/ / /g; $k =~ s/\s+$//; $label->{$s}=$k} | ||
| 507 : | return $cgi->scrolling_list( | ||
| 508 : | -name => 'subsystems', | ||
| 509 : | -values => \@ss, | ||
| 510 : | -labels => $label, | ||
| 511 : | -size => 10, | ||
| 512 : | -multiple=> $multiple, | ||
| 513 : | ); | ||
| 514 : | } | ||
| 515 : | |||
| 516 : | =head2 subsys_names_for_display | ||
| 517 : | |||
| 518 : | Return a list of subsystem names for display. This will take a list as an argument and return a nice clean list for display. | ||
| 519 : | |||
| 520 : | $raelib->subsys_names_for_display(@ss); | ||
| 521 : | or | ||
| 522 : | $raelib->subsys_names_for_display($fig->all_subsystems()); | ||
| 523 : | |||
| 524 : | =cut | ||
| 525 : | |||
| 526 : | sub subsys_names_for_display { | ||
| 527 : | my ($self, @ss)=@_; | ||
| 528 : | foreach (@ss) {s/\_/ /g; 1 while (s/ / /g); s/\s+$//} | ||
| 529 : | return @ss; | ||
| 530 : | } | ||
| 531 : | |||
| 532 : | redwards | 1.17 | =head2 GenBank |
| 533 : | redwards | 1.5 | |
| 534 : | redwards | 1.17 | This object will take a genome number and return a Bio::Seq::RichSeq object that has the whole genome |
| 535 : | in GenBank format. This should be a nice way of getting some data out, but will probably be quite slow | ||
| 536 : | at building the object. | ||
| 537 : | redwards | 1.1 | |
| 538 : | redwards | 1.17 | Note that you need to call this with the genome name and the contig. This will then go through that contig. |
| 539 : | redwards | 1.1 | |
| 540 : | redwards | 1.17 | Something like this should work |
| 541 : | |||
| 542 : | foreach my $contig ($fig->all_contigs($genome)) { | ||
| 543 : | my $seqobj=FIGRob->GenBank($genome, $contig); | ||
| 544 : | # process the contig | ||
| 545 : | } | ||
| 546 : | |||
| 547 : | =cut | ||
| 548 : | |||
| 549 : | sub GenBank { | ||
| 550 : | my ($self, $genome, $contig)=@_; | ||
| 551 : | my $gs=$fig->genus_species($genome); | ||
| 552 : | return unless ($gs); | ||
| 553 : | unless ($contig) { | ||
| 554 : | print STDERR "You didn't provide a contig for $gs. I think that was a mistake. Sorry\n"; | ||
| 555 : | return; | ||
| 556 : | } | ||
| 557 : | my $len=$fig->contig_ln($genome, $contig); | ||
| 558 : | unless ($len) { | ||
| 559 : | print STDERR "$contig from $gs doesn't appear to have a length. Is it right?\n"; | ||
| 560 : | return; | ||
| 561 : | } | ||
| 562 : | |||
| 563 : | |||
| 564 : | # first find all the pegs ... | ||
| 565 : | my $features; # all the features in the genome | ||
| 566 : | my $allpegs; # all the pegs | ||
| 567 : | my $translation; # all the protein sequences | ||
| 568 : | foreach my $peg ($fig->pegs_of($genome)) { | ||
| 569 : | my @location=$fig->feature_location($peg); | ||
| 570 : | my $func=$fig->function_of($peg); | ||
| 571 : | foreach my $loc (@location) { | ||
| 572 : | $loc =~ /^(.*)\_(\d+)\_(\d+)$/; | ||
| 573 : | my ($cg, $start, $stop)=($1, $2, $3); | ||
| 574 : | next unless ($cg eq $contig); | ||
| 575 : | # save this information for later | ||
| 576 : | $features->{'peg'}->{$loc}=$func; | ||
| 577 : | $allpegs->{'peg'}->{$loc}=$peg; | ||
| 578 : | $translation->{'peg'}->{$loc}=$fig->get_translation($peg); | ||
| 579 : | } | ||
| 580 : | } | ||
| 581 : | # ... and all the RNAs | ||
| 582 : | foreach my $peg ($fig->rnas_of($genome)) { | ||
| 583 : | my @location=$fig->feature_location($peg); | ||
| 584 : | my $func=$fig->function_of($peg); | ||
| 585 : | foreach my $loc (@location) { | ||
| 586 : | $loc =~ /^(.*)\_(\d+)\_(\d+)$/; | ||
| 587 : | my ($cg, $start, $stop)=($1, $2, $3); | ||
| 588 : | next unless ($cg eq $contig); | ||
| 589 : | # save this information for later | ||
| 590 : | $features->{'rna'}->{$loc}=$func; | ||
| 591 : | $allpegs->{'rna'}->{$loc}=$peg; | ||
| 592 : | } | ||
| 593 : | } | ||
| 594 : | |||
| 595 : | |||
| 596 : | # now get all the contigs out | ||
| 597 : | my $seq=$fig->dna_seq($genome, $contig."_1_".$len); | ||
| 598 : | my $description = "Contig $contig from " . $fig->genus_species($genome); | ||
| 599 : | my $sobj=Bio::Seq->new( | ||
| 600 : | -seq => $seq, | ||
| 601 : | -id => $contig, | ||
| 602 : | -desc => $description, | ||
| 603 : | -accession_number => $genome | ||
| 604 : | ); | ||
| 605 : | foreach my $prot (keys %{$features->{'peg'}}) { | ||
| 606 : | $prot =~ /^(.*)\_(\d+)\_(\d+)$/; | ||
| 607 : | my ($cg, $start, $stop)=($1, $2, $3); | ||
| 608 : | my $strand=1; | ||
| 609 : | if ($stop < $start) { | ||
| 610 : | ($stop, $start)=($start, $stop); | ||
| 611 : | $strand=-1; | ||
| 612 : | } | ||
| 613 : | |||
| 614 : | my $feat=Bio::SeqFeature::Generic->new( | ||
| 615 : | -start => $start, | ||
| 616 : | -end => $stop, | ||
| 617 : | -strand => $strand, | ||
| 618 : | -primary => 'CDS', | ||
| 619 : | -display_name => $allpegs->{'peg'}->{$prot}, | ||
| 620 : | -source_tag => 'the SEED', | ||
| 621 : | -tag => | ||
| 622 : | { | ||
| 623 : | db_xref => $allpegs->{'peg'}->{$prot}, | ||
| 624 : | note => 'Generated by the Fellowship for the Interpretation of Genomes', | ||
| 625 : | function => $features->{'peg'}->{$prot}, | ||
| 626 : | translation => $translation->{'peg'}->{$prot} | ||
| 627 : | } | ||
| 628 : | ); | ||
| 629 : | |||
| 630 : | $sobj->add_SeqFeature($feat); | ||
| 631 : | } | ||
| 632 : | |||
| 633 : | foreach my $prot (keys %{$features->{'rna'}}) { | ||
| 634 : | $prot =~ /^(.*)\_(\d+)\_(\d+)$/; | ||
| 635 : | my ($cg, $start, $stop)=($1, $2, $3); | ||
| 636 : | my $strand=1; | ||
| 637 : | if ($stop < $start) { | ||
| 638 : | ($stop, $start)=($start, $stop); | ||
| 639 : | $strand=-1; | ||
| 640 : | } | ||
| 641 : | |||
| 642 : | my $feat=Bio::SeqFeature::Generic->new( | ||
| 643 : | -start => $start, | ||
| 644 : | -end => $stop, | ||
| 645 : | -strand => $strand, | ||
| 646 : | -primary => 'RNA', | ||
| 647 : | -source_tag => 'the SEED', | ||
| 648 : | -display_name => $allpegs->{'rna'}->{$prot}, | ||
| 649 : | -tag => | ||
| 650 : | { | ||
| 651 : | db_xref => $allpegs->{'rna'}->{$prot}, | ||
| 652 : | note => 'Generated by the Fellowship for the Interpretation of Genomes', | ||
| 653 : | function => $features->{'rna'}->{$prot}, | ||
| 654 : | } | ||
| 655 : | ); | ||
| 656 : | |||
| 657 : | $sobj->add_SeqFeature($feat); | ||
| 658 : | } | ||
| 659 : | return $sobj; | ||
| 660 : | } | ||
| 661 : | |||
| 662 : | =head2 best_hit | ||
| 663 : | |||
| 664 : | Returns the FIG id of the single best hit to a peg | ||
| 665 : | |||
| 666 : | eg | ||
| 667 : | |||
| 668 : | my $bh=$fr->best_hit($peg); | ||
| 669 : | print 'function is ', scalar $fig->function_of($bh); | ||
| 670 : | |||
| 671 : | =cut | ||
| 672 : | |||
| 673 : | sub best_hit { | ||
| 674 : | my ($self, $peg)=@_; | ||
| 675 : | return unless ($peg); | ||
| 676 : | |||
| 677 : | my ($maxN, $maxP)=(1, 1e-5); | ||
| 678 : | my @sims=$fig->sims($peg, $maxN, $maxP, 'fig'); | ||
| 679 : | return ${$sims[0]}[1]; | ||
| 680 : | } | ||
| 681 : | redwards | 1.1 | |
| 682 : | redwards | 1.23 | |
| 683 : | =head1 read_fasta | ||
| 684 : | |||
| 685 : | Read a fasta format file and return a reference to a hash with the data. The key is the ID and the value is the sequence. If you supply the optional keep comments then the comments (anything after the first white space are returned as a sepaarte hash). | ||
| 686 : | |||
| 687 : | Usage: | ||
| 688 : | my $fasta=$raelib->read_fasta($file); | ||
| 689 : | my ($fasta, $comments)=$raelib->read_fasta($file, 1); | ||
| 690 : | |||
| 691 : | =cut | ||
| 692 : | |||
| 693 : | sub read_fasta { | ||
| 694 : | my ($self, $file, $keepcomments)=@_; | ||
| 695 : | open (IN, $file) || die "Can't open $file"; | ||
| 696 : | my %f; my $t; my $s; my %c; | ||
| 697 : | overbeek | 1.26 | while (<IN>) { |
| 698 : | redwards | 1.23 | chomp; |
| 699 : | if (/^>/) { | ||
| 700 : | if ($s) { | ||
| 701 : | $f{$t}=$s; | ||
| 702 : | undef $s; | ||
| 703 : | } | ||
| 704 : | s/^>(\S+)\s*//; | ||
| 705 : | $t=$1; | ||
| 706 : | $c{$t}=$_ if ($_); | ||
| 707 : | } | ||
| 708 : | else {$s .= $_} | ||
| 709 : | } | ||
| 710 : | $f{$t}=$s; | ||
| 711 : | if ($keepcomments) {return (\%f, \%c)} | ||
| 712 : | else {return \%f} | ||
| 713 : | } | ||
| 714 : | |||
| 715 : | =head1 rc | ||
| 716 : | |||
| 717 : | Reverse complement. It's too easy. | ||
| 718 : | |||
| 719 : | =cut | ||
| 720 : | |||
| 721 : | sub rc { | ||
| 722 : | my ($self, $seq)=@_; | ||
| 723 : | $seq=~tr/GATCgatc/CTAGctag/; | ||
| 724 : | $seq = reverse $seq; | ||
| 725 : | return $seq; | ||
| 726 : | } | ||
| 727 : | |||
| 728 : | redwards | 1.27 | |
| 729 : | =head2 cookies | ||
| 730 : | |||
| 731 : | Handle cookies. This method will get and set the value of the FIG cookie. Cookies are name/value pairs that are stored on the users computer. We then retrieve them using this method. The cookies are passed in as a reference to a hash, and the method returns a tuple of the cookie that can be passed to the browser and a reference to a hash with the data. | ||
| 732 : | |||
| 733 : | If you do not pass any arguments the whole cookie will be returned. | ||
| 734 : | |||
| 735 : | Use as follows: | ||
| 736 : | |||
| 737 : | ($cookie, $data) = raelib->cookie($cgi, \%data); | ||
| 738 : | |||
| 739 : | You do not need to pass in any data, in that case you will just get the cookie back | ||
| 740 : | |||
| 741 : | Underneath, I create a single cookie called FIG which stores all the information. The names and value pairs are stored using = to join name to value and ; to concatenate. This way we can create a single cookie with all the data. I am using the FIG::clean_attribute_key method to remove unwanted characters from the name/value pairs, so don't use them. | ||
| 742 : | |||
| 743 : | Note that for the moment I have put this routine here since it needs to maintain the state of the cookie (i.e. it needs to know what $self is). It should really be in HTML.pm but that is not, as far as I can tell, maintaining states? | ||
| 744 : | |||
| 745 : | =cut | ||
| 746 : | |||
| 747 : | sub cookie { | ||
| 748 : | my ($self, $cgi, $input)=@_; | ||
| 749 : | return unless ($cgi); | ||
| 750 : | $self->{'cookie'}=$cgi->cookie(-name=>"FIG") unless ($self->{'cookie'}); | ||
| 751 : | |||
| 752 : | # first, create a hash from the existing cookie data | ||
| 753 : | my $cookie; | ||
| 754 : | map { | ||
| 755 : | my ($kname, $kvalue)=split /\=/, $_; | ||
| 756 : | $cookie->{$kname}=$kvalue; | ||
| 757 : | } split /\;/, $self->{'cookie'}; | ||
| 758 : | |||
| 759 : | if ($input) | ||
| 760 : | { | ||
| 761 : | # add the values that were passed in | ||
| 762 : | map {$cookie->{FIG->clean_attribute_key($_)}=$input->{$_}} keys %$input; | ||
| 763 : | # put everything back together and set the cookie | ||
| 764 : | my $newcookie=join ";", map {$_ . "=" . $cookie->{$_}} keys %$cookie; | ||
| 765 : | $self->{'cookie'}=$cgi->cookie(-name=>"FIG", -value=>$newcookie, -expires=>'+1y'); | ||
| 766 : | } | ||
| 767 : | |||
| 768 : | return ($self->{'cookie'}, $cookie); | ||
| 769 : | } | ||
| 770 : | |||
| 771 : | |||
| 772 : | |||
| 773 : | overbeek | 1.30 | =head1 commify |
| 774 : | |||
| 775 : | Put commas in numbers. I think this comes straight from the perl cookbook and is very useful for nice displays | ||
| 776 : | |||
| 777 : | =cut | ||
| 778 : | |||
| 779 : | sub commify { | ||
| 780 : | my($self,$n) = @_; | ||
| 781 : | my(@n) = (); | ||
| 782 : | my($i); | ||
| 783 : | |||
| 784 : | for ($i = (length($n) - 3); ($i > 0); $i -= 3) | ||
| 785 : | { | ||
| 786 : | unshift(@n,",",substr($n,$i,3)); | ||
| 787 : | } | ||
| 788 : | unshift(@n,substr($n,0,$i+3)); | ||
| 789 : | return join("",@n); | ||
| 790 : | } | ||
| 791 : | |||
| 792 : | |||
| 793 : | |||
| 794 : | |||
| 795 : | redwards | 1.27 | |
| 796 : | redwards | 1.1 | 1; |
| 797 : | redwards | 1.17 |
| MCS Webmaster | ViewVC Help |
| Powered by ViewVC 1.0.3 |