Bio-Homology-InterologWalk

 view release on metacpan or  search on metacpan

lib/Bio/Homology/InterologWalk.pm  view on Meta::CPAN

          $all_seen{$idIN} = 1;    $all_seen{$idOUT} = 1;
          
          #then I count what fraction of the original ids actually made it to feature any putative interactions
          $old_id_present{$idIN} = 1 if(exists($start_data_set{$idIN}));
          $old_id_present{$idOUT} = 1 if(exists($start_data_set{$idOUT}));
     
          #lastly I store those new ids never seen in the starting dataset
          $new_id_set{$idOUT} = 1 unless(exists($start_data_set{$idOUT})); 
     }
     my $number_of_old_IDs = keys %old_id_present;
     my $percentage = ($number_of_old_IDs / $number_of_elements_start_ds) * 100;
     print("Number of IDs from the original dataset that appear in the network: $number_of_old_IDs\n");
     print("Percentage of IDs from the original dataset that appear in the final dataset: $percentage\n");
     
     my $number_of_new_IDs = keys %new_id_set;
     my $number_of_network_nodes = keys %all_seen;
     $percentage= ($number_of_new_IDs / $number_of_network_nodes) * 100;
     if($onetoone_only){
          print("Number of total UNIQUE IDs in interaction dataset (considering ONE-TO-ONE ortologies only): $number_of_network_nodes\n");
     }else{
          print("Number of total UNIQUE IDs in interaction dataset: $number_of_network_nodes\n");
     }
     print("Number of NEW ids (e.g. not seen in starting data set): $number_of_new_IDs\n");
     print("Percentage of new ids over the total: $percentage\n");
     
     
     #I save all the new ids in a flat file. This might be useful to do some analysis of their functional annotation
     foreach my $id (sort keys %new_id_set){
          #print ("$id\t$new_id_set{$id}\n");
          print $out_data $id, "\n";
     }
     $sth->finish();
     $dbh->disconnect();
     close($start_data);

lib/Bio/Homology/InterologWalk.pm  view on Meta::CPAN

          }

          foreach my $homology_member (@{$genelist}){
               $DF_orthologue_id = $homology_member->stable_id;
               next if ($DF_orthologue_id eq $init_id);#I dont want to print again the gene name
          
               $DF_oname = $homology_member->display_label;

               #OPI
               my $pairwise_alignment_from_multiple = $homology->get_SimpleAlign;
               $DF_opi = $pairwise_alignment_from_multiple->overall_percentage_identity;
               #$opi = sprintf("%.3f", $overall_pid); #rounded
          
               $DF_orthologue_id = '-' if(!$DF_orthologue_id);
               $DF_oname         = '-' if(!$DF_oname); 
               $DF_odesc         = '-' if(!$DF_odesc);
               $DF_dnds          = '-' if(!$DF_dnds); 
               $DF_opi           = '-' if(!$DF_opi);
               $DF_fsa_x         = '-' if(!$DF_fsa_x); 
               $DF_fsa_y         = '-' if(!$DF_fsa_y);
               $DF_nndist        = '-' if(!$DF_nndist);     

lib/Bio/Homology/InterologWalk.pm  view on Meta::CPAN

                                                        );
 Purpose   : This is used to analyse several ancillary data fields obtained alongside the actual 
             putative PPI IDs and collate them into an Interolog Prioritisation Index (IPX), to associate a
             numerical index to each putative PPI based on biological metadata. The index will take into account 
             a number of features related to each of the steps involved in the orthology walk. 
             We can divide the metadata features in two broad classes:
             - features related to the interaction. These include: Interaction Type, Interaction 
               Detection Method, Interaction coming from a SPOKE-expanded complex, interaction recon-
               firmed through multiple taxa, interaction reconfirmed through multiple detection methods
             - features related to the two orthology mappings. These include: orthology type 
               (one-to-one, one-to-many, many-to-one, many-to-many), OPI (percentage identity of the 
               conserved columns - see Bio::SimpleAlign), node to node distance, distance from the 
               first shared ancestor, (under development) dN/dS ratio
             The IPX computation will also involve a normalisation stage. The subroutine requires 
             five arguments (meanscore_x) representing mean values to be used for normalisation.
             The actual means are computed in get_mean_scores(), which is pre-requisite to 
             compute_prioritisation_index().
 Returns   : success/failure
 Argument  : -input_path : path to the input tsv file. A suitable input for this subroutine is the 
             final output of the orthology walk pipeline (see doInterologWalk.pl for usage guidelines).
              input file should have .06out extension

lib/Bio/Homology/InterologWalk.pm  view on Meta::CPAN

             -term_graph :  a Go::Parser graph object obtained from parse_ontology() containing a 
              network representation of the PSI-MI controlled vocabulary of terms.
             -meanscore_em : mean experimental method score for normalisation
             -meanscore_it : mean interaction type score for normalisation  
             -meanscore_dm : mean detection method score for normalisation   
             -meanscore_me_dm : mean 'multiple detection methods' score for normalisation
             -meanscore_me_taxa : mean 'multiple taxa' score for normalisation
 Throws    : -
 Comment   : -

See Also   : L<http://search.cpan.org/~cjfields/BioPerl-1.6.1/Bio/SimpleAlign.pm#overall_percentage_identity>, L</get_mean_scores>, C<doScores.pl> for sample usage

=cut


sub compute_prioritisation_index{
     my %args = @_;
     my $in_path                   = $args{input_path};
     my $out_path                  = $args{output_path};
     my $score_path                = $args{score_path};
     my $graph                     = $args{term_graph};

scripts/Data/psi-mi.obo  view on Meta::CPAN

id: MI:2046
name: lethal dose 50
def: "The LD50 is the dose that kills half (50%) of the animals tested" [PMID:14755292]
subset: Drugable
synonym: "ld50" EXACT PSI-MI-alternate []
synonym: "lethal dose 50 %" EXACT PSI-MI-alternate []
is_a: MI:0640 ! parameter type

[Term]
id: MI:2047
name: percentage of plasma protein binding
def: "Percentage of the drug that is bound in plasma proteins" [PMID:14755292]
subset: Drugable
synonym: "plasma prot binding" EXACT PSI-MI-short []
synonym: "protein binding %" EXACT PSI-MI-alternate []
is_a: MI:0640 ! parameter type

[Term]
id: MI:2048
name: drug biotransformation
def: "The chemical conversion of drugs to other compounds in the body, excluding degradation due to any inherent chemical instability of drugs in biological media." [PMID:14755292]