3 # parse_genecard_results retreives files of search results from ncbi,
4 # and is released under the terms of the GPL version 2, or any later
5 # version, at your option. See the file README and COPYING for more
8 # Copyright 2004 by Don Armstrong <don@donarmstrong.com>.
10 # $Id: ss,v 1.1 2004/06/29 05:26:35 don Exp $
22 parse_genecard_results [options]
28 --dir, -D directory to stick results into [default .]
29 --name, -n file naming scheme [default ${search}_results.$format]
30 --terms, -t file of search terms [default -]
31 --debug, -d debugging level [default 0]
32 --help, -h display this help
33 --man, -m display manual
41 Debug verbosity. (Default 0)
45 Display brief useage information.
55 parse_harvester_results -D ./harvester_results/ -n '${search}_name.html' < search_parameters
57 Will pretty much do what you want
63 use vars qw($DEBUG $REVISION);
66 ($REVISION) = q$LastChangedRevision: 1$ =~ /LastChangedRevision:\s+([^\s]+)/;
67 $DEBUG = 0 unless defined $DEBUG;
73 my %options = (debug => 0,
81 GetOptions(\%options,'keyword|k=s','dir|D=s','debug|d+','help|h|?','man|m',
86 pod2usage() if $options{help};
87 pod2usage({verbose=>2}) if $options{man};
89 $DEBUG = $options{debug};
92 use constant {NAME => 0,
103 if ($options{keywords}) {
105 pod2usage("If the --keywords option is used, exactly one argument (the keyword) must be passed");
107 $option{dir} = "$ARGV[0]_results_genecard";
110 if (not -d $options{dir}) {
111 die "$options{dir} does not exist or is not a directory";
114 my $dir = new IO::Dir $options{dir} or die "Unable to open dir $options{dir}: $!";
116 print join(",", map {qq("$_");} qw(Name RefSeq Location Alias Function Description Keyword DBName Filename)),qq(\n);
118 while ($_ = $dir->read) {
120 next if $file_name =~ /^\./;
121 next unless -f "$options{dir}/$file_name" and -r "$options{dir}/$file_name";
123 my $file = new IO::File "$options{dir}/$file_name", 'r' or die "Unable to open file $file_name";
126 my $result = <$file>;
131 ($results[NAME]) = $result =~ m&(?:Lean|Gene)Card\s+for\s+(?:(?:disorder\s+locus|uncategorized|
132 hugo\s*reserved\s*symbol|cluster|
133 potentially\s*expressed\s*sequence)|(?:predicted\s+|pseudo|rna\s+|)gene)
134 \s*(?:with\s*support\s*|)<FONT\s+COLOR=\"[^\"]+\">\s*<FONT\s+SIZE=\+2>\s*([^\s]+)\s*&xis;
136 $results[NAME] ||= 'NO NAME';
137 # Find REF SEQ number
138 ($results[REFSEQ]) = $result =~ m|http://www.ncbi.nlm.nih.gov/entrez/query.fcgi\?
139 cmd=Search\&db=nucleotide\&doptcmdl=GenBank\&term=([^\"]+)\"|xis;
141 $results[REFSEQ] ||= 'NO REFSEQ';
144 ($results[LOCATION]) = $result =~ m&<I>LocusLink\s+cytogenetic\s+band:</I><b>\s+
145 <a\s+href="[^\"]+"\s+target\s+=\s+"aaa">\s*([^\<]+?)\s*</a>&xis;
147 $results[LOCATION] ||= 'NO LOCATION';
150 my ($alias_table) = $result =~ m|<b>Aliases and Descriptions</b>(.+?)</TR>|is;
153 my @gene_aliases = $alias_table =~ m|<li>\s*([^\(]{0,20}?)\s*\(<FONT|gis;
155 $results[ALIAS] = join('; ', @gene_aliases);
156 $results[ALIAS] ||= 'NO ALIASES';
158 # Find gene function(s)
160 # Swiss prot functions
161 my @functions = $result =~ m&<li><b>Function:</b>\s+(.+?)(?:<li>)|(?:</ul>)&gis;
164 push @functions, (map {s#\s*</a>\s*# #g; $_;} $result =~ m&(GO:\d+\s*</a>.+?)(?:<dd>|<p>)&gis);
165 $results[FUNCTION] = join('; ', map {(defined $_)?($_):()} @functions);
166 $results[FUNCTION] ||= 'NO FUNCTION';
168 # Figure out the keyword used
169 ($results[KEYWORD]) = $file_name =~ /search=([^&]+)/;
171 $results[KEYWORD] ||= 'NO KEYWORD';
173 # Figure out what the description is
174 $results[DESCRIPTION] = '';
177 $results[DBNAME] = 'genecard';
178 $results[FILENAME] = $file_name;
180 print join(',',map {qq("$_")} @results),qq(\n);