X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;ds=inline;f=src%2Fjalview%2Fanalysis%2FGrouping.java;h=066814e626b390897a812565b1262958f61aadc8;hb=7723baa88fa4f365c01e9f0ee9551e278219d2e5;hp=bba668db9e68a827b95320c8a99b887513ac6942;hpb=87deef0fa44f3aa7d7d86ac7241a9d8c93b2408f;p=jalview.git
diff --git a/src/jalview/analysis/Grouping.java b/src/jalview/analysis/Grouping.java
index bba668d..066814e 100644
--- a/src/jalview/analysis/Grouping.java
+++ b/src/jalview/analysis/Grouping.java
@@ -1,273 +1,294 @@
+/*
+ * 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 .
+ * The Jalview Authors are detailed in the 'AUTHORS' file.
+ */
package jalview.analysis;
-import jalview.datamodel.AlignmentI;
-import jalview.datamodel.SequenceFeature;
+import jalview.datamodel.ColumnSelection;
import jalview.datamodel.SequenceGroup;
import jalview.datamodel.SequenceI;
-import java.util.Enumeration;
-import java.util.Hashtable;
+import java.util.ArrayList;
+import java.util.HashMap;
+import java.util.List;
+import java.util.Map;
import java.util.Vector;
/**
- * various methods for defining groups on an alignment based on some other properties
+ * various methods for defining groups on an alignment based on some other
+ * properties
+ *
* @author JimP
- *
+ *
*/
public class Grouping
{
/**
- * Divide the given sequences based on the equivalence of their corresponding selectedChars string. If exgroups is provided, existing groups will be subdivided.
+ * Divide the given sequences based on the equivalence of their corresponding
+ * selectedChars string. If exgroups is provided, existing groups will be
+ * subdivided.
+ *
* @param sequences
* @param selectedChars
- * @param exgroups
+ * @param list
* @return
*/
- public static SequenceGroup[] makeGroupsFrom(SequenceI[] sequences, String[] selectedChars, Vector exgroups)
+ public static SequenceGroup[] makeGroupsFrom(SequenceI[] sequences,
+ String[] selectedChars, List list)
{
- // TODO: determine how to get/recover input data for group generation
- Hashtable gps = new Hashtable();
- int width = 0,i;
- Hashtable pgroup = new Hashtable();
- if (exgroups!=null)
+ // TODO: determine how to get/recover input data for group generation
+ Map> gps = new HashMap>();
+ int width = 0, i;
+ Map pgroup = new HashMap();
+ if (list != null)
{
- SequenceGroup sg;
- for (Enumeration g=exgroups.elements(); g.hasMoreElements(); )
+ for (SequenceGroup sg : list)
{
- sg = (SequenceGroup) g.nextElement();
- for (Enumeration sq = sg.getSequences(null).elements(); sq.hasMoreElements(); )
- pgroup.put(sq.nextElement().toString(), sg);
+ for (SequenceI sq : sg.getSequences(null))
+ {
+ pgroup.put(sq.toString(), sg);
+ }
}
}
for (i = 0; i < sequences.length; i++)
{
String schar = selectedChars[i];
- SequenceGroup pgp = (SequenceGroup) pgroup.get(((Object) sequences[i]).toString());
- if (pgp!=null)
+ SequenceGroup pgp = pgroup.get(((Object) sequences[i]).toString());
+ if (pgp != null)
{
- schar = pgp.getName()+":"+schar;
+ schar = pgp.getName() + ":" + schar;
}
- Vector svec = (Vector) gps.get(schar);
+ List svec = gps.get(schar);
if (svec == null)
{
- svec = new Vector();
+ svec = new ArrayList();
gps.put(schar, svec);
}
- if (width list)
{
- // TODO implement divideByFeature
- /*
- if (method!=AlignmentSorter.FEATURE_SCORE && method!=AlignmentSorter.FEATURE_LABEL && method!=AlignmentSorter.FEATURE_DENSITY)
- {
- throw new Error("Implementation Error - sortByFeature method must be one of FEATURE_SCORE, FEATURE_LABEL or FEATURE_DENSITY.");
- }
- boolean ignoreScore=method!=AlignmentSorter.FEATURE_SCORE;
- StringBuffer scoreLabel = new StringBuffer();
- scoreLabel.append(start+stop+method);
- // This doesn't work yet - we'd like to have a canonical ordering that can be preserved from call to call
- for (int i=0;featureLabels!=null && i> gps = new HashMap>();
+ Map pgroup = new HashMap();
+ if (list != null)
{
- SequenceFeature[] sf = seqs[i].getSequenceFeatures();
- if (sf==null && seqs[i].getDatasetSequence()!=null)
+ for (SequenceGroup sg : list)
{
- sf = seqs[i].getDatasetSequence().getSequenceFeatures();
- }
- if (sf==null)
- {
- sf = new SequenceFeature[0];
- } else {
- SequenceFeature[] tmp = new SequenceFeature[sf.length];
- for (int s=0; s sstop)
- ||
- // or ignore based on selection criteria
- (featureLabels != null && !AlignmentSorter.containsIgnoreCase(sf[f].type, featureLabels))
- || (groupLabels != null
- // problem here: we cannot eliminate null feature group features
- && (sf[f].getFeatureGroup() != null
- && !AlignmentSorter.containsIgnoreCase(sf[f].getFeatureGroup(), groupLabels))))
- {
- // forget about this feature
- sf[f] = null;
- n--;
- } else {
- // or, also take a look at the scores if necessary.
- if (!ignoreScore && sf[f].getScore()!=Float.NaN)
- {
- if (seqScores[i]==0)
- {
- hasScores++;
- }
- seqScores[i]++;
- hasScore[i] = true;
- scores[i] += sf[f].getScore(); // take the first instance of this
- // score.
- }
- }
- }
- SequenceFeature[] fs;
- feats[i] = fs = new SequenceFeature[n];
- if (n>0)
- {
- n=0;
- for (int f=0;f scores[i])
- {
- min = scores[i];
- }
+ pgroup.put(sq.toString(), sg);
}
}
}
-
- if (method==FEATURE_SCORE)
- {
- if (hasScores == 0)
+
+ /*
+ * get selected columns (in the order they were selected);
+ * note this could include right-to-left ranges
+ */
+ int[] spos = new int[cs.getSelected().size()];
+ int width = -1;
+ int i = 0;
+ for (Integer pos : cs.getSelected())
{
- return; // do nothing - no scores present to sort by.
+ spos[i++] = pos.intValue();
}
- // pad score matrix
- if (hasScores < seqs.length)
+
+ for (i = 0; i < sequences.length; i++)
{
- for (int i = 0; i < seqs.length; i++)
+ int slen = sequences[i].getLength();
+ if (width < slen)
{
- if (!hasScore[i])
- {
- scores[i] = (max + i);
- } else {
- int nf=(feats[i]==null) ? 0 :((SequenceFeature[]) feats[i]).length;
- System.err.println("Sorting on Score: seq "+seqs[i].getName()+ " Feats: "+nf+" Score : "+scores[i]);
- }
+ width = slen;
}
- }
- jalview.util.QuickSort.sort(scores, seqs);
- }
- else
- if (method==FEATURE_DENSITY)
+ SequenceGroup pgp = pgroup.get(((Object) sequences[i]).toString());
+ StringBuilder schar = new StringBuilder();
+ if (pgp != null)
+ {
+ schar.append(pgp.getName() + ":");
+ }
+ for (int p : spos)
{
-
- // break ties between equivalent numbers for adjacent sequences by adding 1/Nseq*i on the original order
- double fr = 0.9/(1.0*seqs.length);
- for (int i=0;i= slen)
{
- double nf;
- scores[i] = (0.05+fr*i)+(nf=((feats[i]==null) ? 0.0 :1.0*((SequenceFeature[]) feats[i]).length));
- System.err.println("Sorting on Density: seq "+seqs[i].getName()+ " Feats: "+nf+" Score : "+scores[i]);
+ schar.append("~");
}
- jalview.util.QuickSort.sort(scores, seqs);
- }
- else {
- if (method==FEATURE_LABEL)
+ else
{
- throw new Error("Not yet implemented.");
+ schar.append(sequences[i].getCharAt(p));
}
}
- if (lastSortByFeatureScore ==null || scoreLabel.equals(lastSortByFeatureScore))
- {
- setOrder(alignment, seqs);
+ List svec = gps.get(schar.toString());
+ if (svec == null)
+ {
+ svec = new ArrayList();
+ gps.put(schar.toString(), svec);
+ }
+ svec.add(sequences[i]);
}
- else
+ // make some groups
+ SequenceGroup[] groups = new SequenceGroup[gps.size()];
+ i = 0;
+ for (String key : gps.keySet())
{
- setReverseOrder(alignment, seqs);
+ SequenceGroup group = new SequenceGroup(gps.get(key),
+ "Subseq: " + key, null, true, true, false, 0, width - 1);
+
+ groups[i++] = group;
}
- lastSortByFeatureScore = scoreLabel.toString(); */
+ gps.clear();
+ pgroup.clear();
+ return groups;
}
+ /**
+ * subdivide the given sequences based on the distribution of features
+ *
+ * @param featureLabels
+ * - null or one or more feature types to filter on.
+ * @param groupLabels
+ * - null or set of groups to filter features on
+ * @param start
+ * - range for feature filter
+ * @param stop
+ * - range for feature filter
+ * @param sequences
+ * - sequences to be divided
+ * @param exgroups
+ * - existing groups to be subdivided
+ * @param method
+ * - density, description, score
+ */
+ public static void divideByFeature(String[] featureLabels,
+ String[] groupLabels, int start, int stop, SequenceI[] sequences,
+ Vector exgroups, String method)
+ {
+ // TODO implement divideByFeature
+ /*
+ * if (method!=AlignmentSorter.FEATURE_SCORE &&
+ * method!=AlignmentSorter.FEATURE_LABEL &&
+ * method!=AlignmentSorter.FEATURE_DENSITY) { throw newError(
+ * "Implementation Error - sortByFeature method must be one of FEATURE_SCORE, FEATURE_LABEL or FEATURE_DENSITY."
+ * ); } boolean ignoreScore=method!=AlignmentSorter.FEATURE_SCORE;
+ * StringBuffer scoreLabel = new StringBuffer();
+ * scoreLabel.append(start+stop+method); // This doesn't work yet - we'd
+ * like to have a canonical ordering that can be preserved from call to call
+ * for (int i=0;featureLabels!=null && i sstop) || // or ignore based on
+ * selection criteria (featureLabels != null &&
+ * !AlignmentSorter.containsIgnoreCase(sf[f].type, featureLabels)) ||
+ * (groupLabels != null // problem here: we cannot eliminate null feature
+ * group features && (sf[f].getFeatureGroup() != null &&
+ * !AlignmentSorter.containsIgnoreCase(sf[f].getFeatureGroup(),
+ * groupLabels)))) { // forget about this feature sf[f] = null; n--; } else
+ * { // or, also take a look at the scores if necessary. if (!ignoreScore &&
+ * sf[f].getScore()!=Float.NaN) { if (seqScores[i]==0) { hasScores++; }
+ * seqScores[i]++; hasScore[i] = true; scores[i] += sf[f].getScore(); //
+ * take the first instance of this // score. } } } SequenceFeature[] fs;
+ * feats[i] = fs = new SequenceFeature[n]; if (n>0) { n=0; for (int
+ * f=0;f scores[i]) {
+ * min = scores[i]; } } } }
+ *
+ * if (method==FEATURE_SCORE) { if (hasScores == 0) { return; // do nothing
+ * - no scores present to sort by. } // pad score matrix if (hasScores <
+ * seqs.length) { for (int i = 0; i < seqs.length; i++) { if (!hasScore[i])
+ * { scores[i] = (max + i); } else { int nf=(feats[i]==null) ? 0
+ * :((SequenceFeature[]) feats[i]).length;
+ * System.err.println("Sorting on Score: seq "+seqs[i].getName()+
+ * " Feats: "+nf+" Score : "+scores[i]); } } }
+ *
+ * jalview.util.QuickSort.sort(scores, seqs); } else if
+ * (method==FEATURE_DENSITY) {
+ *
+ * // break ties between equivalent numbers for adjacent sequences by adding
+ * 1/Nseq*i on the original order double fr = 0.9/(1.0*seqs.length); for
+ * (int i=0;i