josemduarte · Oct 26, 2022 · Oct 3, 2022 · Oct 3, 2022 · Oct 3, 2022 · Oct 3, 2022
diff --git a/biojava-core/src/main/java/org/biojava/nbio/core/sequence/features/DBReferenceInfo.java b/biojava-core/src/main/java/org/biojava/nbio/core/sequence/features/DBReferenceInfo.java
@@ -47,7 +47,7 @@ public class DBReferenceInfo extends Qualifier {
 	 * @param id
 	 */
 	public DBReferenceInfo(String database, String id){
-		super("dbxref","");
+		super("db_xref","");
 		this.database = database;
 		this.id = id;
 	}

diff --git a/biojava-core/src/main/java/org/biojava/nbio/core/sequence/io/GenbankReader.java b/biojava-core/src/main/java/org/biojava/nbio/core/sequence/io/GenbankReader.java
@@ -26,6 +26,7 @@
 package org.biojava.nbio.core.sequence.io;
 
 import org.biojava.nbio.core.exceptions.CompoundNotFoundException;
+import org.biojava.nbio.core.sequence.AccessionID;
 import org.biojava.nbio.core.sequence.DataSource;
 import org.biojava.nbio.core.sequence.TaxonomyID;
 import org.biojava.nbio.core.sequence.features.DBReferenceInfo;
@@ -162,8 +163,14 @@ public LinkedHashMap<String,S> process(final int max) throws IOException, Compou
 			if(seqString==null) break;
 			@SuppressWarnings("unchecked")
 			S sequence = (S) sequenceCreator.getSequence(seqString, 0);
-			genbankParser.getSequenceHeaderParser().parseHeader(genbankParser.getHeader(), sequence);
-
+			GenericGenbankHeaderParser<S, C> genbankHeaderParser = genbankParser.getSequenceHeaderParser();			
+			genbankHeaderParser.parseHeader(genbankParser.getHeader(), sequence);			
+			String id = genbankHeaderParser.getAccession();
+			int version = genbankHeaderParser.getVersion();
+			String identifier = genbankHeaderParser.getIdentifier();
+			AccessionID accession = new AccessionID(id , DataSource.GENBANK, version, identifier);
+			sequence.setAccession(accession);
+
 			// add features to new sequence
 			genbankParser.getFeatures().values().stream()
 			.flatMap(List::stream)

diff --git a/biojava-core/src/main/java/org/biojava/nbio/core/sequence/io/GenbankSequenceParser.java b/biojava-core/src/main/java/org/biojava/nbio/core/sequence/io/GenbankSequenceParser.java
@@ -109,8 +109,10 @@ public class GenbankSequenceParser<S extends AbstractSequence<C>, C extends Comp
 	protected static final String PRIMARY = "PRIMARY";
 	protected static final String DBLINK = "DBLINK";
 	protected static final String END_SEQUENCE_TAG = "//";
-	// locus line
-	protected static final Pattern lp = Pattern.compile("^(\\S+)\\s+(\\d+)\\s+(bp|BP|aa|AA)\\s{0,4}(([dmsDMS][sS]-)?(\\S+))?\\s*(circular|CIRCULAR|linear|LINEAR)?\\s*(\\S+)?\\s*(\\S+)?$");
+	// locus line with name that may contain spaces but must start and end with non whitespace character
+	protected static final Pattern lp = Pattern.compile("^(\\S+[\\S ]*\\S*)\\s+(\\d+)\\s+(bp|BP|aa|AA)\\s{0,4}(([dmsDMS][sS]-)?(\\S+))?\\s*(circular|CIRCULAR|linear|LINEAR)?\\s*(\\S+)?\\s*(\\S+)?$");
+	// locus line with no name
+	protected static final Pattern lp2 = Pattern.compile("^(\\d+)\\s+(bp|BP|aa|AA)\\s{0,4}(([dmsDMS][sS]-)?(\\S+))?\\s*(circular|CIRCULAR|linear|LINEAR)?\\s*(\\S+)?\\s*(\\S+)?$");	
 	// version line
 	protected static final Pattern vp = Pattern.compile("^(\\S*?)(\\.(\\d+))?(\\s+GI:(\\S+))?$");
 	// reference line
@@ -198,10 +200,12 @@ private void parseFeatureTag(List<String[]> section) {
 				if (gbFeature == null) {
 					throw new ParserException("Malformed GenBank file: found a qualifier without feature.");
 				}
+				Boolean needsQuotes = false;
 				key = key.substring(1); // strip leading slash
-				val = val.replaceAll("\\s*[\\n\\r]+\\s*", " ").trim();
+				val = val.replaceAll("\\s*[\\n\\r]+\\s*", " ").trim();				
 				if (val.endsWith("\"")) {
 					val = val.substring(1, val.length() - 1); // strip quotes
+					needsQuotes = true; // as the value has quotes then set that it needs quotes when written back out
 				}
 				// parameter on old feature
 				if (key.equals("db_xref")) {
@@ -210,6 +214,7 @@ private void parseFeatureTag(List<String[]> section) {
 						String dbname = m.group(1);
 						String raccession = m.group(2);
 						DBReferenceInfo xref = new DBReferenceInfo(dbname, raccession);
+						xref.setNeedsQuotes(needsQuotes);
 						gbFeature.addQualifier(key, xref);
 
 						ArrayList<DBReferenceInfo> listDBEntry = new ArrayList<>();
@@ -219,17 +224,17 @@ private void parseFeatureTag(List<String[]> section) {
 						throw new ParserException("Bad dbxref");
 					}
 				} else if (key.equalsIgnoreCase("organism")) {
-					Qualifier q = new Qualifier(key, val.replace('\n', ' '));
+					Qualifier q = new Qualifier(key, val.replace('\n', ' '), needsQuotes);
 					gbFeature.addQualifier(key, q);
 				} else {
 					if (key.equalsIgnoreCase("translation") || key.equals("anticodon")
 							|| key.equals("transl_except")) {
 						// strip spaces from sequence
 						val = val.replaceAll("\\s+", "");
-						Qualifier q = new Qualifier(key, val);
+						Qualifier q = new Qualifier(key, val, needsQuotes);
 						gbFeature.addQualifier(key, q);
 					} else {
-						Qualifier q = new Qualifier(key, val);
+						Qualifier q = new Qualifier(key, val, needsQuotes);
 						gbFeature.addQualifier(key, q);
 					}
 				}
@@ -304,9 +309,12 @@ private void parseLocusTag(List<String[]> section) {
 		String loc = section.get(0)[1];
 		header = loc;
 		Matcher m = lp.matcher(loc);
+		Matcher m2 = lp2.matcher(loc);		
 		if (m.matches()) {
-			headerParser.setName(m.group(1));
-			headerParser.setAccession(m.group(1)); // default if no accession found
+			//remove any preceding or trailing whitespace from the locus name
+			String name = m.group(1).trim().replaceAll(" ","_");		
+			headerParser.setName(name);
+			headerParser.setAccession(name); // default if no accession found			
 			long sequenceLength = Long.valueOf(m.group(2));
 			String lengthUnits = m.group(3);
 			String type = m.group(6);
@@ -333,6 +341,37 @@ private void parseLocusTag(List<String[]> section) {
 
 			log.debug("compound type: {}", compoundType.getClass().getSimpleName());
 
+		} else if (m2.matches()) {
+			// Locus Name Missing - use different Locus regex
+			headerParser.setName("");
+			headerParser.setAccession(""); // default if no accession found			
+			long sequenceLength = Long.valueOf(m2.group(1));
+			String lengthUnits = m2.group(2);
+			String type = m2.group(5);
+
+			if (lengthUnits.equalsIgnoreCase("aa")) {
+				compoundType = AminoAcidCompoundSet.getAminoAcidCompoundSet();
+			} else if (lengthUnits.equalsIgnoreCase("bp")) {
+				if (type != null) {
+					if (type.contains("RNA")) {
+						compoundType = RNACompoundSet.getRNACompoundSet();
+					} else {
+						compoundType = DNACompoundSet.getDNACompoundSet();
+					}
+				} else {
+					compoundType = DNACompoundSet.getDNACompoundSet();
+				}
+			}
+
+			if (m2.group(6) != null) isCircularSequence = m2.group(6).equalsIgnoreCase("circular");
+
+			// configure location parser with needed information
+			locationParser.setSequenceLength(sequenceLength);
+			locationParser.setSequenceCircular(isCircularSequence);
+
+			log.debug("compound type: {}", compoundType.getClass().getSimpleName());
+
+
 		} else {
 			throw new ParserException("Bad locus line");
 		}

diff --git a/biojava-core/src/main/java/org/biojava/nbio/core/sequence/io/GenbankWriter.java b/biojava-core/src/main/java/org/biojava/nbio/core/sequence/io/GenbankWriter.java
@@ -92,10 +92,9 @@ public void process() throws Exception {
 		PrintWriter writer = new PrintWriter(os);
 		for (S sequence : sequences) {
 			String header = headerFormat.getHeader(sequence);
-			writer.format(header);
+			writer.print(header);
 			writer.println();
-			// os.write(lineSep);
-
+			// os.write(lineSep);			
 			/*
 			 * if isinstance(record.seq, UnknownSeq): #We have already recorded
 			 * the length, and there is no need #to record a long sequence of

diff --git a/biojava-core/src/main/java/org/biojava/nbio/core/sequence/io/GenbankWriterHelper.java b/biojava-core/src/main/java/org/biojava/nbio/core/sequence/io/GenbankWriterHelper.java
@@ -135,6 +135,24 @@ public static void writeNucleotideSequence(OutputStream outputStream,
 		genbankWriter.process();
 	}
 
+	/**
+	 * Write a collection of NucleotideSequences to a file using the NucleotideSequences
+	 * original header as the LOCUS line rather than generating it
+	 *
+	 * @param outputStream
+	 * @param dnaSequences
+	 * @throws Exception
+	 */
+
+	public static void writeNucleotideSequenceOriginal(OutputStream outputStream, Collection<DNASequence> dnaSequences)
+			throws Exception {
+		GenericGenbankHeaderFormat<DNASequence, NucleotideCompound> genericGenbankHeaderFormat = new GenericGenbankHeaderFormat<DNASequence, NucleotideCompound>(
+				true);
+		GenbankWriter<DNASequence, NucleotideCompound> genbankWriter = new GenbankWriter<DNASequence, NucleotideCompound>(
+				outputStream, dnaSequences, genericGenbankHeaderFormat);
+		genbankWriter.process();
+	}
+
 	/**
 	 * Write a sequence to a file
 	 *

diff --git a/biojava-core/src/main/java/org/biojava/nbio/core/sequence/io/GenericGenbankHeaderFormat.java b/biojava-core/src/main/java/org/biojava/nbio/core/sequence/io/GenericGenbankHeaderFormat.java
@@ -20,6 +20,7 @@
  */
 package org.biojava.nbio.core.sequence.io;
 
+import org.biojava.nbio.core.sequence.AccessionID;
 import org.biojava.nbio.core.sequence.compound.AminoAcidCompoundSet;
 import org.biojava.nbio.core.sequence.compound.DNACompoundSet;
 import org.biojava.nbio.core.sequence.compound.RNACompoundSet;
@@ -38,6 +39,7 @@ public class GenericGenbankHeaderFormat<S extends AbstractSequence<C>, C extends
 	private static final int HEADER_WIDTH = 12;
 	private static final String lineSep = "%n";
 	private String seqType = null;
+	private boolean useOriginalHeader = false;
 
 	public GenericGenbankHeaderFormat() {
 		seqType = null;
@@ -46,6 +48,10 @@ public GenericGenbankHeaderFormat() {
 	public GenericGenbankHeaderFormat(String seqType) {
 		this.seqType = seqType;
 	}
+
+	public GenericGenbankHeaderFormat(boolean useOriginalHeader) {
+		this.useOriginalHeader = useOriginalHeader;
+	}
 
 	/**
 	 * Used in the the 'header' of each GenBank record.
@@ -222,6 +228,22 @@ private String _write_the_first_line(S sequence) {
 		 */
 	}
 
+	/**
+	 * Write the original LOCUS line.
+	 *
+	 * @param sequence
+	 */
+	private String _write_original_first_line(S sequence) {
+
+		StringBuilder sb = new StringBuilder();
+		Formatter formatter = new Formatter(sb, Locale.US);
+		formatter.format("LOCUS       %s" + lineSep,
+				StringManipulationHelper.padRight(sequence.getOriginalHeader(), 16));
+		String output = formatter.toString();
+		formatter.close();
+		return output;
+	}
+
 	/**
 	 * This is a bit complicated due to the range of possible ways people might
 	 * have done their annotation... Currently the parser uses a single string
@@ -243,12 +265,30 @@ private String _write_comment(S sequence) {
 
 	@Override
 	public String getHeader(S sequence) {
-		String header = _write_the_first_line(sequence);
-		String acc_with_version;
+
+		String header;
+
+		if (useOriginalHeader) {
+			header = _write_original_first_line(sequence);
+		} else {
+			header = _write_the_first_line(sequence);
+		}
+
+		AccessionID accessionIdObj = sequence.getAccession();
 		String accession;
+		String acc_with_version;
+
 		try {
-			acc_with_version = sequence.getAccession().getID();
-			accession = acc_with_version.split("\\.", 1)[0];
+			accession = accessionIdObj.getID();
+
+			if (accessionIdObj.getIdentifier() != null) {
+				acc_with_version = sequence.getAccession().getID() + "." + sequence.getAccession().getVersion() + " GI:"
+						+ accessionIdObj.getIdentifier();
+
+			} else {
+				acc_with_version = sequence.getAccession().getID() + "." + sequence.getAccession().getVersion();
+			}
+
 		} catch (Exception e) {
 			acc_with_version = "";
 			accession = "";
@@ -262,13 +302,6 @@ public String getHeader(S sequence) {
 		header += _write_multi_line("VERSION", acc_with_version);
 
 		/*
-		 * gi = self._get_annotation_str(record, "gi", just_first=True)
-		 *
-		 * self._write_single_line("ACCESSION", accession) if gi != ".":
-		 * self._write_single_line("VERSION", "%s  GI:%s" \ % (acc_with_version,
-		 * gi)) else: self._write_single_line("VERSION", "%s" %
-		 * (acc_with_version))
-		 *
 		 * #The NCBI only expect two types of link so far, #e.g. "Project:28471"
 		 * and "Trace Assembly Archive:123456" #TODO - Filter the dbxrefs list
 		 * to just these? self._write_multi_entries("DBLINK", record.dbxrefs)
@@ -319,7 +352,7 @@ public String getHeader(S sequence) {
 			header += _write_feature(feature, rec_length);
 		}
 
-		return header;
+		return String.format(header);
 	}
 
 }