[Bio] / FigKernelPackages / raelib.pm Repository:
ViewVC logotype

Annotation of /FigKernelPackages/raelib.pm

Parent Directory Parent Directory | Revision Log Revision Log


Revision 1.9 - (view) (download) (as text)

1 : redwards 1.1 # -*- perl -*-
2 :    
3 :     =pod
4 :    
5 :     =head1
6 :    
7 :     Some routines and things that Rob uses. Please feel free to use at will and incorporate into
8 :     your own code or move them into FIG.pm or elsewhere.
9 :    
10 :     =cut
11 :    
12 :     package raelib;
13 :     use strict;
14 :     use FIG;
15 :     my $fig=new FIG;
16 :    
17 : redwards 1.5 =head2 new
18 :    
19 :     Just instantiate the object and return $self
20 :    
21 :     =cut
22 :    
23 :     sub new {
24 :     my $self=shift;
25 :     return $self
26 :     }
27 :    
28 :    
29 :    
30 :    
31 : redwards 1.4 =head2 features_on_contig
32 :    
33 :     Returns a reference to an array containing all the features on a contig in a genome.
34 :    
35 :     use:
36 :    
37 :     my $arrayref=$rae->features_on_contig($genome, $contig);
38 :    
39 :     or
40 :    
41 :     foreach my $peg (@{$rae->features_on_contig($genome, $contig)}) {
42 :     ... blah blah ...
43 :     }
44 :    
45 :     returns undef if contig is not a part of genome or there is nothing to return, otherwise returns a list of pegs
46 :    
47 :     v. experimental and guaranteed not to work!
48 :    
49 :     =cut
50 :    
51 :     sub features_on_contig {
52 :     my ($self, $genome, $contig)=@_;
53 :     # were this in FIG.pm you'd use this line:
54 :     #my $rdbH = $self->db_handle;
55 :    
56 :     my $rdbH = $fig->db_handle;
57 :     my $relational_db_response=$rdbH->SQL('SELECT id FROM features WHERE (genome = \'' . $genome . '\' AND location ~* \'' . $contig . '\')');
58 :     # this is complicated. A reference to an array of references to arrays, and we only want the first element.
59 :     # simplify.
60 :     my @results;
61 :     foreach my $res (@$relational_db_response) {push @results, $res->[0]}
62 :     return \@results;
63 :     }
64 :    
65 :    
66 :    
67 :    
68 :    
69 :    
70 :    
71 : redwards 1.7 =head2 pirsfcorrespondence
72 : redwards 1.1
73 : redwards 1.7 Generate the pirsf->fig id correspondence. This is only done once and the correspondence file is written.
74 : redwards 1.1 This is so that we can easily go back and forth.
75 :    
76 : redwards 1.7 The correspondence has PIR ID \t FIG ID\n, and is probably based on ftp://ftp.pir.georgetown.edu/pir_databases/pirsf/data/pirsfinfo.dat
77 : redwards 1.1
78 : redwards 1.9 This method takes three arguments:
79 :     from : pirsfinfo.dat file
80 :     to : file to write information to
81 :     verbose : report on progress
82 :    
83 :     Returns the number of lines in the pirsinfo file that were read.
84 :    
85 : redwards 1.1 =cut
86 :    
87 : redwards 1.7 sub pirsfcorrespondence {
88 : redwards 1.9 my ($self, $from, $to, $verbose)=@_;
89 : redwards 1.1 die "File $from does not exist as called in $0" unless (-e $from);
90 :     open (IN, $from) || die "Can't open $from";
91 : redwards 1.8 open (OUT, ">$to") || die "Can't write to $to";
92 : redwards 1.9 my $linecount;
93 : redwards 1.1 while (<IN>) {
94 : redwards 1.9 $linecount++;
95 :     unless ($linecount % 10000) {print STDERR "Correspondence of $linecount lines calculated\n"}
96 : redwards 1.1 if (/^>/) {print OUT; next}
97 :     chomp;
98 :     my $done;
99 : redwards 1.2 foreach my $peg ($fig->by_alias("uni|$_")) {
100 : redwards 1.1 print OUT $_, "\t", $peg, "\n";
101 :     $done=1;
102 :     }
103 :     unless ($done) {print OUT $_, "\t\n"}
104 :     }
105 :     close IN;
106 :     close OUT;
107 : redwards 1.9 return $linecount;
108 : redwards 1.1 }
109 :    
110 :    
111 :     =head2 ss_by_id
112 :    
113 :     Generate a list of subsystems that a peg occurs in. This is a ; separated list.
114 :     This is a wrapper that removes roles and ignores essential things
115 :    
116 :     =cut
117 :    
118 :     sub ss_by_id {
119 :     my ($self, $peg)=@_;
120 :     my $ssout;
121 :     foreach my $ss (sort $fig->subsystems_for_peg($peg))
122 :     {
123 :     next if ($$ss[0] =~ /essential/i); # Ignore the Essential B-subtilis subsystems
124 :     $ssout.=$$ss[0]."; ";
125 :     }
126 :     $ssout =~ s/; $//;
127 :     return $ssout;
128 :     }
129 :    
130 : redwards 1.3 =head2 ss_by_homol
131 :    
132 :     Generate a list of subsystems that homologs of a peg occur in. This is a ; separated list.
133 :     This is also a wrapper around sims and ss, but makes everything unified
134 :    
135 :     =cut
136 :    
137 :     sub ss_by_homol {
138 :     my ($self, $peg)=@_;
139 :     return unless ($peg);
140 :     my ($maxN, $maxP)=(50, 1e-20);
141 :    
142 :     # find the sims
143 :     my @sims=$fig->sims($peg, $maxN, $maxP, 'fig');
144 :    
145 :     # we are only going to keep the best hit for each peg
146 :     # in a subsystem
147 :     my $best_ss_score; my $best_ss_id;
148 :     foreach my $sim (@sims)
149 :     {
150 :     my $simpeg=$$sim[1];
151 :     my $simscore=$$sim[10];
152 :     my @subsys=$fig->subsystems_for_peg($simpeg);
153 :     foreach my $ss (@subsys)
154 :     {
155 :     if (! defined $best_ss_score->{$$ss[0]}) {$best_ss_score->{$$ss[0]}=$simscore; $best_ss_id->{$$ss[0]}=$simpeg}
156 :     elsif ($best_ss_score->{$$ss[0]} > $simscore)
157 :     {
158 :     $best_ss_score->{$$ss[0]}=$simscore;
159 :     $best_ss_id->{$$ss[0]}=$simpeg;
160 :     }
161 :     }
162 :     }
163 :    
164 :     my $ssoutput=join "", (map {"$_ (".$best_ss_id->{$_}."), "} keys %$best_ss_id);
165 :    
166 :     $ssoutput =~ s/, $//;
167 :     return $ssoutput;
168 :     }
169 :    
170 :     =head2 tagvalue
171 :    
172 :     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)
173 :    
174 :     e.g. $values=raelib->tagvalue($peg, "PIRSF"); print join "\n", @$values;
175 :    
176 :     Returns an empty array if no tag/value appropriate.
177 :    
178 :     Just because I use this a lot I don't want to waste rewriting it.
179 :    
180 :     =cut
181 :    
182 :     sub tagvalue {
183 :     my ($self, $peg, $tag)=@_;
184 :     my @return;
185 :     my @attr=$fig->feature_attributes($peg);
186 :     foreach my $attr (@attr) {
187 :     my ($gottag, $val, $link)=@$attr;
188 :     push @return, $val if ($gottag eq $tag);
189 :     }
190 :     return wantarray ? @return : join "; ", @return;
191 :     }
192 : redwards 1.1
193 : redwards 1.5 =head2 locations_on_contig
194 :    
195 :     Return the locations of a sequence on a contig.
196 :    
197 :     This will look for exact matches to a sequence on a contig, and return a reference to an array that has all the locations.
198 :    
199 :     my $locations=$raelib->locations_on_contig($genome, $contig, 'GATC', undef);
200 :     foreach my $bp (@$locations) { ... do something ... }
201 :    
202 :     first argument : genome number
203 :     second argument : contig name
204 :     third argument : sequence to look for
205 :     fourth argument : beginning position to start looking from (can be undef)
206 :     fifth argument : end position to stop looking from (can be undef)
207 :     sixth argument : check reverse complement (0 or undef will check forward, 1 or true will check rc)
208 :    
209 :     Note, the position is calculated before the sequence is rc'd
210 :    
211 :     =cut
212 :    
213 :     sub locations_on_contig {
214 :     my ($self, $genome, $contig, $sequence, $from, $to, $check_reverse)=@_;
215 :     my $return=[];
216 :    
217 :     # get the dna sequence of the contig, and make sure it is uppercase
218 :     my $contig_ln=$fig->contig_ln($genome, $contig);
219 :     return $return unless ($contig_ln);
220 :     unless ($from) {$from=1}
221 :     unless ($to) {$to=$contig_ln}
222 :     if ($from > $to) {($from, $to)=($to, $from)}
223 :     my $dna_seq=$fig->dna_seq($genome, $contig."_".$from."_".$to);
224 :     $dna_seq=uc($dna_seq);
225 :    
226 :     # if we want to check the rc, we actually rc the query
227 :     $sequence=$fig->reverse_comp($sequence) if ($check_reverse);
228 :     $sequence=uc($sequence);
229 :    
230 :     # now find all the matches
231 :     my $posn=index($dna_seq, $sequence, 0);
232 :     while ($posn > -1) {
233 :     push @$return, $posn;
234 :     $posn=index($dna_seq, $sequence, $posn+1);
235 :     }
236 :     return $return;
237 :     }
238 :    
239 :    
240 :     =head2 scrolling_org_list
241 :    
242 :     This is the list from index.cgi, that I call often. It has one minor modification: the value returned is solely the organisms id and does not contain genus_species information. I abstracted this here: 1, so I could call it often, and 2, so I could edit it once.
243 :    
244 :     use like this push @$html, $raelib->scrolling_org_list($cgi, $multiple);
245 :    
246 :     multiple selections will only be set if $multiple is true
247 :    
248 :     =cut
249 :    
250 :     sub scrolling_org_list {
251 :     my ($self, $cgi, $multiple)=@_;
252 :     unless ($multiple) {$multiple=0}
253 :    
254 :     my @display = ( 'All', 'Archaea', 'Bacteria', 'Eucarya', 'Viruses', 'Environmental samples' );
255 :    
256 :     #
257 :     # Canonical names must match the keywords used in the DBMS. They are
258 :     # defined in compute_genome_counts.pl
259 :     #
260 :     my %canonical = (
261 :     'All' => undef,
262 :     'Archaea' => 'Archaea',
263 :     'Bacteria' => 'Bacteria',
264 :     'Eucarya' => 'Eukaryota',
265 :     'Viruses' => 'Virus',
266 :     'Environmental samples' => 'Environmental Sample'
267 :     );
268 :    
269 :     my $req_dom = $cgi->param( 'domain' ) || 'All';
270 :     my @domains = $cgi->radio_group( -name => 'domain',
271 :     -default => $req_dom,
272 :     -override => 1,
273 :     -values => [ @display ]
274 :     );
275 :    
276 :     my $n_domain = 0;
277 :     my %dom_num = map { ( $_, $n_domain++ ) } @display;
278 :     my $req_dom_num = $dom_num{ $req_dom } || 0;
279 :    
280 :     #
281 :     # Viruses and Environmental samples must have completeness = All (that is
282 :     # how they are in the database). Otherwise, default is Only "complete".
283 :     #
284 :     my $req_comp = ( $req_dom_num > $dom_num{ 'Eucarya' } ) ? 'All'
285 :     : $cgi->param( 'complete' ) || 'Only "complete"';
286 :     my @complete = $cgi->radio_group( -name => 'complete',
287 :     -default => $req_comp,
288 :     -override => 1,
289 :     -values => [ 'All', 'Only "complete"' ]
290 :     );
291 :     #
292 :     # Use $fig->genomes( complete, restricted, domain ) to get org list:
293 :     #
294 :     my $complete = ( $req_comp =~ /^all$/i ) ? undef : "complete";
295 :    
296 :     my $orgs; my $label;
297 :     @$orgs = $fig->genomes( $complete, undef, $canonical{ $req_dom } );
298 :    
299 :     foreach (@$orgs) {
300 :     my $gs = $fig->genus_species($_);
301 :     my $gc=scalar $fig->all_contigs($_);
302 :     $label->{$_} = "$gs ($_) [$gc contigs]";
303 :     }
304 :    
305 :     @$orgs = sort {$label->{$a} cmp $label->{$b}} @$orgs;
306 :    
307 :     my $n_genomes = @$orgs;
308 :    
309 :     return ( "<TABLE>\n",
310 :     " <TR>\n",
311 :     " <TD>",
312 : redwards 1.6 $cgi->scrolling_list( -name => 'korgs',
313 :     -values => $orgs,
314 :     -labels => $label,
315 :     -size => 10,
316 :     -multiple => $multiple,
317 : redwards 1.5 ), $cgi->br,
318 :     "$n_genomes genomes shown ",
319 :     $cgi->submit( 'Update List' ), $cgi->reset, $cgi->br,
320 :     " </TD>",
321 :     " <TD>",
322 :     join( "<br>", "<b>Domain(s) to show:</b>", @domains), "<br>\n",
323 :     join( "<br>", "<b>Completeness?</b>", @complete), "\n",
324 :     "</TD>",
325 :     " </TR>\n",
326 :     "</TABLE>\n",
327 :     $cgi->hr
328 :     );
329 :     }
330 :    
331 :    
332 : redwards 1.1
333 :    
334 :    
335 :     1;

MCS Webmaster
ViewVC Help
Powered by ViewVC 1.0.3