#!/usr/local/bin/perl ############################################################################### # Program : GetProteinsByPep # Author : Eric Deutsch # $Id: GetPeptides 5436 2007-07-03 23:56:02Z sloeveni $ # # Description : This program that allows users to # get proteins from the PeptideAtlas based on various criteria. # # SBEAMS is Copyright (C) 2000-2005 Institute for Systems Biology # This program is governed by the terms of the GNU General Public License (GPL) # version 2 as published by the Free Software Foundation. It is provided # WITHOUT ANY WARRANTY. See the full description of GPL terms in the # LICENSE file distributed with this software. # ############################################################################### ############################################################################### # Set up all needed modules and objects ############################################################################### use strict; use Getopt::Long; use FindBin; use lib "$FindBin::Bin/../../lib/perl"; use vars qw ($sbeams $sbeamsMOD $q $current_contact_id $current_username $PROG_NAME $USAGE %OPTIONS $QUIET $VERBOSE $DEBUG $DATABASE $TABLE_NAME $PROGRAM_FILE_NAME $CATEGORY $DB_TABLE_NAME @MENU_OPTIONS); use SBEAMS::Connection qw($q $log); use SBEAMS::Connection::Settings; use SBEAMS::Connection::Tables; use SBEAMS::Connection::TabMenu; use SBEAMS::PeptideAtlas; use SBEAMS::PeptideAtlas::Settings; use SBEAMS::PeptideAtlas::Tables; $sbeams = new SBEAMS::Connection; $sbeamsMOD = new SBEAMS::PeptideAtlas; $sbeamsMOD->setSBEAMS($sbeams); $sbeams->setSBEAMS_SUBDIR($SBEAMS_SUBDIR); ############################################################################### # Set program name and usage banner for command line use ############################################################################### $PROG_NAME = $FindBin::Script; $USAGE = <Authenticate() and exit if it fails or continue if it works. ############################################################################### sub main { #### Do the SBEAMS authentication and exit if a username is not returned exit unless ($current_username = $sbeams->Authenticate( #permitted_work_groups_ref=>['PeptideAtlas_user','PeptideAtlas_admin'], # connect_read_only=>1, allow_anonymous_access=>1, )); #### Read in the default input parameters my %parameters; $parameters{uploaded_file_not_saved} = 1; my $n_params_found = $sbeams->parse_input_parameters( q=>$q,parameters_ref=>\%parameters); #$sbeams->printDebuggingInfo($q); #### Process generic "state" parameters before we start $sbeams->processStandardParameters(parameters_ref=>\%parameters); #### Decide what action to take based on information so far if ($parameters{action} eq "???") { # Some action } else { my $project_id = $sbeamsMOD->getProjectID( atlas_build_id => $parameters{atlas_build_id} ); $sbeamsMOD->display_page_header(project_id => $project_id); handle_request(ref_parameters=>\%parameters); $sbeamsMOD->display_page_footer(); } } # end main ############################################################################### # Handle Request ############################################################################### sub handle_request { my %args = @_; #### Process the arguments list my $ref_parameters = $args{'ref_parameters'} || die "ref_parameters not passed"; my %parameters = %{$ref_parameters}; #### Show current user context information print "
\n" if ($sbeams->output_mode() eq 'html'); #$sbeams->printUserContext(); #### Get the HTML to display the tabs my $tabMenu = $sbeamsMOD->getTabMenu( parameters_ref => \%parameters, program_name => $PROG_NAME, ); print $tabMenu->asHTML() if ($sbeams->output_mode() eq 'html'); #### Get the current atlas_build_id based on parameters or session my $atlas_build_id = $sbeamsMOD->getCurrentAtlasBuildID( parameters_ref => \%parameters, ); if (defined($atlas_build_id) && $atlas_build_id < 0) { return; } $parameters{atlas_build_id} = $atlas_build_id; #### Define some generic variables my ($i,$element,$key,$value,$line,$result,$sql); #### Define some variables for a query and resultset my %resultset = (); my $resultset_ref = \%resultset; my (%url_cols,%hidden_cols,%max_widths,$show_sql); #### Read in the standard form values my $apply_action = $parameters{'action'} || $parameters{'apply_action'}; my $TABLE_NAME = $parameters{'QUERY_NAME'}; #### Set some specific settings for this program my $CATEGORY="Get Proteins By Pep"; $TABLE_NAME="AT_GetProteinsByPep" unless ($TABLE_NAME); ($PROGRAM_FILE_NAME) = $sbeamsMOD->returnTableInfo($TABLE_NAME,"PROGRAM_FILE_NAME"); my $base_url = "$CGI_BASE_DIR/$SBEAMS_SUBDIR/$PROGRAM_FILE_NAME"; #### Get the columns and input types for this table/query my @columns = $sbeamsMOD->returnTableInfo($TABLE_NAME,"ordered_columns"); my %input_types = $sbeamsMOD->returnTableInfo($TABLE_NAME,"input_types"); #### Read the input parameters for each column my $n_params_found = $sbeams->parse_input_parameters( q=>$q,parameters_ref=>\%parameters, columns_ref=>\@columns,input_types_ref=>\%input_types); #$sbeams->printDebuggingInfo($q); #### If the apply action was to recall a previous resultset, do it my %rs_params = $sbeams->parseResultSetParams(q=>$q); if ($apply_action eq "VIEWRESULTSET") { $sbeams->readResultSet( resultset_file=>$rs_params{set_name}, resultset_ref=>$resultset_ref, query_parameters_ref=>\%parameters ); $n_params_found = 99; } #### Set some reasonable defaults if no parameters supplied unless ($n_params_found) { $parameters{input_form_format} = "minimum_detail"; } #### Apply any parameter adjustment logic # None #### Display the user-interaction input form $sbeams->display_input_form( TABLE_NAME=>$TABLE_NAME, CATEGORY=>$CATEGORY, apply_action=>$apply_action, PROGRAM_FILE_NAME=>$PROG_NAME, parameters_ref=>\%parameters, input_types_ref=>\%input_types, mask_user_context=> '1', ); #### Display the form action buttons $sbeams->display_form_buttons(TABLE_NAME=>$TABLE_NAME); #### Finish the upper part of the page and go begin the full-width #### data portion of the page $sbeams->display_page_footer(close_tables=>'YES', separator_bar=>'YES',display_footer=>'NO'); ######################################################################### #### Process all the constraints #### If no atlas_build_id was selected, stop here unless ($parameters{atlas_build_id}) { $sbeams->reportException( state => 'ERROR', type => 'INSUFFICIENT CONSTRAINTS', message => 'You must select at least one Atlas Build', ); return; } #### Build ATLAS_BUILD constraint my $atlas_build_clause = $sbeams->parseConstraint2SQL( constraint_column=>"AB.atlas_build_id", constraint_type=>"int_list", constraint_name=>"Atlas Build", constraint_value=>$parameters{atlas_build_id} ); return if ($atlas_build_clause eq '-1'); #### Build BIOSEQUENCE_NAME constraint my $biosequence_name_clause = $sbeams->parseConstraint2SQL( constraint_column=>"BS.biosequence_name", constraint_type=>"plain_text", constraint_name=>"BioSequence Name", constraint_value=>$parameters{biosequence_name_constraint} ); return if ($biosequence_name_clause eq '-1'); #### Build BIOSEQUENCE_GENE_NAME constraint my $biosequence_gene_name_clause = $sbeams->parseConstraint2SQL( constraint_column=>"BS.biosequence_gene_name", constraint_type=>"plain_text", constraint_name=>"BioSequence Gene Name", constraint_value=>$parameters{biosequence_gene_name_constraint} ); return if ($biosequence_gene_name_clause eq '-1'); #### Build BIOSEQUENCE_DESC constraint my $biosequence_desc_clause = $sbeams->parseConstraint2SQL( constraint_column=>"BS.biosequence_desc", constraint_type=>"plain_text", constraint_name=>"BioSequence Description", constraint_value=>$parameters{biosequence_desc_constraint} ); return if ($biosequence_desc_clause eq '-1'); #### Build BEST_PROBABILITY constraint my $best_probability_clause = $sbeams->parseConstraint2SQL( constraint_column=>"MAX(PI.best_probability)", constraint_type=>"flexible_float", constraint_name=>"Best Probability", constraint_value=>$parameters{best_probability_constraint} ); return if ($best_probability_clause eq '-1'); #### Build N_OBSERVATIONS constraint my $n_observations_per_peptide_clause = $sbeams->parseConstraint2SQL( constraint_column=>"PI.n_observations", constraint_type=>"flexible_int", constraint_name=>"Number of Observations", constraint_value=>$parameters{n_observations_per_peptide_constraint} ); return if ($n_observations_per_peptide_clause eq '-1'); #### Build N_OBSERVATIONS constraint my $n_observations_clause = $sbeams->parseConstraint2SQL( constraint_column=>"SUM(PI.n_observations)", constraint_type=>"flexible_int", constraint_name=>"Number of Observations", constraint_value=>$parameters{n_observations_constraint} ); return if ($n_observations_clause eq '-1'); #### Build n_protein_mappings constraint my $n_protein_mappings_clause = $sbeams->parseConstraint2SQL( constraint_column=>"MIN(PI.n_protein_mappings)", constraint_type=>"flexible_int", constraint_name=>"n_protein_mappings", constraint_value=>$parameters{n_protein_mappings_constraint} ); return if ($n_protein_mappings_clause eq '-1'); #### Build n_genome_locations constraint my $n_genome_locations_clause = $sbeams->parseConstraint2SQL( constraint_column=>"MIN(PI.n_genome_locations)", constraint_type=>"flexible_int", constraint_name=>"n_genome_locations", constraint_value=>$parameters{n_genome_locations_constraint} ); return if ($n_genome_locations_clause eq '-1'); #### Build CHROMOSOME constraint my $chromosome_clause = $sbeams->parseConstraint2SQL( constraint_column=>"PM.chromosome", constraint_type=>"plain_text", constraint_name=>"Chromosome", constraint_value=>$parameters{chromosome_constraint} ); return if ($chromosome_clause eq '-1'); #### Build START_IN_CHROMOSOME constraint my $start_in_chromosome_clause = $sbeams->parseConstraint2SQL( constraint_column=>"PM.start_in_chromosome", constraint_type=>"flexible_int", constraint_name=>"Start in Chromosome", constraint_value=>$parameters{start_in_chromosome_constraint} ); return if ($start_in_chromosome_clause eq '-1'); #### Build END_IN_CHROMOSOME constraint my $end_in_chromosome_clause = $sbeams->parseConstraint2SQL( constraint_column=>"PM.end_in_chromosome", constraint_type=>"flexible_int", constraint_name=>"end in Chromosome", constraint_value=>$parameters{End_in_chromosome_constraint} ); return if ($end_in_chromosome_clause eq '-1'); #### Build STRAND constraint my $strand_clause = $sbeams->parseConstraint2SQL( constraint_column=>"PM.strand", constraint_type=>"plain_text", constraint_name=>"Strand", constraint_value=>$parameters{strand_constraint} ); return if ($chromosome_clause eq '-1'); #### Build SAMPLE_ID constraint my $samples_clause = $sbeams->parseConstraint2SQL( constraint_column=>"PIS.sample_id", constraint_type=>"int_list", constraint_name=>"Samples ", constraint_value=>$parameters{sample_ids} ); return if ($samples_clause eq '-1'); ## get organism_id to pass on to url_cols my $tsql = qq~ SELECT BS.organism_id FROM $TBAT_BIOSEQUENCE_SET BS JOIN $TBAT_ATLAS_BUILD AB ON (AB.biosequence_set_id = BS.biosequence_set_id) where AB.atlas_build_id ='$parameters{atlas_build_id}' ~; my ($organism_id) = $sbeams->selectOneColumn($tsql) or die "\nERROR: Unable to find the organism_id" . " with $tsql\n\n"; $parameters{organism_id} = $organism_id; ## handle file upload and clause for sql for $parameters{upload_file} my $biosequence_names_clause; my %protein_hash; if ( $parameters{upload_file} ) { ## upload the file to a file handler my $fh = $q->upload('upload_file'); if (!$fh && $q->cgi_error) { print $q->header(-status=>$q->cgi_error); } if ( (-T $fh) && (-s $fh < 1000000)) ##size constraint of 10 MB, restrict $count < 30000 { my $count = 0; my $read_file=0; my $prt; ## protein list while ($prt=<$fh>) { chomp($prt); $prt =~ s/\s+$//; if ($prt) { $protein_hash{$prt} = $prt; $count = $count + 1; } last if ($count > 30000); } } ## join with a commas: my $protein_list = ""; foreach my $pr (keys %protein_hash) { $protein_list = "'$protein_hash{$pr}',$protein_list"; } ## trim off last comma: $protein_list =~ s/(.*)(,)$/$1/; unless ($protein_list =~ m/%/) { $log->debug(" in unless % ") ; $biosequence_names_clause = " AND BS.biosequence_name IN ( $protein_list )"; } if ($protein_list =~ m/%/) { my @tmp = split("," , $protein_list); $biosequence_names_clause = " AND ( BS.biosequence_name LIKE $tmp[0] "; for (my $i = 1; $i < scalar(@tmp); $i++ ){ if ( $tmp[$i] ){ $biosequence_names_clause = $biosequence_names_clause . " OR BS.biosequence_name LIKE $tmp[$i] "; } } $biosequence_names_clause = $biosequence_names_clause . " ) --end"; } } # if upload file #### Build ROWCOUNT constraint $parameters{row_limit} = 5000 unless ($parameters{row_limit} > 0 && $parameters{row_limit}<=1000000); my $limit_clause = "TOP $parameters{row_limit}"; #### Disable row limits $limit_clause = ""; #### Define some variables needed to build the query my @column_array; my $mapping_constraints = qq~ $chromosome_clause $start_in_chromosome_clause $end_in_chromosome_clause $strand_clause $samples_clause ~; #### Define the desired columns in the query #### [friendly name used in url_cols,SQL,displayed column title] @column_array = ( ["biosequence_name","BS.biosequence_name","Biosequence Name"], # ["representative_peptide_sequence","MIN(P.peptide_sequence)","Representative Peptide Sequence"], ["best_peptide_probability","MAX(STR(PI.best_probability,7,3))","Best Peptide Prob"], ["n_observations","SUM(PI.n_observations)","N Obs"], # ["n_samples","COUNT(PIS.peptide_instance_id)","N Samples"], ["n_distinct_peptides","COUNT(DISTINCT P.peptide_sequence)","N Distinct Peptides"], ["min_protein_mappings","MIN(PI.n_protein_mappings)","Min Protein Mappings"], ["min_genome_locations","MIN(PI.n_genome_locations)","Min Unique Locations on Genome"], ["biosequence_description","MAX(CAST(BS.biosequence_desc AS varchar(255)))","Biosequence Description"], ["biosequence_accessor","DBX.accessor","biosequence_accessor"], ["biosequence_accessor_suffix","DBX.accessor_suffix","biosequence_accessor_suffix"], ["biosequence_accession","BS.biosequence_accession","biosequence_accession"], ["organism_full_name","O.full_name","Organism"], ); #### Set flag to display SQL statement if user selected if ( $parameters{display_options} =~ /ShowSQL/ ) { $show_sql = 1; } #### Build the columns part of the SQL statement my %colnameidx = (); my @column_titles = (); ## Sends @column_array_ref to build_SQL_columns_list, which ## (1) appends the 2nd element in array to $columns_clause ## (2) fills %colnameidx_ref as a hash with key = 1st element ## and value = 3rd element, and (3) fills @column_titles_ref ## array with the 3rd element my $columns_clause = $sbeams->build_SQL_columns_list( column_array_ref=>\@column_array, colnameidx_ref=>\%colnameidx, column_titles_ref=>\@column_titles ); #### Define the SQL statement $sql = qq~ SELECT $limit_clause $columns_clause FROM $TBAT_PEPTIDE_INSTANCE PI INNER JOIN $TBAT_PEPTIDE P ON ( PI.peptide_id = P.peptide_id ) INNER JOIN $TBAT_PEPTIDE_INSTANCE_SAMPLE PIS ON ( PI.peptide_instance_id = PIS.peptide_instance_id ) INNER JOIN $TBAT_ATLAS_BUILD AB ON ( PI.atlas_build_id = AB.atlas_build_id ) LEFT JOIN $TBAT_BIOSEQUENCE_SET BSS ON ( AB.biosequence_set_id = BSS.biosequence_set_id ) LEFT JOIN $TB_ORGANISM O ON ( BSS.organism_id = O.organism_id ) LEFT JOIN $TBAT_PEPTIDE_MAPPING PM ON ( PI.peptide_instance_id = PM.peptide_instance_id ) LEFT JOIN $TBAT_BIOSEQUENCE BS ON ( PM.matched_biosequence_id = BS.biosequence_id ) LEFT JOIN $TB_DBXREF DBX ON ( BS.dbxref_id = DBX.dbxref_id ) WHERE 1 = 1 $atlas_build_clause $biosequence_name_clause $biosequence_names_clause $biosequence_gene_name_clause $biosequence_desc_clause $n_observations_per_peptide_clause $mapping_constraints GROUP BY BS.biosequence_name, DBX.accessor,DBX.accessor_suffix,BS.biosequence_accession,O.full_name HAVING 1 = 1 $best_probability_clause $n_observations_clause $n_protein_mappings_clause $n_genome_locations_clause ORDER BY BS.biosequence_name ~; #### Certain types of actions should be passed to links my $pass_action = "QUERY"; $pass_action = $apply_action if ($apply_action =~ /QUERY/i); #### Pass nearly all of the constraints down to a child query my @parameters_to_pass; my $parameters_list = ''; while ( ($key,$value) = each %input_types ) { if ($key ne 'sort_order' && $key ne 'display_options') { if ($parameters{$key}) { push(@parameters_to_pass,"$key=$parameters{$key}"); } } } if (@parameters_to_pass) { $parameters_list = join('&',@parameters_to_pass); } ## tailor the chromosomal url link to a genome browser by the organism name: my $organism_full_name = getOrganismFullName(organism_id => $organism_id); my $chrom_url_link; if ($organism_full_name =~ /^Halobacterium/) { $chrom_url_link = "http://www.genome.jp/kegg-bin/show_genomemap?ORG=hal&ACCESSION=\%$colnameidx{biosequence_name}V"; } else { ## assume it's an organism supported by Ensembl $chrom_url_link = "http://www.ensembl.org/$organism_full_name/contigview?c=\%$colnameidx{chromosome}V\:\%$colnameidx{start_in_chromosome}V&w=10000"; } #### Define the hypertext links for columns that need them %url_cols = ( 'Biosequence Name' => "$CGI_BASE_DIR/PeptideAtlas/GetProtein?protein_name=\%$colnameidx{biosequence_name}V&apply_action=$pass_action", 'Biosequence Name_ATAG' => 'TARGET="Win1" ONMOUSEOVER="window.status=\'Show more information about this protein\'; return true"', 'Representative Peptide Sequence' => "$CGI_BASE_DIR/PeptideAtlas/GetPeptide?_tab=3&atlas_build_id=$parameters{atlas_build_id}&searchWithinThis=Peptide+Sequence&searchForThis=\%$colnameidx{peptide_sequence}V&action=QUERY", 'Representative Peptide Sequence_ATAG' => 'TARGET="Win1" ONMOUSEOVER="window.status=\'Show more information for this peptide\'; return true"', ); #### Define columns that should be hidden in the output table %hidden_cols = ( 'biosequence_accession' => 1, 'biosequence_accessor' => 1, 'biosequence_accessor_suffix' => 1, 'Organism' => 1, 'PeptideAtlas Name' => 1, ); ######################################################################### #### If QUERY or VIEWRESULTSET was selected, display the data if ($apply_action =~ /QUERY/i || $apply_action eq "VIEWRESULTSET") { #### If the action contained QUERY, then fetch the results from #### the database if ($apply_action =~ /QUERY/i) { #### Show the SQL that will be or was executed $sbeams->display_sql(sql=>$sql) if ($show_sql); #### Fetch the results from the database server $sbeams->fetchResultSet( sql_query=>$sql, resultset_ref=>$resultset_ref, ); #### Post process the resultset postProcessResultset( rs_params_ref=>\%rs_params, resultset_ref=>$resultset_ref, query_parameters_ref=>\%parameters, column_titles_ref=>\@column_titles, ) if ($parameters{display_options} =~ /DASFormat/i); #### Store the resultset and parameters to disk resultset cache $rs_params{set_name} = "SETME"; $sbeams->writeResultSet( resultset_file_ref=>\$rs_params{set_name}, resultset_ref=>$resultset_ref, query_parameters_ref=>\%parameters, resultset_params_ref=>\%rs_params, query_name=>"$SBEAMS_SUBDIR/$PROGRAM_FILE_NAME", column_titles_ref=>\@column_titles, ); } #### Display the resultset $sbeams->displayResultSet( resultset_ref=>$resultset_ref, query_parameters_ref=>\%parameters, rs_params_ref=>\%rs_params, url_cols_ref=>\%url_cols, hidden_cols_ref=>\%hidden_cols, max_widths=>\%max_widths, column_titles_ref=>\@column_titles, base_url=>$base_url, ); #### Display the resultset controls $sbeams->displayResultSetControls( resultset_ref=>$resultset_ref, query_parameters_ref=>\%parameters, rs_params_ref=>\%rs_params, base_url=>$base_url, ); #### Display a plot of data from the resultset $sbeams->displayResultSetPlot( rs_params_ref=>\%rs_params, resultset_ref=>$resultset_ref, query_parameters_ref=>\%parameters, column_titles_ref=>\@column_titles, base_url=>$base_url, ); #### If QUERY was not selected, then tell the user to enter some parameters } else { if ($sbeams->invocation_mode() eq 'http') { print "

Select parameters above and press QUERY

\n"; } else { print "You need to supply some parameters to contrain the query\n"; } } } # end handle_request ############################################################################### # evalSQL # # Callback for translating Perl variables into their values, # especially the global table variables to table names ############################################################################### sub evalSQL { my $sql = shift; return eval "\"$sql\""; } # end evalSQL ############################################################################### # postProcessResultset # # Perform some additional processing on the resultset that would otherwise # be very awkward to do in SQL. ############################################################################### sub postProcessResultset { my %args = @_; my ($i,$element,$key,$value,$line,$result,$sql); #### Process the arguments list my $resultset_ref = $args{'resultset_ref'}; my $rs_params_ref = $args{'rs_params_ref'}; my $query_parameters_ref = $args{'query_parameters_ref'}; my $column_titles_ref = $args{'column_titles_ref'}; my %rs_params = %{$rs_params_ref}; my %parameters = %{$query_parameters_ref}; my $n_rows = scalar(@{$resultset_ref->{data_ref}}); my $cols = $resultset_ref->{column_hash_ref}; my $peptide_accession; my $nprot; my $nloc; for (my $i=0; $i<$n_rows; $i++) { $peptide_accession = $resultset_ref->{data_ref}->[$i]->[$cols->{peptide_accession}]; $nprot = $resultset_ref->{data_ref}->[$i]->[$cols->{n_protein_mappings}]; $nloc = $resultset_ref->{data_ref}->[$i]->[$cols->{n_genome_locations}]; if ($nprot > 1) { if ($nloc > 1) { $peptide_accession .= "(degen_NLoc=$nloc)"; } else { $peptide_accession .= "(degen)"; } $resultset_ref->{data_ref}->[$i]->[$cols->{peptide_accession}] = $peptide_accession; } elsif ($nloc > 1) { die("ERROR: nprot <=2 but $nloc > 1. This is wrong. Please report this error."); } } return 1; } # end postProcessResult ####################################################################### sub getOrganismFullName { my %args = @_; my $organism_id = $args{organism_id}; my $sql = qq~ SELECT full_name FROM $TB_ORGANISM WHERE organism_id = '$organism_id' AND record_status != 'D' ~; my ($full_name) = $sbeams->selectOneColumn($sql); ## replace spaces with _ $full_name =~ s/ /\_/g; return $full_name; }