Merge branch 'develop' version 1.0.9

README.org

@@ -190,6 +190,8 @@ The font is from [[fontlibrary.org]]:
  - [[https://fontlibrary.org/en/font/ranchers][Ranchers]] ([[http://scripts.sil.org/cms/scripts/page.php?site_id=nrsi&id=OFL][SIL Open Font License]])
 ** Changes
 [[https://travis-ci.org/molbiodiv/bcgTree][https://travis-ci.org/molbiodiv/bcgTree.svg?branch=master]]
+*** v1.0.9 <2017-03-03 Fr>
+ - Add parameters --min-proteomes and --all-proteomes (#21)
 *** v1.0.8 <2016-09-07 Mi>
  - Set default bootstraps to 100
  - Add description for reproduction of results in paper

bcgTreeGUI/src/bcgTree/gui/BcgTree.java

@@ -29,6 +29,7 @@
 import javax.imageio.ImageIO;
 import javax.swing.ImageIcon;
 import javax.swing.JButton;
+import javax.swing.JCheckBox;
 import javax.swing.JFileChooser;
 import javax.swing.JFrame;
 import javax.swing.JLabel;
@@ -44,7 +45,7 @@
 
 public class BcgTree extends JFrame {
 	private static final long serialVersionUID = 1L;
-	public static final String VERSION = "1.0.8";
+	public static final String VERSION = "1.0.9";
 	private static final int DEFAULT_TEXTFIELD_COLUMNS = 30;
 
 	public static void main(String[] args) {
@@ -74,6 +75,8 @@ public static void main(String[] args) {
 	private JTextField randomSeedXTextField;
 	private JTextField randomSeedPTextField;
 	private JTextField hmmfileTextField;
+	private JSpinner minProteomesSpinner;
+	private JCheckBox allProteomesCheckbox;
 
 	public BcgTree(){
 		self = this;
@@ -208,7 +211,7 @@ public void actionPerformed(ActionEvent e) {
 	}
 
 	private JPanel getAdvancedSettingsPanel(){
-		JPanel advancedSettingsPanel = new JPanel(new GridLayout(5, 2));
+		JPanel advancedSettingsPanel = new JPanel(new GridLayout(7, 2));
 		advancedSettingsPanel.add(new JLabel("--bootstraps"));
 		bootstrapSpinner = new JSpinner(new SpinnerNumberModel(10, 1, 1000, 1));
 		advancedSettingsPanel.add(bootstrapSpinner);
@@ -224,6 +227,18 @@ private JPanel getAdvancedSettingsPanel(){
 		advancedSettingsPanel.add(new JLabel("--hmmfile"));
 		hmmfileTextField = new JTextField(System.getProperty("user.dir")+"/../data/essential.hmm", DEFAULT_TEXTFIELD_COLUMNS);
 		advancedSettingsPanel.add(hmmfileTextField);
+		advancedSettingsPanel.add(new JLabel("--min-proteomes"));
+		minProteomesSpinner = new JSpinner(new SpinnerNumberModel(2, 1, 10000, 1));
+		advancedSettingsPanel.add(minProteomesSpinner);
+		advancedSettingsPanel.add(new JLabel("--all-proteomes"));
+		allProteomesCheckbox = new JCheckBox();
+		allProteomesCheckbox.addActionListener(new ActionListener() {
+			@Override
+			public void actionPerformed(ActionEvent e) {
+				minProteomesSpinner.setEnabled(!allProteomesCheckbox.isSelected());
+			}
+		});
+		advancedSettingsPanel.add(allProteomesCheckbox);
 		return advancedSettingsPanel;
 	}
 
@@ -255,6 +270,11 @@ public void actionPerformed(ActionEvent e) {
 				writer.println("--bootstraps="+bootstrapSpinner.getValue());
 				writer.println("--threads="+threadsSpinner.getValue());
 				writer.println("--hmmfile=\""+hmmfileTextField.getText()+"\"");
+				if(allProteomesCheckbox.isSelected()){
+					writer.println("--all-proteomes");
+				} else {
+					writer.println("--min-proteomes="+minProteomesSpinner.getValue());
+				}
 				String pSeed = randomSeedPTextField.getText();
 				if(! pSeed.equals("")){
 					writer.println("--raxml-p-parsimonyRandomSeed="+pSeed);

bin/bcgTree.pl

@@ -54,23 +54,23 @@ =head1 OPTIONS
 
 $options{'outdir=s'} = \( my $opt_outdir="bcgTree" );
 
-=item [--help] 
+=item [--help]
 
 show help
 
 =cut
 
 $options{'help|h|?'} = \( my $opt_help );
 
-=item [--version] 
+=item [--version]
 
 show version number of bcgTree and exit
 
 =cut
 
 $options{'version'} = \( my $opt_version );
 
-=item [--check-external-programs] 
+=item [--check-external-programs]
 
 Check if all of the required external programs can be found and are executable, then exit.
 Report table with program, status (ok or !fail!) and path.
@@ -138,6 +138,30 @@ =head1 OPTIONS
 
 =back
 
+=item [--min-proteomes=<INT>]
+
+Minimum number of proteomes in which a gene must occur in order to be kept. Default: 2
+All genes with less hits are discarded prior to the alignment step.
+This option is ignored if --all-proteomes is set.
+
+=cut
+
+$options{'min-proteomes=i'} = \( my $opt_min_proteomes = 2 );
+
+=back
+
+=item [--all-proteomes]
+
+Sets --min-proteomes to the total number of proteomes supplied. Default: not set
+All genes that do not hit all of the proteomes are discarded prior to the alignment step.
+If set --min-proteomes is ignored.
+
+=cut
+
+$options{'all-proteomes'} = \( my $opt_all_proteomes = 0 );
+
+=back
+
 =item [--hmmfile=<PATH>]
 
 Path to HMM file to be used for hmmsearch. Default: <bcgTreeDir>/data/essential.hmm
@@ -204,6 +228,7 @@ =head1 CODE
 );
 
 my $L = Log::Log4perl::get_logger();
+$opt_min_proteomes = 0+(keys %{$opt_proteome}) if($opt_all_proteomes);
 my $bcgTree = bcgTree->new({
 	'proteome' => $opt_proteome,
 	'outdir' => $opt_outdir,
@@ -216,7 +241,8 @@ =head1 CODE
 	'bootstraps' => $opt_bootstraps,
 	'hmmfile' => $opt_hmmfile,
 	'raxml-p' => $opt_raxml_p,
-	'raxml-x' => $opt_raxml_x
+	'raxml-x' => $opt_raxml_x,
+    'min-proteomes' => $opt_min_proteomes
 });
 $bcgTree->check_existence_of_fasta_files();
 $bcgTree->rename_fasta_headers();
@@ -239,7 +265,7 @@ sub check_external_programs{
 		my $result = 'ok';
 		if(! -X $path){
 			$result = '!fail!';
-			$fail = 1;	
+			$fail = 1;
 		}
 		printf "%-10s%6s\t%s\n", $p, $result, $path;
 	}

lib/bcgTree.pm

@@ -11,7 +11,7 @@ use File::Path qw(make_path);
 use File::Spec;
 use Fasta::Parser;
 
-our $VERSION = '1.0.8';
+our $VERSION = '1.0.9';
 
 my $L = Log::Log4perl::get_logger();
 
@@ -152,13 +152,13 @@ sub collect_best_hmm_hits{
 			$count++;
 		}
 		$L->info("Wrote $count ids for gene $g");
-		if($count < 2){
+		if($count < $self->{'min-proteomes'}){
 			$L->warn("Gene $g only present in $count proteome - removing from further analyses");
 			delete $gene_id_map{$g};
 		}
 		close OUT or $L->logdie("Error closing $out/$g.ids. $!");
 	}
-	# Add the remaining list of genes to the object 
+	# Add the remaining list of genes to the object
 	$self->{genes} = [keys %gene_id_map];
 	$L->logdie("Not a single gene is present in more than one proteome - Exiting") if((keys %gene_id_map) < 1);
 	# Write absence presence list