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,
74 GetOptions(\%options,'keyword|k=s','debug|d+','help|h|?','man|m');
77 pod2usage() if $options{help};
78 pod2usage({verbose=>2}) if $options{man};
80 $DEBUG = $options{debug};
83 use constant {NAME => 0,
94 my @csv_fields = qw(name hits rzscore refseq location alias database terms description function);
98 for my $file_name (@ARGV) {
99 my $file = new IO::File $file_name, 'r' or die "Unable to open file $file_name $!";
102 my @gene = map {s/^\"//; s/\"$//; $_;} split /(?<=\")\,(?=\")/, $_;
103 $genes{$gene[NAME]}{name} = $gene[NAME];
104 $genes{$gene[NAME]}{database}{$gene[DBNAME]}++;
105 $genes{$gene[NAME]}{hits}++;
106 $genes{$gene[NAME]}{terms}{$gene[KEYWORD]}++;
107 $genes{$gene[NAME]}{terms}{$gene[KEYWORD].'['.$gene[DBNAME].']'}++;
108 add_unique_parts($genes{$gene[NAME]},'refseq',$gene[REFSEQ]);
109 add_if_better($genes{$gene[NAME]},'description',$gene[DESCRIPTION]);
110 add_if_better($genes{$gene[NAME]},'location',$gene[LOCATION]);
111 add_unique_parts($genes{$gene[NAME]},'function',split(/; /, $gene[FUNCTION]));
112 add_unique_parts($genes{$gene[NAME]},'alias', split(/; /, $gene[ALIAS]));
116 print join(',',map {qq("$_")} @csv_fields),qq(\n);
117 for my $gene (keys %genes) {
118 $genes{$gene}{rzscore} = scalar grep {$_ !~ /\[/} keys %{$genes{$gene}{terms}};
119 next if $genes{$gene}{rzscore} == 1 and exists $genes{$gene}{terms}{antigen};
120 $genes{$gene}{rzscore} -= 1 if exists $genes{$gene}{terms}{antigen};
121 print STDOUT join (',',
122 map {s/"//g; qq("$_")}
125 if (ref $value eq 'HASH') {
126 join('; ',map {qq($_:$$value{$_})} keys %$value);
128 elsif (ref $value eq 'ARRAY') {
134 } @{$genes{$gene}}{@csv_fields}
139 sub add_unique_parts{
140 my ($hr,$key,@values) = @_;
141 if (not defined $$hr{key}) {
142 $$hr{$key} = [@values];
146 @temp_hash{@{$$hr{$key}}} = (1) x scalar @{$$hr{$key}};
147 $temp_hash{@values} = (1) x scalar @values;
148 $$hr{$key} = [keys %temp_hash];
153 my ($hash, $key, $value) = @_;
154 if (not defined $$hash{$key}) {
155 $$hash{$key} = $value;
157 elsif (length $$hash{$key} < length $value and $value !~ /^NO\s+(LOCATION|DESCRIPTION)+$/) {
158 $$hash{$key} = $value;