*/\r
public class SeqsetUtils\r
{\r
+\r
/**\r
* Store essential properties of a sequence in a hashtable for later recovery\r
* Keys are Name, Start, End, SeqFeatures, PdbId\r
sqinfo.put("Name", seq.getName());\r
sqinfo.put("Start", new Integer(seq.getStart()));\r
sqinfo.put("End", new Integer(seq.getEnd()));\r
- sqinfo.put("SeqFeatures", seq.getSequenceFeatures());\r
+ sqinfo.put("SeqFeatures", (seq.getSequenceFeatures() !=null) ? seq.getSequenceFeatures() : new Vector());\r
sqinfo.put("PdbId",\r
- (seq.getPDBId() != null) ? seq.getPDBId() : new String(""));\r
-\r
+ (seq.getPDBId() != null) ? seq.getPDBId() : new Vector());\r
+ sqinfo.put("datasetSequence", (seq.getDatasetSequence() !=null) ? seq.getDatasetSequence() : new Sequence("THISISAPLACEHOLDER",""));\r
return sqinfo;\r
}\r
\r
public static boolean SeqCharacterUnhash(SequenceI sq, Hashtable sqinfo)\r
{\r
boolean namePresent = true;\r
+ if (sqinfo==null)\r
+ return false;\r
String oldname = (String) sqinfo.get("Name");\r
Integer start = (Integer) sqinfo.get("Start");\r
Integer end = (Integer) sqinfo.get("End");\r
java.util.Vector sfeatures = (java.util.Vector) sqinfo.get(\r
"SeqFeatures");\r
- String pdbid = (String) sqinfo.get("PdbId");\r
-\r
+ Vector pdbid = (Vector) sqinfo.get("PdbId");\r
+ Sequence seqds = (Sequence) sqinfo.get("datasetSequence");\r
if (oldname == null)\r
{\r
namePresent = false;\r
{\r
sq.setName(oldname);\r
}\r
-\r
- if (!pdbid.equals(""))\r
+ if (pdbid!=null && pdbid.size()>0)\r
{\r
sq.setPDBId(pdbid);\r
}\r
sq.setEnd(end.intValue());\r
}\r
\r
- if (sfeatures != null)\r
+ if ((sfeatures != null) && (sfeatures.size()>0))\r
{\r
sq.setSequenceFeatures(sfeatures);\r
}\r
+ if ((seqds!=null) && !(seqds.getName().equals("THISISAPLACEHOLDER") && seqds.getLength()==0)) {\r
+ sq.setDatasetSequence(seqds);\r
+ }\r
\r
return namePresent;\r
}\r
return new String("Sequence" + i);\r
}\r
\r
+ /**\r
+ * Generates a hash of SeqCharacterHash properties for each sequence\r
+ * in a sequence set, and optionally renames the sequences to an\r
+ * unambiguous 'safe' name.\r
+ * @param sequences SequenceI[]\r
+ * @param write_names boolean set this to rename each of the sequences to its unique_name(index) name\r
+ * @return Hashtable to be passed to @see deuniquify to recover original names (and properties) for renamed sequences\r
+ */\r
public static Hashtable uniquify(SequenceI[] sequences, boolean write_names)\r
{\r
// Generate a safely named sequence set and a hash to recover the sequence names\r
Hashtable map = new Hashtable();\r
- String[] un_names = new String[sequences.length];\r
+ //String[] un_names = new String[sequences.length];\r
\r
- if (!write_names)\r
+ for (int i = 0; i < sequences.length; i++)\r
{\r
- for (int i = 0; i < sequences.length; i++)\r
- {\r
- String safename = new String("Sequence" + i);\r
- map.put(safename, SeqCharacterHash(sequences[i]));\r
+ String safename = unique_name(i);\r
+ map.put(safename, SeqCharacterHash(sequences[i]));\r
\r
- if (write_names)\r
- {\r
- sequences[i].setName(safename);\r
- }\r
+ if (write_names)\r
+ {\r
+ sequences[i].setName(safename);\r
}\r
}\r
\r
+\r
return map;\r
}\r
-\r
+ /**\r
+ * recover unsafe sequence names and original properties for a sequence\r
+ * set using a map generated by @see uniquify(sequences,true)\r
+ * @param map Hashtable\r
+ * @param sequences SequenceI[]\r
+ * @return boolean\r
+ */\r
public static boolean deuniquify(Hashtable map, SequenceI[] sequences)\r
{\r
- // recover unsafe sequence names for a sequence set\r
- boolean allfound = true;\r
-\r
- for (int i = 0; i < sequences.length; i++)\r
- {\r
- if (map.containsKey(sequences[i].getName()))\r
- {\r
- Hashtable sqinfo = (Hashtable) map.get(sequences[i].getName());\r
- SeqCharacterUnhash(sequences[i], sqinfo);\r
- }\r
- else\r
- {\r
- allfound = false;\r
+ jalview.analysis.SequenceIdMatcher matcher = new SequenceIdMatcher(sequences);\r
+ SequenceI msq = null;\r
+ Enumeration keys = map.keys();\r
+ Vector unmatched = new Vector();\r
+ for (int i=0, j=sequences.length; i<j; i++)\r
+ unmatched.add(sequences[i]);\r
+ while (keys.hasMoreElements()) {\r
+ Object key = keys.nextElement();\r
+ if (key instanceof String) {\r
+ if ((msq = matcher.findIdMatch((String) key))!=null) {\r
+ Hashtable sqinfo = (Hashtable) map.get(key);\r
+ unmatched.remove(msq);\r
+ SeqCharacterUnhash(msq, sqinfo);\r
+ }\r
+ else\r
+ {\r
+ System.err.println("Can't find '"+((String) key)+"' in uniquified alignment");\r
+ }\r
}\r
}\r
+ if (unmatched.size()>0) {\r
+ System.err.println("Did not find matches for :");\r
+ for (Enumeration i = unmatched.elements(); i.hasMoreElements(); System.out.println(((SequenceI) i.nextElement()).getName()))\r
+ ;\r
+ return false;\r
+ }\r
\r
- return allfound;\r
+ return true;\r
+ }\r
+ /**\r
+ * returns a subset of the sequenceI seuqences,\r
+ * including only those that contain at least one residue.\r
+ * @param sequences SequenceI[]\r
+ * @return SequenceI[]\r
+ */\r
+ public static SequenceI[] getNonEmptySequenceSet(SequenceI[] sequences) {\r
+ // Identify first row of alignment with residues for prediction\r
+ boolean ungapped[] = new boolean[sequences.length];\r
+ int msflen=0;\r
+ for (int i=0,j=sequences.length; i<j;i++) {\r
+ String tempseq = jalview.analysis.AlignSeq.extractGaps(jalview.util.Comparison.GapChars, sequences[i].getSequence());\r
+ if (tempseq.length()==0)\r
+ ungapped[i]=false;\r
+ else {\r
+ ungapped[i]=true;\r
+ msflen++;\r
+ }\r
+ }\r
+ if (msflen==0)\r
+ return null; // no minimal set\r
+ // compose minimal set\r
+ SequenceI[] mset = new SequenceI[msflen];\r
+ for (int i=0,j=sequences.length,k=0; i<j;i++) {\r
+ if (ungapped[i])\r
+ mset[k++] = sequences[i];\r
+ }\r
+ ungapped = null;\r
+ return mset;\r
}\r
}\r
git://source.jalview.org / jalview.git / blobdiff