3 # combine_results, is part of the gene search suite, and is released
4 # under the terms of the GPL version 2, or any later version, at your
5 # option. See the file README and COPYING for more information.
7 # Copyright 2006,2007 by Don Armstrong <don@donarmstrong.com>.
19 combine_results -- combines parsed result files; outputs to stdout.
23 combine_results parsed_results_1.txt [parsedresultfiles ...]
26 --debug, -d debugging level [default 0]
27 --help, -h display this help
28 --man, -m display manual
36 Debug verbosity. (Default 0)
40 Display brief useage information.
50 combine_results foo_1.txt
52 Will pretty much do what you want
61 $DEBUG = 0 unless defined $DEBUG;
64 use XML::Parser::Expat;
67 # XXX parse config file
69 my %options = (debug => 0,
75 GetOptions(\%options,'keywords|k=s@','debug|d+','help|h|?','man|m');
78 pod2usage() if $options{help};
79 pod2usage({verbose=>2}) if $options{man};
81 $DEBUG = $options{debug};
84 use constant {NAME => 0,
95 my @csv_fields = qw(name hits rzscore refseq location alias database terms description function);
99 for my $file_name (@ARGV) {
100 my $file = new IO::File $file_name, 'r' or die "Unable to open file $file_name $!";
103 my @gene = map {s/^\"//; s/\"$//; $_;} split /(?<=\")\,(?=\")/, $_;
104 $genes{$gene[NAME]}{name} = $gene[NAME];
105 $genes{$gene[NAME]}{database}{$gene[DBNAME]}++;
106 $genes{$gene[NAME]}{hits}++;
107 $genes{$gene[NAME]}{terms}{$gene[KEYWORD]}++;
108 $genes{$gene[NAME]}{terms}{$gene[KEYWORD].'['.$gene[DBNAME].']'}++;
109 add_unique_parts($genes{$gene[NAME]},'refseq',$gene[REFSEQ]);
110 add_if_better($genes{$gene[NAME]},'description',$gene[DESCRIPTION]);
111 add_if_better($genes{$gene[NAME]},'location',$gene[LOCATION]);
112 add_unique_parts($genes{$gene[NAME]},'function',split(/; /, $gene[FUNCTION]));
113 add_unique_parts($genes{$gene[NAME]},'alias', split(/; /, $gene[ALIAS]));
117 print join(',',map {qq("$_")} @csv_fields),qq(\n);
118 for my $gene (keys %genes) {
119 $genes{$gene}{rzscore} = scalar grep {$_ !~ /\[/} keys %{$genes{$gene}{terms}};
120 print STDOUT join (',',
121 map {s/"//g; qq("$_")}
124 if (ref $value eq 'HASH') {
125 join('; ',map {qq($_:$$value{$_})} keys %$value);
127 elsif (ref $value eq 'ARRAY') {
133 } @{$genes{$gene}}{@csv_fields}
138 sub add_unique_parts{
139 my ($hr,$key,@values) = @_;
140 if (not defined $$hr{key}) {
141 $$hr{$key} = [@values];
145 @temp_hash{@{$$hr{$key}}} = (1) x scalar @{$$hr{$key}};
146 $temp_hash{@values} = (1) x scalar @values;
147 $$hr{$key} = [keys %temp_hash];
152 my ($hash, $key, $value) = @_;
153 if (not defined $$hash{$key}) {
154 $$hash{$key} = $value;
156 elsif (length $$hash{$key} < length $value and $value !~ /^NO\s+(LOCATION|DESCRIPTION)+$/) {
157 $$hash{$key} = $value;