X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Fanalysis%2FSeqsetUtils.java;h=5a246f8d84f83fd644cc6a03ebd1d8da250e8f01;hb=cf04d0714f5d7686f47201d42d63df80b4ebf10d;hp=cdcecc0b9fe40cf9d6c61432cce46ab41b4a8b2a;hpb=33c59ebddbc51bbd26bc4c8576ec17c1498b19e8;p=jalview.git

diff --git a/src/jalview/analysis/SeqsetUtils.java b/src/jalview/analysis/SeqsetUtils.java
index cdcecc0..5a246f8 100755
--- a/src/jalview/analysis/SeqsetUtils.java
+++ b/src/jalview/analysis/SeqsetUtils.java
@@ -1,52 +1,321 @@
+/*
+ * Jalview - A Sequence Alignment Editor and Viewer ($$Version-Rel$$)
+ * Copyright (C) $$Year-Rel$$ The Jalview Authors
+ * 
+ * This file is part of Jalview.
+ * 
+ * Jalview is free software: you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License 
+ * as published by the Free Software Foundation, either version 3
+ * of the License, or (at your option) any later version.
+ *  
+ * Jalview is distributed in the hope that it will be useful, but 
+ * WITHOUT ANY WARRANTY; without even the implied warranty 
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR 
+ * PURPOSE.  See the GNU General Public License for more details.
+ * 
+ * You should have received a copy of the GNU General Public License
+ * along with Jalview.  If not, see <http://www.gnu.org/licenses/>.
+ * The Jalview Authors are detailed in the 'AUTHORS' file.
+ */
 package jalview.analysis;
 
+import jalview.bin.Cache;
+import jalview.bin.Console;
+import jalview.datamodel.AlignmentAnnotation;
+import jalview.datamodel.HiddenMarkovModel;
+import jalview.datamodel.PDBEntry;
+import jalview.datamodel.Sequence;
+import jalview.datamodel.SequenceFeature;
 import jalview.datamodel.SequenceI;
+
+import java.util.ArrayList;
+import java.util.Enumeration;
+import java.util.HashMap;
 import java.util.Hashtable;
+import java.util.Iterator;
+import java.util.List;
+import java.util.Map;
+import java.util.Objects;
+import java.util.Optional;
+import java.util.Vector;
+import static java.lang.String.format;
 
-/**
- * <p>Title: </p>
- *
- * <p>Description: </p>
- *
- * <p>Copyright: Copyright (c) 2004</p>
- *
- * <p>Company: Dundee University</p>
- *
- * @author not attributable
- * @version 1.0
- */
 public class SeqsetUtils
 {
-  public static Hashtable uniquify(SequenceI[] sequences)
+  public static class SequenceInfo {
+    private String name;
+    private int start;
+    private int end;
+    private Optional<String> description = Optional.empty();
+    private Optional<List<SequenceFeature>> features = Optional.empty();
+    private Optional<List<PDBEntry>> pdbId = Optional.empty();
+    private Optional<SequenceI> dataset = Optional.empty();
+    private Optional<HiddenMarkovModel> hmm = Optional.empty();
+    private Optional<AlignmentAnnotation[]> searchScores = Optional.empty();
+
+    private SequenceInfo(String name, int start, int end) {
+      this.name = name;
+      this.start = start;
+      this.end = end;
+    }
+  }
+
+  /**
+   * Store essential properties of a sequence in a hashtable for later recovery
+   * Keys are Name, Start, End, SeqFeatures, PdbId, HMM
+   * 
+   * @param seq
+   *          SequenceI
+   * @return Hashtable
+   */
+  public static SequenceInfo SeqCharacterHash(SequenceI seq)
+  {
+    SequenceInfo sqinfo = new SequenceInfo(seq.getName(), seq.getStart(), seq.getEnd());
+    sqinfo.description = Optional.ofNullable(seq.getDescription());
+    sqinfo.dataset = Optional.ofNullable(seq.getDatasetSequence());
+    if (!sqinfo.dataset.isPresent())
+    {
+      ArrayList<SequenceFeature> feats = new ArrayList<>(
+          seq.getFeatures().getAllFeatures());
+      sqinfo.features = Optional.of(feats);
+      sqinfo.pdbId = Optional.of(Objects.requireNonNullElse(
+          seq.getAllPDBEntries(), new ArrayList<>()));
+    }
+    if (seq.hasHMMProfile())
+    {
+      sqinfo.hmm = Optional.of(seq.getHMM());
+    }
+    sqinfo.searchScores = Optional.ofNullable(seq.getAnnotation("Search Scores"));
+    return sqinfo;
+  }
+
+  /**
+   * Recover essential properties of a sequence from a hashtable TODO: replace
+   * these methods with something more elegant.
+   * 
+   * @param sq
+   *          SequenceI
+   * @param sqinfo
+   *          Hashtable
+   * @return boolean true if name was not updated from sqinfo Name entry
+   */
+  public static boolean SeqCharacterUnhash(SequenceI sq, SequenceInfo sqinfo)
+  {
+    if (sqinfo == null)
+    {
+      return false;
+    }
+    if (sqinfo.name != null)
+    {
+      sq.setName(sqinfo.name);
+    }
+    sq.setStart(sqinfo.start);
+    sq.setEnd(sqinfo.end);
+    if (sqinfo.pdbId.isPresent() && !sqinfo.pdbId.get().isEmpty())
+      sq.setPDBId(new Vector<>(sqinfo.pdbId.get()));
+    if (sqinfo.features.isPresent() && !sqinfo.features.get().isEmpty())
+      sq.setSequenceFeatures(sqinfo.features.get());
+    if (sqinfo.description.isPresent())
+      sq.setDescription(sqinfo.description.get());
+    if (sqinfo.dataset.isPresent())
+    {
+      if (sqinfo.features.isPresent())
+      {
+        Console.warn("Setting dataset sequence for a sequence which has " +
+            "sequence features. Dataset sequence features will not be visible.");
+        assert false;
+      }
+      sq.setDatasetSequence(sqinfo.dataset.get());
+    }
+    if (sqinfo.hmm.isPresent())
+      sq.setHMM(new HiddenMarkovModel(sqinfo.hmm.get(), sq));
+    if (sqinfo.searchScores.isPresent())
+    {
+      for (AlignmentAnnotation score : sqinfo.searchScores.get())
+      {
+        sq.addAlignmentAnnotation(score);
+      }
+    }
+    return sqinfo.name != null;
+  }
+
+  /**
+   * Form of the unique name used in uniquify for the i'th sequence in an
+   * ordered vector of sequences.
+   * 
+   * @param i
+   *          int
+   * @return String
+   */
+  public static String unique_name(int i)
+  {
+    return String.format("Sequence%d", i);
+  }
+
+  /**
+   * Generates a hash of SeqCharacterHash properties for each sequence in a
+   * sequence set, and optionally renames the sequences to an unambiguous 'safe'
+   * name.
+   * 
+   * @param sequences
+   *          SequenceI[]
+   * @param write_names
+   *          boolean set this to rename each of the sequences to its
+   *          unique_name(index) name
+   * @return Hashtable to be passed to
+   * @see deuniquify to recover original names (and properties) for renamed
+   *      sequences
+   */
+  public static Map<String, SequenceInfo> uniquify(SequenceI[] sequences,
+          boolean write_names)
   {
-    // Generate a safely named sequence set and a hash to recover the sequence names
-    Hashtable map = new Hashtable();
+    // Generate a safely named sequence set and a hash to recover the sequence
+    // names
+    HashMap<String, SequenceInfo> map = new HashMap<>();
+    // String[] un_names = new String[sequences.length];
+
     for (int i = 0; i < sequences.length; i++)
     {
-      String safename = new String("Sequence" + i);
-      map.put(safename, sequences[i].getName());
-      sequences[i].setName(safename);
+      String safename = unique_name(i);
+      map.put(safename, SeqCharacterHash(sequences[i]));
+
+      if (write_names)
+      {
+        sequences[i].setName(safename);
+      }
     }
+
     return map;
   }
 
-  public static boolean deuniquify(Hashtable map, SequenceI[] sequences)
+  /**
+   * recover unsafe sequence names and original properties for a sequence set
+   * using a map generated by
+   * 
+   * @see uniquify(sequences,true)
+   * @param map
+   *          Hashtable
+   * @param sequences
+   *          SequenceI[]
+   * @return boolean
+   */
+  public static boolean deuniquify(Map<String, SequenceInfo> map,
+      SequenceI[] sequences)
   {
-    // recover unsafe sequence names for a sequence set
-    boolean allfound = true;
-    for (int i = 0; i < sequences.length; i++)
+    return deuniquify(map, sequences, true);
+  }
+
+  /**
+   * recover unsafe sequence names and original properties for a sequence set
+   * using a map generated by
+   * 
+   * @see uniquify(sequences,true)
+   * @param map
+   *          Hashtable
+   * @param sequences
+   *          SequenceI[]
+   * @param quiet
+   *          when false, don't complain about sequences without any data in the
+   *          map.
+   * @return boolean
+   */
+  public static boolean deuniquify(Map<String, SequenceInfo> map,
+      SequenceI[] sequences, boolean quiet)
+  {
+    jalview.analysis.SequenceIdMatcher matcher = new SequenceIdMatcher(
+            sequences);
+    SequenceI msq = null;
+    Iterator<String> keys = map.keySet().iterator();
+    Vector<SequenceI> unmatched = new Vector<>();
+    for (int i = 0, j = sequences.length; i < j; i++)
+    {
+      unmatched.addElement(sequences[i]);
+    }
+    while (keys.hasNext())
     {
-      if (map.containsKey(sequences[i].getName()))
+      String key = keys.next();
+      try {
+        if ((msq = matcher.findIdMatch((String) key)) != null)
+        {
+          SequenceInfo sqinfo = map.get(key);
+          unmatched.removeElement(msq);
+          SeqCharacterUnhash(msq, sqinfo);
+        }
+        else
+        {
+          if (!quiet)
+          {
+            Console.warn(format("Can't find '%s' in uniquified alignment",
+                key));
+          }
+        }
+      } catch (ClassCastException ccastex) {
+        if (!quiet)
+        {
+          Console.error("Unexpected object in SeqSet map : "+ key.getClass());
+        }
+      }
+    }
+    if (unmatched.size() > 0 && !quiet)
+    {
+      StringBuilder sb = new StringBuilder("Did not find match for sequences: ");
+      Enumeration<SequenceI> i = unmatched.elements();
+      sb.append(i.nextElement().getName());
+      for (; i.hasMoreElements();)
       {
-        String unsafename = (String) map.get(sequences[i].getName());
-        sequences[i].setName(unsafename);
+        sb.append(", " + i.nextElement().getName());
+      }
+      Console.warn(sb.toString());
+      return false;
+    }
+
+    return true;
+  }
+
+  /**
+   * returns a subset of the sequenceI seuqences, including only those that
+   * contain at least one residue.
+   * 
+   * @param sequences
+   *          SequenceI[]
+   * @return SequenceI[]
+   */
+  public static SequenceI[] getNonEmptySequenceSet(SequenceI[] sequences)
+  {
+    // Identify first row of alignment with residues for prediction
+    boolean ungapped[] = new boolean[sequences.length];
+    int msflen = 0;
+    for (int i = 0, j = sequences.length; i < j; i++)
+    {
+      String tempseq = jalview.analysis.AlignSeq.extractGaps(
+              jalview.util.Comparison.GapChars,
+              sequences[i].getSequenceAsString());
+
+      if (tempseq.length() == 0)
+      {
+        ungapped[i] = false;
       }
       else
       {
-        allfound = false;
+        ungapped[i] = true;
+        msflen++;
       }
     }
-    return allfound;
+    if (msflen == 0)
+    {
+      return null; // no minimal set
+    }
+    // compose minimal set
+    SequenceI[] mset = new SequenceI[msflen];
+    for (int i = 0, j = sequences.length, k = 0; i < j; i++)
+    {
+      if (ungapped[i])
+      {
+        mset[k++] = sequences[i];
+      }
+    }
+    ungapped = null;
+    return mset;
   }
-
 }