Handle an empty Element column in a PDB file

lafita · lafita · commit db1839e0e2dd · 2016-07-27T11:24:13.000+02:00
Fix biojava#537
diff --git a/biojava-integrationtest/src/test/java/org/biojava/nbio/structure/test/PDBFileParserTest.java b/biojava-integrationtest/src/test/java/org/biojava/nbio/structure/test/PDBFileParserTest.java
@@ -579,6 +579,24 @@ public void testMissingElements() throws IOException {
 				"ATOM     14  SG  CYS L   1      12.247  14.885  22.538  1.00 20.55"+newline+
 				"TER                                                               "+newline;
 		
+		// A two residue structure with empty Element column
+		String emptyElement =
+				"ATOM      1  N   ASP L   1A     11.095  19.341  20.188  1.00 30.14            "+newline+
+				"ATOM      2  CA  ASP L   1A     10.070  18.634  19.379  1.00 28.34            "+newline+
+				"ATOM      3  C   ASP L   1A      9.846  17.102  19.503  1.00 26.08            "+newline+
+				"ATOM      4  O   ASP L   1A      8.744  16.584  19.162  1.00 23.47            "+newline+
+				"ATOM      5  CB  ASP L   1A     10.255  18.858  17.853  1.00 37.55            "+newline+
+				"ATOM      6  CG  ASP L   1A      8.836  19.264  17.401  1.00 42.76            "+newline+
+				"ATOM      7  OD1 ASP L   1A      8.058  19.292  18.400  1.00 44.03            "+newline+
+				"ATOM      8  OD2 ASP L   1A      8.616  19.668  16.244  1.00 46.88            "+newline+
+				"ATOM      9  N   CYS L   1      10.835  16.440  20.113  1.00 23.72            "+newline+
+				"ATOM     10  CA  CYS L   1      10.769  14.970  20.210  1.00 20.89            "+newline+
+				"ATOM     11  C   CYS L   1       9.580  14.524  21.006  1.00 18.64            "+newline+
+				"ATOM     12  O   CYS L   1       9.110  15.220  21.912  1.00 19.03            "+newline+
+				"ATOM     13  CB  CYS L   1      12.117  14.468  20.771  1.00 21.77            "+newline+
+				"ATOM     14  SG  CYS L   1      12.247  14.885  22.538  1.00 20.55            "+newline+
+				"TER                                                                             "+newline;
+		
 		String original =
 				"ATOM      1  N   ASP L   1A     11.095  19.341  20.188  1.00 30.14           N"+newline+
 				"ATOM      2  CA  ASP L   1A     10.070  18.634  19.379  1.00 28.34           C"+newline+
@@ -600,8 +618,13 @@ public void testMissingElements() throws IOException {
 		BufferedReader br = new BufferedReader(new StringReader(missingElement));
 		Structure s = parser.parsePDBFile(br);
 		String pdb = s.toPDB();
+		assertTrue("the Element column has not been filled correctly", pdb.equals(original));
 		
-		assertTrue("the created PDB file does not match the input file", pdb.equals(original));
+		
+		br = new BufferedReader(new StringReader(emptyElement));
+		s = parser.parsePDBFile(br);
+		pdb = s.toPDB();
+		assertTrue("the Element column has not been filled correctly", pdb.equals(original));
 		
 	}
 }
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
@@ -1828,17 +1828,27 @@ private void  pdb_ATOM_Handler(String line)	{
 		// missing (i.e. misformatted PDB file), then parse the
 		// element from the chemical component.
 		Element element = Element.R;
+		boolean guessElement = true;
 		if ( line.length() > 77 ) {
 			// parse element from element field
-			String elementSymbol = line.substring (76, 78).trim();
-			try {
-				element = Element.valueOfIgnoreCase(elementSymbol);
-			}  catch (IllegalArgumentException e){
-				logger.warn("Element {} was not recognised. Assigning generic element R to it", elementSymbol);
+			String elementSymbol = line.substring(76, 78).trim();
+			if (elementSymbol.equals("")) {
+				logger.warn("Element column was empty. Assigning atom element "
+						+ "from Chemical Component Dictionary information", elementSymbol);
+			} else {
+				try {
+					element = Element.valueOfIgnoreCase(elementSymbol);
+					guessElement = false;
+				}  catch (IllegalArgumentException e){
+					logger.warn("Element {} was not recognised. Assigning atom element "
+							+ "from Chemical Component Dictionary information", elementSymbol);
+				}
 			}
 		} else {
 			logger.warn("Missformatted PDB file: element column is not present. "
-					+ "Assigning atom elements from Chemical Component Dictionary information");
+					+ "Assigning atom element from Chemical Component Dictionary information");
+		}
+		if (guessElement) {
 			String elementSymbol = null;
 			if (currentGroup.getChemComp() != null) {
 				for (ChemCompAtom a : currentGroup.getChemComp().getAtoms()) {
