import jalview.xml.binding.sifts.Entry.Entity.Segment.ListMapRegion.MapRegion;
import jalview.xml.binding.sifts.Entry.Entity.Segment.ListResidue.Residue;
import jalview.xml.binding.sifts.Entry.Entity.Segment.ListResidue.Residue.CrossRefDb;
+import jalview.xml.binding.sifts.Entry.Entity.Segment.ListResidue.Residue.ResidueDetail;
import jalview.xml.binding.sifts.Entry.ListDB.Db;
import java.io.File;
private String segStartEnd;
- private static final String UNIPROT_COORDINATE_SYS = "UniProt";
-
- private static final String PDB_COORDINATE_SYS = "PDBresnum";
-
- private String seqCoordSys = UNIPROT_COORDINATE_SYS;
+ private CoordinateSys seqCoordSys = CoordinateSys.UNIPROT;
private static final int BUFFER_SIZE = 4096;
private final static String NEWLINE = System.lineSeparator();
+ private String curSourceDBRef;
+
+ private HashSet<String> curDBRefAccessionIdsString;
+
+ public enum CoordinateSys
+ {
+ UNIPROT("UniProt"), PDB("PDBresnum"), PDBe("PDBe");
+ private String name;
+
+ private CoordinateSys(String name)
+ {
+ this.name = name;
+ }
+
+ public String getName()
+ {
+ return name;
+ }
+ };
+
+ public enum ResidueDetailType
+ {
+ NAME_SEC_STRUCTURE("nameSecondaryStructure"), CODE_SEC_STRUCTURE(
+ "codeSecondaryStructure"), ANNOTATION("Annotation");
+ private String code;
+
+ private ResidueDetailType(String code)
+ {
+ this.code = code;
+ }
+
+ public String getCode()
+ {
+ return code;
+ }
+ };
+
/**
* Fetch SIFTs file for the given PDB Id and construct an instance of
* SiftsClient
*
* @param pdbId
+ * @throws SiftsException
*/
- public SiftsClient(PDBfile pdb)
+ public SiftsClient(PDBfile pdb) throws SiftsException
{
this.pdb = pdb;
this.pdbId = pdb.id;
- try
- {
- File siftsFile = getSiftsFile(pdbId);
- siftsEntry = parseSIFTs(siftsFile);
- } catch (Exception e)
- {
- e.printStackTrace();
- }
+ File siftsFile = getSiftsFile(pdbId);
+ siftsEntry = parseSIFTs(siftsFile);
}
/**
- * Construct an instance of SiftsClient using the supplied SIFTs file -
- * the SIFTs file should correspond to the given PDB Id
+ * Construct an instance of SiftsClient using the supplied SIFTs file - the
+ * SIFTs file should correspond to the given PDB Id
*
* @param pdbId
* @param siftsFile
+ * @throws SiftsException
+ * @throws Exception
*/
- public SiftsClient(PDBfile pdb, File siftsFile)
+ public SiftsClient(PDBfile pdb, File siftsFile) throws SiftsException
{
this.pdb = pdb;
this.pdbId = pdb.id;
- try
- {
- siftsEntry = parseSIFTs(siftsFile);
- } catch (Exception e)
- {
- e.printStackTrace();
- }
-
+ siftsEntry = parseSIFTs(siftsFile);
}
/**
* @throws Exception
* if a problem occurs while parsing the SIFTs XML
*/
- private Entry parseSIFTs(File siftFile) throws Exception
+ private Entry parseSIFTs(File siftFile) throws SiftsException
{
try
{
} catch (JAXBException e)
{
e.printStackTrace();
+ throw new SiftsException(e.getMessage());
} catch (FileNotFoundException e)
{
e.printStackTrace();
+ throw new SiftsException(e.getMessage());
} catch (XMLStreamException e)
{
e.printStackTrace();
+ throw new SiftsException(e.getMessage());
} catch (FactoryConfigurationError e)
{
e.printStackTrace();
+ throw new SiftsException(e.getMessage());
} catch (IOException e)
{
e.printStackTrace();
+ throw new SiftsException(e.getMessage());
}
- throw new Exception("Error parsing siftFile");
}
/**
*
* @param pdbId
* @return SIFTs XML file
+ * @throws SiftsException
*/
- public static File getSiftsFile(String pdbId)
+ public static File getSiftsFile(String pdbId) throws SiftsException
{
File siftsFile = new File(SIFTS_DOWNLOAD_DIR + pdbId.toLowerCase()
+ ".xml.gz");
*
* @param pdbId
* @return downloaded SIFTs XML file
+ * @throws SiftsException
*/
- public static File downloadSiftsFile(String pdbId)
+ public static File downloadSiftsFile(String pdbId) throws SiftsException
{
String siftFile = pdbId + ".xml.gz";
String siftsFileFTPURL = SIFTS_FTP_BASE_URL + siftFile;
System.out.println(">>> File downloaded : " + downloadedSiftsFile);
} catch (IOException ex)
{
- ex.printStackTrace();
+ throw new SiftsException(ex.getMessage());
}
return new File(downloadedSiftsFile);
}
continue;
}
if (isFoundInSiftsEntry(dbRef.getAccessionId())
- && (dbRef.getSource().equalsIgnoreCase("uniprot") || dbRef
- .getSource().equalsIgnoreCase("pdb")))
+ && (dbRef.getSource().equalsIgnoreCase(DBRefSource.UNIPROT) || dbRef
+ .getSource().equalsIgnoreCase(DBRefSource.PDB)))
{
return dbRef;
}
java.io.PrintStream os)
throws SiftsException
{
-
System.out.println("Generating mappings for : " + entityId);
Entity entity = null;
entity = getEntityById(entityId);
// set sequence coordinate system - default value is UniProt
if (sourceDBRef.getSource().equalsIgnoreCase(DBRefSource.PDB))
{
- seqCoordSys = PDB_COORDINATE_SYS;
+ seqCoordSys = CoordinateSys.PDB;
}
- ArrayList<String> dbRefAccessionIdsString = new ArrayList<String>();
+ HashSet<String> dbRefAccessionIdsString = new HashSet<String>();
for (DBRefEntry dbref : seq.getDBRefs())
{
- dbRefAccessionIdsString.add(dbref.getAccessionId());
+ dbRefAccessionIdsString.add(dbref.getAccessionId().toLowerCase());
}
+ dbRefAccessionIdsString.add(sourceDBRef.getAccessionId().toLowerCase());
+
+ curDBRefAccessionIdsString = dbRefAccessionIdsString;
+ curSourceDBRef = sourceDBRef.getAccessionId();
// initialise all mapping positions to unassigned
for (int residuePos[] : mapping)
residuePos[PDB_RES_POS] = UNASSIGNED;
residuePos[PDB_ATOM_POS] = UNASSIGNED;
}
-
TreeMap<Integer, String> resNumMap = new TreeMap<Integer, String>();
List<Segment> segments = entity.getSegment();
for (Segment segment : segments)
{
int currSeqIndex = UNASSIGNED;
List<CrossRefDb> cRefDbs = residue.getCrossRefDb();
+ CrossRefDb pdbRefDb = null;
for (CrossRefDb cRefDb : cRefDbs)
{
- if (cRefDb.getDbCoordSys().equalsIgnoreCase(seqCoordSys)
- && dbRefAccessionIdsString.contains(cRefDb
- .getDbAccessionId()))
+ if (cRefDb.getDbSource().equalsIgnoreCase(DBRefSource.PDB))
+ {
+ pdbRefDb = cRefDb;
+ }
+ if (cRefDb.getDbCoordSys()
+ .equalsIgnoreCase(seqCoordSys.getName())
+ && hasAccessionId(cRefDb.getDbAccessionId()))
{
String resNumIndexString = cRefDb.getDbResNum()
.equalsIgnoreCase("None") ? String.valueOf(UNASSIGNED)
: cRefDb.getDbResNum();
currSeqIndex = Integer.valueOf(resNumIndexString);
- break;
+ if (pdbRefDb != null)
+ {
+ break;// exit loop if pdb and uniprot are already found
+ }
}
}
if (currSeqIndex == UNASSIGNED)
}
if (currSeqIndex > seq.getStart() && currSeqIndex <= seq.getEnd())
{
- int resNum = Integer.valueOf(residue.getDbResNum());
- mapping[currSeqIndex][PDB_RES_POS] = Integer.valueOf(resNum);
+ int resNum;
+ try
+ {
+ resNum = (pdbRefDb == null) ? Integer.valueOf(residue
+ .getDbResNum()) : Integer.valueOf(pdbRefDb.getDbResNum());
+ } catch (NumberFormatException nfe)
+ {
+ resNum = (pdbRefDb == null) ? Integer.valueOf(residue
+ .getDbResNum()) : Integer.valueOf(pdbRefDb
+ .getDbResNum().split("[a-zA-Z]")[0]);
+ }
+ try
+ {
+ mapping[currSeqIndex][PDB_RES_POS] = Integer
+ .valueOf(resNum);
+ } catch (ArrayIndexOutOfBoundsException e)
+ {
+ // do nothing..
+ }
char resCharCode = ResidueProperties
.getSingleCharacterCode(residue.getDbResName());
resNumMap.put(currSeqIndex, String.valueOf(resCharCode));
}
padWithGaps(resNumMap);
int counter = 0;
- int seqStart = 0;
- int seqEnd = 0;
- int pdbStart = 0;
- int pdbEnd = 0;
+ int seqStart = UNASSIGNED;
+ int seqEnd = UNASSIGNED;
+ int pdbStart = UNASSIGNED;
+ int pdbEnd = UNASSIGNED;
boolean startDetected = false;
for (int[] x : mapping)
{
targetStrucSeqs.append(res);
}
- try
+ if (os != null)
+ {
+ MappingOutputPojo mop = new MappingOutputPojo();
+ mop.setSeqStart(seqStart);
+ mop.setSeqEnd(seqEnd);
+ mop.setSeqName(seq.getName());
+ mop.setSeqResidue(matchedSeq);
+
+ mop.setStrStart(pdbStart);
+ mop.setStrEnd(pdbEnd);
+ mop.setStrName(structId);
+ mop.setStrResidue(targetStrucSeqs.toString());
+
+ mop.setType("pep");
+ os.print(getMappingOutput(mop).toString());
+ }
+ return mapping;
+ }
+
+ private boolean isResidueObserved(Residue residue)
+ {
+ String annotation = getResidueAnnotaiton(residue,
+ ResidueDetailType.ANNOTATION);
+ if (annotation == null)
+ {
+ return true;
+ }
+ if (!annotation.equalsIgnoreCase("Not_Found")
+ && annotation.equalsIgnoreCase("Not_Observed"))
+ {
+ return false;
+ }
+ return true;
+ }
+
+ private String getResidueAnnotaiton(Residue residue,
+ ResidueDetailType type)
+ {
+ List<ResidueDetail> resDetails = residue.getResidueDetail();
+ for (ResidueDetail resDetail : resDetails)
{
- if (os != null)
+ if (resDetail.getProperty().equalsIgnoreCase(type.getCode()))
{
- MappingOutputPojo mop = new MappingOutputPojo();
- mop.setSeqStart(seqStart);
- mop.setSeqEnd(seqEnd);
- mop.setSeqName(seq.getName());
- mop.setSeqResidue(matchedSeq);
-
- mop.setStrStart(pdbStart);
- mop.setStrEnd(pdbEnd);
- mop.setStrName(structId);
- mop.setStrResidue(targetStrucSeqs.toString());
-
- mop.setType("pep");
- os.print(getMappingOutput(mop).toString());
+ return resDetail.getContent();
}
- } catch (Exception ex)
- {
- ex.printStackTrace();
}
- return mapping;
+ return "Not_Found";
+ }
+
+ private boolean hasAccessionId(String accession)
+ {
+ boolean isStrictMatch = true;
+ return isStrictMatch ? curSourceDBRef.equalsIgnoreCase(accession)
+ : curDBRefAccessionIdsString.contains(accession.toLowerCase());
}
@Override
*/
void padWithGaps(TreeMap<Integer, String> resNumMap)
{
+ if (resNumMap == null || resNumMap.isEmpty())
+ {
+ return;
+ }
Integer[] keys = resNumMap.keySet().toArray(new Integer[0]);
Arrays.sort(keys);
int firstIndex = keys[0];
output.append(NEWLINE).append(NEWLINE);
}
float pid = (float) matchedSeqCount / seqRes.length() * 100;
+ if (pid < 2)
+ {
+ throw new SiftsException("Low PID detected for SIFTs mapping...");
+ }
output.append("Length of alignment = " + seqRes.length())
.append(NEWLINE);
output.append(new Format("Percentage ID = %2.2f").form(pid));