import jalview.io.StructureFile;
import jalview.schemes.ResidueProperties;
import jalview.structure.StructureMapping;
+import jalview.util.Comparison;
+import jalview.util.DBRefUtils;
import jalview.util.Format;
import jalview.xml.binding.sifts.Entry;
import jalview.xml.binding.sifts.Entry.Entity;
public class SiftsClient implements SiftsClientI
{
+ /*
+ * for use in mocking out file fetch for tests only
+ * - reset to null after testing!
+ */
+ private static File mockSiftsFile;
+
private Entry siftsEntry;
private StructureFile pdb;
private String structId;
- // private String segStartEnd;
-
private CoordinateSys seqCoordSys = CoordinateSys.UNIPROT;
private static final int BUFFER_SIZE = 4096;
private HashSet<String> curDBRefAccessionIdsString;
- public enum CoordinateSys
+ private enum CoordinateSys
{
UNIPROT("UniProt"), PDB("PDBresnum"), PDBe("PDBe");
private String name;
}
};
- public enum ResidueDetailType
+ private enum ResidueDetailType
{
NAME_SEC_STRUCTURE("nameSecondaryStructure"), CODE_SEC_STRUCTURE(
"codeSecondaryStructure"), ANNOTATION("Annotation");
siftsEntry = parseSIFTs(siftsFile);
}
-
/**
* Parse the given SIFTs File and return a JAXB POJO of parsed data
*
*/
public static File getSiftsFile(String pdbId) throws SiftsException
{
+ /*
+ * return mocked file if it has been set
+ */
+ if (mockSiftsFile != null)
+ {
+ return mockSiftsFile;
+ }
+
String siftsFileName = SiftsSettings.getSiftDownloadDirectory()
+ pdbId.toLowerCase() + ".xml.gz";
File siftsFile = new File(siftsFileName);
return new File(siftsFileName);
}
}
+ else
+ {
+ return siftsFile;
+ }
}
try
{
{
siftsDownloadDir.mkdirs();
}
- // System.out.println(">> Download ftp url : " + siftsFileFTPURL);
- URL url = new URL(siftsFileFTPURL);
- URLConnection conn = url.openConnection();
- InputStream inputStream = conn.getInputStream();
- FileOutputStream outputStream = new FileOutputStream(
- downloadedSiftsFile);
- byte[] buffer = new byte[BUFFER_SIZE];
- int bytesRead = -1;
- while ((bytesRead = inputStream.read(buffer)) != -1)
- {
- outputStream.write(buffer, 0, bytesRead);
- }
- outputStream.close();
- inputStream.close();
- // System.out.println(">>> File downloaded : " + downloadedSiftsFile);
+ // System.out.println(">> Download ftp url : " + siftsFileFTPURL);
+ // long now = System.currentTimeMillis();
+ URL url = new URL(siftsFileFTPURL);
+ URLConnection conn = url.openConnection();
+ InputStream inputStream = conn.getInputStream();
+ FileOutputStream outputStream = new FileOutputStream(
+ downloadedSiftsFile);
+ byte[] buffer = new byte[BUFFER_SIZE];
+ int bytesRead = -1;
+ while ((bytesRead = inputStream.read(buffer)) != -1)
+ {
+ outputStream.write(buffer, 0, bytesRead);
+ }
+ outputStream.close();
+ inputStream.close();
+// System.out.println(">>> File downloaded : " + downloadedSiftsFile
+// + " took " + (System.currentTimeMillis() - now) + "ms");
return new File(downloadedSiftsFile);
}
public DBRefEntryI getValidSourceDBRef(SequenceI seq)
throws SiftsException
{
- DBRefEntryI sourceDBRef = null;
- sourceDBRef = seq.getSourceDBRef();
- if (sourceDBRef != null && isValidDBRefEntry(sourceDBRef))
+ List<DBRefEntry> dbRefs = seq.getPrimaryDBRefs();
+ if (dbRefs == null || dbRefs.size() < 1)
{
- return sourceDBRef;
+ throw new SiftsException(
+ "Source DBRef could not be determined. DBRefs might not have been retrieved.");
}
- else
+
+ for (DBRefEntry dbRef : dbRefs)
{
- DBRefEntry[] dbRefs = seq.getDBRefs();
- if (dbRefs == null || dbRefs.length < 1)
+ if (dbRef == null || dbRef.getAccessionId() == null
+ || dbRef.getSource() == null)
{
- throw new SiftsException(
- "Source DBRef could not be determined. DBRefs might not have been retrieved.");
+ continue;
}
-
- for (DBRefEntryI dbRef : dbRefs)
+ String canonicalSource = DBRefUtils.getCanonicalName(dbRef
+ .getSource());
+ if (isValidDBRefEntry(dbRef)
+ && (canonicalSource.equalsIgnoreCase(DBRefSource.UNIPROT) || canonicalSource
+ .equalsIgnoreCase(DBRefSource.PDB)))
{
- if (dbRef == null || dbRef.getAccessionId() == null
- || dbRef.getSource() == null)
- {
- continue;
- }
- if (isFoundInSiftsEntry(dbRef.getAccessionId())
- && (dbRef.getSource().equalsIgnoreCase(DBRefSource.UNIPROT) || dbRef
- .getSource().equalsIgnoreCase(DBRefSource.PDB)))
- {
- seq.setSourceDBRef(dbRef);
- return dbRef;
- }
+ return dbRef;
}
}
- if (sourceDBRef != null && isValidDBRefEntry(sourceDBRef))
- {
- return sourceDBRef;
- }
throw new SiftsException("Could not get source DB Ref");
}
String pdbFile, String chain) throws SiftsException
{
structId = (chain == null) ? pdbId : pdbId + "|" + chain;
- System.out.println("Getting mapping for: " + pdbId + "|" + chain
- + " : seq- " + seq.getName());
+ System.out.println("Getting SIFTS mapping for " + structId + ": seq "
+ + seq.getName());
final StringBuilder mappingDetails = new StringBuilder(128);
PrintStream ps = new PrintStream(System.out)
String originalSeq = AlignSeq.extractGaps(
jalview.util.Comparison.GapChars, seq.getSequenceAsString());
HashMap<Integer, int[]> mapping = new HashMap<Integer, int[]>();
- DBRefEntryI sourceDBRef = seq.getSourceDBRef();
+ DBRefEntryI sourceDBRef;
sourceDBRef = getValidSourceDBRef(seq);
// TODO ensure sequence start/end is in the same coordinate system and
// consistent with the choosen sourceDBRef
TreeMap<Integer, String> resNumMap = new TreeMap<Integer, String>();
List<Segment> segments = entity.getSegment();
+ SegmentHelperPojo shp = new SegmentHelperPojo(seq, mapping, resNumMap,
+ omitNonObserved, nonObservedShiftIndex);
+ processSegments(segments, shp);
+ try
+ {
+ populateAtomPositions(entityId, mapping);
+ } catch (Exception e)
+ {
+ e.printStackTrace();
+ }
+ if (seqCoordSys == CoordinateSys.UNIPROT)
+ {
+ padWithGaps(resNumMap, omitNonObserved);
+ }
+ int seqStart = UNASSIGNED;
+ int seqEnd = UNASSIGNED;
+ int pdbStart = UNASSIGNED;
+ int pdbEnd = UNASSIGNED;
+
+ if (mapping.isEmpty())
+ {
+ throw new SiftsException("SIFTS mapping failed");
+ }
+
+ Integer[] keys = mapping.keySet().toArray(new Integer[0]);
+ Arrays.sort(keys);
+ seqStart = keys[0];
+ seqEnd = keys[keys.length - 1];
+
+ String matchedSeq = originalSeq;
+ if (seqStart != UNASSIGNED)
+ {
+ pdbStart = mapping.get(seqStart)[PDB_RES_POS];
+ pdbEnd = mapping.get(seqEnd)[PDB_RES_POS];
+ int orignalSeqStart = seq.getStart();
+ if (orignalSeqStart >= 1)
+ {
+ int subSeqStart = (seqStart >= orignalSeqStart) ? seqStart
+ - orignalSeqStart : 0;
+ int subSeqEnd = seqEnd - (orignalSeqStart - 1);
+ subSeqEnd = originalSeq.length() < subSeqEnd ? originalSeq.length()
+ : subSeqEnd;
+ matchedSeq = originalSeq.substring(subSeqStart, subSeqEnd);
+ }
+ else
+ {
+ matchedSeq = originalSeq.substring(1, originalSeq.length());
+ }
+ }
+
+ StringBuilder targetStrucSeqs = new StringBuilder();
+ for (String res : resNumMap.values())
+ {
+ targetStrucSeqs.append(res);
+ }
+
+ 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());
+ os.println();
+ }
+ return mapping;
+ }
+
+ void processSegments(List<Segment> segments, SegmentHelperPojo shp)
+ {
+ SequenceI seq = shp.getSeq();
+ HashMap<Integer, int[]> mapping = shp.getMapping();
+ TreeMap<Integer, String> resNumMap = shp.getResNumMap();
+ List<Integer> omitNonObserved = shp.getOmitNonObserved();
+ int nonObservedShiftIndex = shp.getNonObservedShiftIndex();
for (Segment segment : segments)
{
- // segStartEnd = segment.getStart() + " - " + segment.getEnd();
- // System.out.println("Mapping segments : " + segment.getSegId() + "\\"
+ // System.out.println("Mapping segments : " + segment.getSegId() + "\\"s
// + segStartEnd);
List<Residue> residues = segment.getListResidue().getResidue();
for (Residue residue : residues)
}
}
}
- try
- {
- populateAtomPositions(entityId, mapping);
- } catch (Exception e)
- {
- e.printStackTrace();
- }
- if (seqCoordSys == CoordinateSys.UNIPROT)
- {
- padWithGaps(resNumMap, omitNonObserved);
- }
- int seqStart = UNASSIGNED;
- int seqEnd = UNASSIGNED;
- int pdbStart = UNASSIGNED;
- int pdbEnd = UNASSIGNED;
-
- Integer[] keys = mapping.keySet().toArray(new Integer[0]);
- Arrays.sort(keys);
- if (keys.length < 1)
- {
- throw new SiftsException(">>> Empty SIFTS mapping generated!!");
- }
- seqStart = keys[0];
- seqEnd = keys[keys.length - 1];
-
- String matchedSeq = originalSeq;
- if (seqStart != UNASSIGNED)
- {
- pdbStart = mapping.get(seqStart)[PDB_RES_POS];
- pdbEnd = mapping.get(seqEnd)[PDB_RES_POS];
- int orignalSeqStart = seq.getStart();
- if (orignalSeqStart >= 1)
- {
- int subSeqStart = (seqStart >= orignalSeqStart) ? seqStart
- - orignalSeqStart : 0;
- int subSeqEnd = seqEnd - (orignalSeqStart - 1);
- subSeqEnd = originalSeq.length() < subSeqEnd ? originalSeq.length()
- : subSeqEnd;
- matchedSeq = originalSeq.substring(subSeqStart, subSeqEnd);
- }
- else
- {
- matchedSeq = originalSeq.substring(1, originalSeq.length());
- }
- }
-
- StringBuilder targetStrucSeqs = new StringBuilder();
- for (String res : resNumMap.values())
- {
- targetStrucSeqs.append(res);
- }
-
- if (os != null)
- {
- MappingOutputPojo mop = new MappingOutputPojo();
- mop.setSeqStart(pdbStart);
- mop.setSeqEnd(pdbEnd);
- mop.setSeqName(seq.getName());
- mop.setSeqResidue(matchedSeq);
-
- mop.setStrStart(seqStart);
- mop.setStrEnd(seqEnd);
- mop.setStrName(structId);
- mop.setStrResidue(targetStrucSeqs.toString());
-
- mop.setType("pep");
- os.print(getMappingOutput(mop).toString());
- os.println();
- }
- return mapping;
}
/**
return;
}
Integer[] keys = resNumMap.keySet().toArray(new Integer[0]);
- Arrays.sort(keys);
+ // Arrays.sort(keys);
int firstIndex = keys[0];
int lastIndex = keys[keys.length - 1];
// System.out.println("Min value " + firstIndex);
}
}
-
-
@Override
public Entity getEntityById(String id) throws SiftsException
{
- // Sometimes SIFTS mappings are wrongly swapped between different chains of
- // a PDB entry. This results to wrong mappings being generated. The boolean
- // flag 'isGetEntityIdDirectly, determines whether an entity to process is
- // determined by a greedy heuristic search or by just matching the Chain Id
- // directly against the entity Id tag. Setting the default value to 'false'
- // utilise the heuristic search which always produces correct mappings but
- // less optimised processing, where as changing the value to 'true'
- // optimises performance but might result to incorrect mapping in some cases
- // where SIFTS mappings are wrongly swapped between different chains.
- // boolean isGetEntityIdDirectly = false;
- // if (isGetEntityIdDirectly)
- // {
- // List<Entity> entities = siftsEntry.getEntity();
- // for (Entity entity : entities)
- // {
- // if (!entity.getEntityId().equalsIgnoreCase(id))
- // {
- // continue;
- // }
- // return entity;
- // }
- // }
+ // Determines an entity to process by performing a heuristic matching of all
+ // Entities with the given chainId and choosing the best matching Entity
Entity entity = getEntityByMostOptimalMatchedId(id);
if (entity != null)
{
}
}
}
- sPojo[count].pid = 100 * (sPojo[count].chainIdFreq / sPojo[count].resCount);
+ sPojo[count].pid = (100 * sPojo[count].chainIdFreq)
+ / sPojo[count].resCount;
++count;
}
Arrays.sort(sPojo, Collections.reverseOrder());
if (sPojo[0].entityId != null)
{
+ if (sPojo[0].pid < 1)
+ {
+ return null;
+ }
for (Entity entity : entities)
{
if (!entity.getEntityId().equalsIgnoreCase(sPojo[0].entityId))
return null;
}
- public class SiftsEntitySortPojo implements
+ private class SiftsEntitySortPojo implements
Comparable<SiftsEntitySortPojo>
{
public String entityId;
}
}
+ private class SegmentHelperPojo
+ {
+ private SequenceI seq;
+
+ private HashMap<Integer, int[]> mapping;
+
+ private TreeMap<Integer, String> resNumMap;
+
+ private List<Integer> omitNonObserved;
+
+ private int nonObservedShiftIndex;
+
+ public SegmentHelperPojo(SequenceI seq,
+ HashMap<Integer, int[]> mapping,
+ TreeMap<Integer, String> resNumMap,
+ List<Integer> omitNonObserved, int nonObservedShiftIndex)
+ {
+ setSeq(seq);
+ setMapping(mapping);
+ setResNumMap(resNumMap);
+ setOmitNonObserved(omitNonObserved);
+ setNonObservedShiftIndex(nonObservedShiftIndex);
+ }
+
+ public SequenceI getSeq()
+ {
+ return seq;
+ }
+
+ public void setSeq(SequenceI seq)
+ {
+ this.seq = seq;
+ }
+
+ public HashMap<Integer, int[]> getMapping()
+ {
+ return mapping;
+ }
+
+ public void setMapping(HashMap<Integer, int[]> mapping)
+ {
+ this.mapping = mapping;
+ }
+
+ public TreeMap<Integer, String> getResNumMap()
+ {
+ return resNumMap;
+ }
+
+ public void setResNumMap(TreeMap<Integer, String> resNumMap)
+ {
+ this.resNumMap = resNumMap;
+ }
+
+ public List<Integer> getOmitNonObserved()
+ {
+ return omitNonObserved;
+ }
+
+ public void setOmitNonObserved(List<Integer> omitNonObserved)
+ {
+ this.omitNonObserved = omitNonObserved;
+ }
+
+ public int getNonObservedShiftIndex()
+ {
+ return nonObservedShiftIndex;
+ }
+
+ public void setNonObservedShiftIndex(int nonObservedShiftIndex)
+ {
+ this.nonObservedShiftIndex = nonObservedShiftIndex;
+ }
+ }
@Override
public StringBuffer getMappingOutput(MappingOutputPojo mp)
{
if ((i + (j * len)) < seqRes.length())
{
- if (seqRes.charAt(i + (j * len)) == strRes
- .charAt(i + (j * len))
+ boolean sameChar = Comparison.isSameResidue(
+ seqRes.charAt(i + (j * len)),
+ strRes.charAt(i + (j * len)), false);
+ if (sameChar
&& !jalview.util.Comparison.isGap(seqRes.charAt(i
+ (j * len))))
{
return siftsEntry.getDbVersion();
}
+ public static void setMockSiftsFile(File file)
+ {
+ mockSiftsFile = file;
+ }
+
}