biojava · josemduarte · Sep 20, 2021 · Jul 26, 2021 · Aug 3, 2021 · Aug 3, 2021
diff --git a/...ationtest/src/test/java/org/biojava/nbio/structure/test/io/TestLongPdbVsMmCifParsing.java b/...ationtest/src/test/java/org/biojava/nbio/structure/test/io/TestLongPdbVsMmCifParsing.java
@@ -287,7 +287,7 @@ private void testHeader(Structure sPdb, Structure sCif) {
 		// description is set in CIF parser to same as classification (_struct_keywords.pdbx_keywords field)
 		// while in PDB parser it is simply not set
 		//assertNotNull("pdb description null",hPdb.getDescription());
-		assertNotNull("cif description null",hCif.getDescription());
+		//assertNotNull("cif description null",hCif.getDescription());  //If we will enable this test again, we may use assertNull instead
 		//assertEquals("failed getDescription:",hPdb.getDescription().toLowerCase(), hCif.getDescription().toLowerCase());
 
 		assertEquals("failed getDepDate:",hPdb.getDepDate(), hCif.getDepDate());

diff --git a/.../src/test/java/org/biojava/nbio/structure/test/io/cif/CifFileSupplierIntegrationTest.java b/.../src/test/java/org/biojava/nbio/structure/test/io/cif/CifFileSupplierIntegrationTest.java
@@ -21,6 +21,7 @@
 import java.net.URL;
 import java.util.Arrays;
 
+import static org.junit.Assert.assertArrayEquals;
 import static org.junit.Assert.assertEquals;
 import static org.junit.Assert.assertNotNull;
 
@@ -47,6 +48,18 @@ public void test1A2C() throws IOException {
         testRoundTrip("1A2C");
     }
 
+    @Test
+    public void test3CDL() throws IOException {
+    	// a structure with insertion codes
+    	testRoundTrip("3CDL");
+    }
+
+    @Test
+    public void test6ELW() throws IOException {
+    	// a structure with insertion codes
+    	testRoundTrip("6ELW");
+    }
+
     private static void testRoundTrip(String pdbId) throws IOException {
         URL url = new URL("https://files.rcsb.org/download/" + pdbId + ".cif");
         Structure originalStruct = CifStructureConverter.fromURL(url);
@@ -57,6 +70,10 @@ private static void testRoundTrip(String pdbId) throws IOException {
         assertNotNull(readStruct);
         assertEquals(originalStruct.getChains().size(), readStruct.getChains().size());
         assertEquals(originalStruct.nrModels(), readStruct.nrModels());
+
+        assertArrayEquals("Keywords Are not preserved", 
+        		originalStruct.getPDBHeader().getKeywords().toArray(), 
+        		readStruct.getPDBHeader().getKeywords().toArray());
 
         for (int i = 0; i < originalStruct.nrModels(); i++) {
             assertEquals(originalStruct.getModel(i).size(), readStruct.getModel(i).size());

diff --git a/biojava-structure/src/main/java/org/biojava/nbio/structure/PDBHeader.java b/biojava-structure/src/main/java/org/biojava/nbio/structure/PDBHeader.java
@@ -35,7 +35,12 @@
 
 /**
  * A class that contains PDB Header information.
- *
+ * In contrast to what the name suggests, this class does not represent a
+ * direct mapping of the Header section of the PDB legacy file format.
+ * Instead, it holds the information that is not directly related to the 
+ * structure data. Such information may exist in some cases and may not exist in
+ * other cases.
+ * 
  * @author Andreas Prlic
  * @since 1.6
  *
@@ -47,7 +52,10 @@ public class PDBHeader implements PDBRecord {
 	private static final Logger logger = LoggerFactory.getLogger(PDBHeader.class);
 
 	private String title;
+	/**@deprecated This field should not be used. It will be removed later. 
+	 * Use {@link #getKeywords()} instead. */
 	private String description;
+	private List<String> keywords;
 	private String idCode;
 	private String classification;
 
@@ -92,6 +100,8 @@ public PDBHeader(){
 		bioAssemblies = new LinkedHashMap<Integer, BioAssemblyInfo>();
 		crystallographicInfo = new PDBCrystallographicInfo();
 
+		keywords       = new ArrayList<>();
+
 	}
 
 	/** String representation
@@ -589,9 +599,20 @@ public String getTitle() {
 	public void setTitle(String title) {
 		this.title = title;
 	}
+
+	/**@deprecated will be removed later. Use {@link #getKeywords()} if you use 
+	 * <code>description</code> to keep the keywords.
+	 * @return
+	 */
+	@Deprecated
 	public String getDescription() {
 		return description;
 	}
+	/**@deprecated will be removed later. Use {@link #getKeywords()} if you use 
+	 * <code>description</code> to keep the keywords.
+	 * @param description
+	 */
+	@Deprecated
 	public void setDescription(String description) {
 		this.description = description;
 	}
@@ -683,4 +704,22 @@ public float getRwork() {
 	public void setRwork(float rWork) {
 		this.rWork = rWork;
 	}
+
+	/**
+	 * Gets the keywords (KEYWODS) record of the structure
+	 * @return The keywords in a <code>List&lt;String&gt;</code>
+	 * @since 6.0.0
+	 */
+	public List<String> getKeywords() {
+		return keywords;
+	}
+
+	/**
+	 * Sets the KEYWODS record of the structure.
+	 * @param keywords The keywords in a <code>List&lt;String&gt; to set.</code>
+	 * @since 6.0.0
+	 */
+	public void setKeywords(List<String> keywords) {
+		this.keywords = keywords;
+	}
 }
diff --git a/biojava-structure/src/main/java/org/biojava/nbio/structure/Structure.java b/biojava-structure/src/main/java/org/biojava/nbio/structure/Structure.java
@@ -665,7 +665,9 @@ public interface Structure extends Cloneable, Serializable {
 	EntityInfo getEntityById(int entityId);
 
 	/**
-	 * Return the header information for this PDB file
+	 * Return the header information for this PDB file.
+	 * <b>N.B.</b> Take care when you blindly use the returned object from this method,
+	 * because it might be null in some cases.
 	 *
 	 * @return the PDBHeader object
 	 */
@@ -788,4 +790,5 @@ public interface Structure extends Cloneable, Serializable {
 	 *  returned the same value as {@link #getPDBCode()}
 	 */
 	String getIdentifier();
+
 }
diff --git a/biojava-structure/src/main/java/org/biojava/nbio/structure/io/PDBFileParser.java b/biojava-structure/src/main/java/org/biojava/nbio/structure/io/PDBFileParser.java
@@ -181,6 +181,7 @@ public class PDBFileParser  {
 	private List<String> compndLines = new ArrayList<String>();
 	private List<String> sourceLines = new ArrayList<String>();
 	private List<String> journalLines = new ArrayList<String>();
+	private List<String> keywordsLines = new ArrayList<String>();
 	private List<DBRef> dbrefs;
 	private Map<String, Site> siteMap = new LinkedHashMap<String, Site>();
 	private Map<String, List<ResidueNumber>> siteToResidueMap = new LinkedHashMap<String, List<ResidueNumber>>();
@@ -2442,6 +2443,8 @@ private void pdb_SITE_Handler(String line){
 
 	//Site variable related to parsing the REMARK 800 records.
 	Site site;
+
+	private String[] keywords;
 	private void pdb_REMARK_800_Handler(String line){
 
 		if (params.isHeaderOnly()) return;
@@ -2595,6 +2598,7 @@ public  Structure parsePDBFile(BufferedReader buf)
 		current_compound = null;
 		sourceLines.clear();
 		compndLines.clear();
+		keywordsLines.clear();
 		isLastCompndLine = false;
 		isLastSourceLine = false;
 		prevMolId = -1;
@@ -2660,6 +2664,8 @@ else if (recordName.equals("SOURCE"))
 					sourceLines.add(line); //pdb_SOURCE_Handler
 				else if (recordName.equals("COMPND"))
 					compndLines.add(line); //pdb_COMPND_Handler
+				else if (recordName.equals("KEYWDS"))
+					keywordsLines.add(line);
 				else if (recordName.equals("JRNL"))
 					pdb_JRNL_Handler(line);
 				else if (recordName.equals("EXPDTA"))
@@ -2693,6 +2699,8 @@ else if ( params.isParseSecStruc()) {
 		}
 
 		makeCompounds(compndLines, sourceLines);
+
+		handlePDBKeywords(keywordsLines);
 
 		triggerEndFileChecks();
 
@@ -2757,6 +2765,52 @@ private void makeCompounds(List<String> compoundList,
 
 	}
 
+	/**Parse KEYWODS record of the PDB file.<br>
+	 * A keyword may be split over two lines. whether a keyword ends by the end 
+	 * of a line or it is aplit over two lines, a <code>space</code> is added 
+	 * between the 2 lines's contents, unless the first line ends in 
+	 * a '-' character.
+	 * <pre>
+	 * Record Format
+	 * COLUMNS       DATA  TYPE     FIELD         DEFINITION 
+	 *	---------------------------------------------------------------------------------
+	 *	 1 -  6       Record name    "KEYWDS" 
+	 *	 9 - 10       Continuation   continuation  Allows concatenation of records if necessary.
+	 *	11 - 79       List           keywds        Comma-separated list of keywords relevant
+	 *	                                           to the entry.      
+	 * Example
+	 * 	         1         2         3         4         5         6         7         8
+	 *	12345678901234567890123456789012345678901234567890123456789012345678901234567890
+	 *	KEYWDS    LYASE,  TRICARBOXYLIC ACID CYCLE, MITOCHONDRION, OXIDATIVE
+	 *	KEYWDS   2 METABOLISM
+	 * </pre>
+	 * @param lines The KEWODS record lines.
+	 * @author Amr ALHOSSARY
+	 */
+	private void handlePDBKeywords(List<String> lines) {
+		StringBuilder fullList = new StringBuilder();
+		for (String line : lines) {
+			String kwList = line.substring(10).trim();
+			if(kwList.length() > 0) {
+				if(fullList.length() > 0 && fullList.indexOf("-", fullList.length()-1) < 0) {
+					fullList.append(' ');
+				}
+				fullList.append(kwList);
+			}
+		}
+		String fulllengthList = fullList.toString();
+		keywords = fulllengthList.split("( )*,( )*");
+		ArrayList<String> lst = new ArrayList<String>(keywords.length);
+		for (String keyword : keywords) {
+			if(keyword.length() == 0) {
+				logger.warn("Keyword empty in structure "+ structure.getIdentifier().toString());
+				continue;
+			}
+			lst.add(keyword);
+		}
+		pdbHeader.setKeywords(lst);
+	}
+
 	/**
 	 * Handles creation of all bonds. Looks at LINK records, SSBOND (Disulfide
 	 * bonds), peptide bonds, and intra-residue bonds.

diff --git a/...va-structure/src/main/java/org/biojava/nbio/structure/io/cif/AbstractCifFileSupplier.java b/...va-structure/src/main/java/org/biojava/nbio/structure/io/cif/AbstractCifFileSupplier.java
@@ -45,6 +45,10 @@ protected CifFile getInternal(Structure structure, List<WrappedAtom> wrappedAtom
         MmCifBlockBuilder blockBuilder = CifBuilder.enterFile(StandardSchemata.MMCIF)
                 .enterBlock(structure.getPDBCode());
 
+        blockBuilder.enterStructKeywords().enterText()
+        .add(String.join(", ", structure.getPDBHeader().getKeywords()))
+        .leaveColumn().leaveCategory();
+
         if (atomSite.isDefined() && atomSite.getRowCount() > 0) {
             // set atom site
             blockBuilder.addCategory(atomSite);

diff --git a/...a-structure/src/main/java/org/biojava/nbio/structure/io/cif/CifStructureConsumerImpl.java b/...a-structure/src/main/java/org/biojava/nbio/structure/io/cif/CifStructureConsumerImpl.java
@@ -1,5 +1,24 @@
 package org.biojava.nbio.structure.io.cif;
 
+import java.time.LocalDate;
+import java.time.ZoneId;
+import java.time.format.DateTimeFormatter;
+import java.time.format.DateTimeFormatterBuilder;
+import java.util.ArrayList;
+import java.util.Date;
+import java.util.HashMap;
+import java.util.LinkedHashMap;
+import java.util.List;
+import java.util.Locale;
+import java.util.Map;
+import java.util.NoSuchElementException;
+import java.util.Optional;
+import java.util.OptionalInt;
+import java.util.stream.Collectors;
+import java.util.stream.IntStream;
+
+import javax.vecmath.Matrix4d;
+
 import org.biojava.nbio.structure.AminoAcid;
 import org.biojava.nbio.structure.AminoAcidImpl;
 import org.biojava.nbio.structure.Atom;
@@ -88,15 +107,6 @@
 import org.slf4j.Logger;
 import org.slf4j.LoggerFactory;
 
-import javax.vecmath.Matrix4d;
-import java.time.LocalDate;
-import java.time.ZoneId;
-import java.time.format.DateTimeFormatter;
-import java.time.format.DateTimeFormatterBuilder;
-import java.util.*;
-import java.util.stream.Collectors;
-import java.util.stream.IntStream;
-
 /**
  * An implementation of a CifFileConsumer for BioJava. Will process the information provided by a CifFile instance and
  * use it to build up a {@link Structure} object.
@@ -885,12 +895,24 @@ public void consumeStructConnType(StructConnType structConnType) {
 
     @Override
     public void consumeStructKeywords(StructKeywords structKeywords) {
+        ArrayList<String> keywordsList = new ArrayList<String>();
+
+        StrColumn text = structKeywords.getText();
+        if (text.isDefined()) {
+            String keywords = text.get(0);
+            String[] strings = keywords.split(" *, *");
+            for (String string : strings) {
+                keywordsList.add(string.trim());
+            }
+        }
+        structure.getPDBHeader().setKeywords(keywordsList);
+
         StrColumn pdbxKeywords = structKeywords.getPdbxKeywords();
-        // TODO what is the correct format for these?
         if (pdbxKeywords.isDefined()) {
-            String keywords = pdbxKeywords.values().collect(Collectors.joining(", "));
-            pdbHeader.setDescription(keywords);
+            String keywords = pdbxKeywords.get(0);
             pdbHeader.setClassification(keywords);
+            //This field should be left empty. TODO The next line should be removed later
+            pdbHeader.setDescription(keywords);
         }
     }
 

diff --git a/biojava-structure/src/test/java/org/biojava/nbio/structure/TestKeywords.java b/biojava-structure/src/test/java/org/biojava/nbio/structure/TestKeywords.java
@@ -0,0 +1,41 @@
+package org.biojava.nbio.structure;
+
+import static org.junit.Assert.assertEquals;
+
+import java.io.IOException;
+import java.io.InputStream;
+import java.util.List;
+import java.util.zip.GZIPInputStream;
+
+import org.biojava.nbio.structure.io.PDBFileParser;
+import org.junit.Test;
+
+public class TestKeywords {
+
+	@Test
+	public void testKeywordsOnFiveLines () throws IOException {
+		String fileName = "/3cdl.pdb";
+		InputStream inStream = this.getClass().getResourceAsStream(fileName);
+
+		PDBFileParser pdbpars = new PDBFileParser();
+		Structure structure = pdbpars.parsePDBFile(inStream);
+		List<String> keywords = structure.getPDBHeader().getKeywords();
+		assertEquals(12, keywords.size());
+		assertEquals("TRANSCRIPTION REGULATOR", keywords.get(11));
+	}
+
+	@Test
+	public void testDash() throws IOException {
+		String fileName;
+		fileName = "/pdb6elw-26lines.ent.gz";
+		InputStream resourceAsStream = getClass().getResourceAsStream(fileName);
+		GZIPInputStream inStream = new GZIPInputStream(resourceAsStream);
+
+		Structure structure =  new PDBFileParser().parsePDBFile(inStream);
+
+		List<String> keywords = structure.getPDBHeader().getKeywords();
+		assertEquals(6, keywords.size());
+		assertEquals("THIOREDOXIN-FOLD", keywords.get(3));
+		assertEquals("ANTI-OXIDATVE DEFENSE SYSTEM", keywords.get(4));
+	}
+}
diff --git a/biojava-structure/src/test/resources/pdb6elw-26lines.ent.gz b/biojava-structure/src/test/resources/pdb6elw-26lines.ent.gz