/*
- * Jalview - A Sequence Alignment Editor and Viewer (Version 2.7)
- * Copyright (C) 2011 J Procter, AM Waterhouse, J Engelhardt, LM Lui, G Barton, M Clamp, S Searle
- *
+ * 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
+ * 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/>.
+ *
+ * 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 java.util.*;
-
-import jalview.datamodel.*;
-import jalview.util.*;
+import jalview.analysis.scoremodels.PIDModel;
+import jalview.analysis.scoremodels.SimilarityParams;
+import jalview.datamodel.AlignmentAnnotation;
+import jalview.datamodel.AlignmentI;
+import jalview.datamodel.AlignmentOrder;
+import jalview.datamodel.SequenceFeature;
+import jalview.datamodel.SequenceGroup;
+import jalview.datamodel.SequenceI;
+import jalview.datamodel.SequenceNode;
+import jalview.util.MessageManager;
+import jalview.util.QuickSort;
+
+import java.util.ArrayList;
+import java.util.Arrays;
+import java.util.List;
/**
* Routines for manipulating the order of a multiple sequence alignment TODO:
static boolean sortOrderAscending = true;
- static NJTree lastTree = null;
+ static TreeModel lastTree = null;
static boolean sortTreeAscending = true;
private static boolean sortLengthAscending;
/**
- * Sort by Percentage Identity w.r.t. s
- *
+ * Sorts sequences in the alignment by Percentage Identity with the given
+ * reference sequence, sorting the highest identity to the top
+ *
* @param align
* AlignmentI
* @param s
* SequenceI
- * @param tosort
- * sequences from align that are to be sorted.
- */
- public static void sortByPID(AlignmentI align, SequenceI s,
- SequenceI[] tosort)
- {
- sortByPID(align, s, tosort, 0, -1);
- }
-
- /**
- * Sort by Percentage Identity w.r.t. s
- *
- * @param align
- * AlignmentI
- * @param s
- * SequenceI
- * @param tosort
- * sequences from align that are to be sorted.
- * @param start
- * start column (0 for beginning
* @param end
*/
- public static void sortByPID(AlignmentI align, SequenceI s,
- SequenceI[] tosort, int start, int end)
+ public static void sortByPID(AlignmentI align, SequenceI s)
{
int nSeq = align.getHeight();
float[] scores = new float[nSeq];
SequenceI[] seqs = new SequenceI[nSeq];
+ String refSeq = s.getSequenceAsString();
+ SimilarityParams pidParams = new SimilarityParams(true, true, true,
+ true);
for (int i = 0; i < nSeq; i++)
{
- scores[i] = Comparison.PID(align.getSequenceAt(i)
- .getSequenceAsString(), s.getSequenceAsString());
+ scores[i] = (float) PIDModel.computePID(
+ align.getSequenceAt(i).getSequenceAsString(), refSeq,
+ pidParams);
seqs[i] = align.getSequenceAt(i);
}
- QuickSort.sort(scores, 0, scores.length - 1, seqs);
+ QuickSort.sort(scores, seqs);
setReverseOrder(align, seqs);
}
/**
* Reverse the order of the sort
- *
+ *
* @param align
* DOCUMENT ME!
* @param seqs
/**
* Sets the Alignment object with the given sequences
- *
+ *
* @param align
* Alignment object to be updated
* @param tmp
* sequences as a vector
*/
- private static void setOrder(AlignmentI align, Vector tmp)
+ private static void setOrder(AlignmentI align, List<SequenceI> tmp)
{
setOrder(align, vectorSubsetToArray(tmp, align.getSequences()));
}
/**
* Sets the Alignment object with the given sequences
- *
+ *
* @param align
* DOCUMENT ME!
* @param seqs
/**
* Sorts by ID. Numbers are sorted before letters.
- *
+ *
* @param align
* The alignment object to sort
*/
/**
* Sorts by sequence length
- *
+ *
* @param align
* The alignment object to sort
*/
* Sorts the alignment by size of group. <br>
* Maintains the order of sequences in each group by order in given alignment
* object.
- *
+ *
* @param align
* sorts the given alignment object by group
*/
{
// MAINTAINS ORIGNAL SEQUENCE ORDER,
// ORDERS BY GROUP SIZE
- Vector groups = new Vector();
+ List<SequenceGroup> groups = new ArrayList<SequenceGroup>();
if (groups.hashCode() != lastGroupHash)
{
{
for (int j = 0; j < groups.size(); j++)
{
- SequenceGroup sg2 = (SequenceGroup) groups.elementAt(j);
+ SequenceGroup sg2 = groups.get(j);
if (sg.getSize() > sg2.getSize())
{
- groups.insertElementAt(sg, j);
+ groups.add(j, sg);
break;
}
if (!groups.contains(sg))
{
- groups.addElement(sg);
+ groups.add(sg);
}
}
// NOW ADD SEQUENCES MAINTAINING ALIGNMENT ORDER
// /////////////////////////////////////////////
- Vector seqs = new Vector();
+ List<SequenceI> seqs = new ArrayList<SequenceI>();
for (int i = 0; i < groups.size(); i++)
{
- SequenceGroup sg = (SequenceGroup) groups.elementAt(i);
+ SequenceGroup sg = groups.get(i);
SequenceI[] orderedseqs = sg.getSequencesInOrder(align);
for (int j = 0; j < orderedseqs.length; j++)
{
- seqs.addElement(orderedseqs[j]);
+ seqs.add(orderedseqs[j]);
}
}
}
/**
- * Converts Vector to array. java 1.18 does not have Vector.toArray()
- *
- * @param tmp
- * Vector of SequenceI objects
- *
- * @return array of Sequence[]
- */
- private static SequenceI[] vectorToArray(Vector tmp)
- {
- SequenceI[] seqs = new SequenceI[tmp.size()];
-
- for (int i = 0; i < tmp.size(); i++)
- {
- seqs[i] = (SequenceI) tmp.elementAt(i);
- }
-
- return seqs;
- }
-
- /**
* Select sequences in order from tmp that is present in mask, and any
- * remaining seqeunces in mask not in tmp
- *
+ * remaining sequences in mask not in tmp
+ *
* @param tmp
* thread safe collection of sequences
* @param mask
* thread safe collection of sequences
- *
+ *
* @return intersect(tmp,mask)+intersect(complement(tmp),mask)
*/
private static SequenceI[] vectorSubsetToArray(List<SequenceI> tmp,
List<SequenceI> mask)
{
+ // or?
+ // tmp2 = tmp.retainAll(mask);
+ // return tmp2.addAll(mask.removeAll(tmp2))
+
ArrayList<SequenceI> seqs = new ArrayList<SequenceI>();
int i, idx;
boolean[] tmask = new boolean[mask.size()];
/**
* Sorts by a given AlignmentOrder object
- *
+ *
* @param align
* Alignment to order
* @param order
public static void sortBy(AlignmentI align, AlignmentOrder order)
{
// Get an ordered vector of sequences which may also be present in align
- Vector tmp = order.getOrder();
+ List<SequenceI> tmp = order.getOrder();
if (lastOrder == order)
{
}
else
{
- setReverseOrder(align, vectorSubsetToArray(tmp, align.getSequences()));
+ setReverseOrder(align,
+ vectorSubsetToArray(tmp, align.getSequences()));
}
}
/**
* DOCUMENT ME!
- *
+ *
* @param align
* alignment to order
* @param tree
* tree which has
- *
+ *
* @return DOCUMENT ME!
*/
- private static Vector getOrderByTree(AlignmentI align, NJTree tree)
+ private static List<SequenceI> getOrderByTree(AlignmentI align,
+ TreeModel tree)
{
int nSeq = align.getHeight();
- Vector tmp = new Vector();
+ List<SequenceI> tmp = new ArrayList<SequenceI>();
tmp = _sortByTree(tree.getTopNode(), tmp, align.getSequences());
if (tmp.size() != nSeq)
{
- System.err
- .println("WARNING: tmp.size()="
- + tmp.size()
- + " != nseq="
- + nSeq
- + " in getOrderByTree - tree contains sequences not in alignment");
+ System.err.println("WARNING: tmp.size()=" + tmp.size() + " != nseq="
+ + nSeq
+ + " in getOrderByTree - tree contains sequences not in alignment");
}
}
/**
* Sorts the alignment by a given tree
- *
+ *
* @param align
* alignment to order
* @param tree
* tree which has
*/
- public static void sortByTree(AlignmentI align, NJTree tree)
+ public static void sortByTree(AlignmentI align, TreeModel tree)
{
- Vector tmp = getOrderByTree(align, tree);
+ List<SequenceI> tmp = getOrderByTree(align, tree);
// tmp should properly permute align with tree.
if (lastTree != tree)
}
else
{
- setReverseOrder(align, vectorSubsetToArray(tmp, align.getSequences()));
+ setReverseOrder(align,
+ vectorSubsetToArray(tmp, align.getSequences()));
}
}
/**
* DOCUMENT ME!
- *
+ *
* @param align
* DOCUMENT ME!
- * @param seqs
+ * @param tmp
* DOCUMENT ME!
*/
- private static void addStrays(AlignmentI align, Vector seqs)
+ private static void addStrays(AlignmentI align, List<SequenceI> tmp)
{
int nSeq = align.getHeight();
for (int i = 0; i < nSeq; i++)
{
- if (!seqs.contains(align.getSequenceAt(i)))
+ if (!tmp.contains(align.getSequenceAt(i)))
{
- seqs.addElement(align.getSequenceAt(i));
+ tmp.add(align.getSequenceAt(i));
}
}
- if (nSeq != seqs.size())
+ if (nSeq != tmp.size())
{
System.err
.println("ERROR: Size still not right even after addStrays");
/**
* DOCUMENT ME!
- *
+ *
* @param node
* DOCUMENT ME!
* @param tmp
* DOCUMENT ME!
* @param seqset
* DOCUMENT ME!
- *
+ *
* @return DOCUMENT ME!
*/
- private static Vector _sortByTree(SequenceNode node, Vector tmp,
- List<SequenceI> seqset)
+ private static List<SequenceI> _sortByTree(SequenceNode node,
+ List<SequenceI> tmp, List<SequenceI> seqset)
{
if (node == null)
{
// seqset.size()==0 ||
// seqset.contains(tmp)))
{
- tmp.addElement(node.element());
+ tmp.add((SequenceI) node.element());
}
}
}
/**
* Sort sequence in order of increasing score attribute for annotation with a
* particular scoreLabel. Or reverse if same label was used previously
- *
+ *
* @param scoreLabel
* exact label for sequence associated AlignmentAnnotation scores to
* use for sorting.
/**
* sort the alignment using the features on each sequence found between start
* and stop with the given featureLabel (and optional group qualifier)
- *
+ *
* @param featureLabel
* (may not be null)
* @param groupLabel
public static void sortByFeature(String featureLabel, String groupLabel,
int start, int stop, AlignmentI alignment, String method)
{
- sortByFeature(featureLabel == null ? null : new String[]
- { featureLabel }, groupLabel == null ? null : new String[]
- { groupLabel }, start, stop, alignment, method);
+ sortByFeature(
+ featureLabel == null ? null : Arrays.asList(new String[]
+ { featureLabel }),
+ groupLabel == null ? null : Arrays.asList(new String[]
+ { groupLabel }), start, stop, alignment, method);
}
private static boolean containsIgnoreCase(final String lab,
- final String[] labs)
+ final List<String> labs)
{
if (labs == null)
{
{
return false;
}
- for (int q = 0; q < labs.length; q++)
+ for (String label : labs)
{
- if (labs[q] != null && lab.equalsIgnoreCase(labs[q]))
+ if (lab.equalsIgnoreCase(label))
{
return true;
}
return false;
}
- public static void sortByFeature(String[] featureLabels,
- String[] groupLabels, int start, int stop, AlignmentI alignment,
- String method)
+ public static void sortByFeature(List<String> featureLabels,
+ List<String> groupLabels, int start, int stop,
+ AlignmentI alignment, String method)
{
if (method != FEATURE_SCORE && method != FEATURE_LABEL
&& method != FEATURE_DENSITY)
{
- throw new Error(
- "Implementation Error - sortByFeature method must be one of FEATURE_SCORE, FEATURE_LABEL or FEATURE_DENSITY.");
+ throw new Error(MessageManager
+ .getString("error.implementation_error_sortbyfeature"));
}
+
boolean ignoreScore = method != FEATURE_SCORE;
StringBuffer scoreLabel = new StringBuffer();
scoreLabel.append(start + stop + method);
// This doesn't quite 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 < featureLabels.length; i++)
+ if (featureLabels != null)
{
- scoreLabel.append(featureLabels[i] == null ? "null"
- : featureLabels[i]);
+ for (String label : featureLabels)
+ {
+ scoreLabel.append(label);
+ }
}
- for (int i = 0; groupLabels != null && i < groupLabels.length; i++)
+ if (groupLabels != null)
{
- scoreLabel.append(groupLabels[i] == null ? "null" : groupLabels[i]);
+ for (String label : groupLabels)
+ {
+ scoreLabel.append(label);
+ }
}
+
+ /*
+ * if resorting the same feature, toggle sort order
+ */
+ if (lastSortByFeatureScore == null
+ || !scoreLabel.toString().equals(lastSortByFeatureScore))
+ {
+ sortByFeatureScoreAscending = true;
+ }
+ else
+ {
+ sortByFeatureScoreAscending = !sortByFeatureScoreAscending;
+ }
+ lastSortByFeatureScore = scoreLabel.toString();
+
SequenceI[] seqs = alignment.getSequencesArray();
boolean[] hasScore = new boolean[seqs.length]; // per sequence score
for (int i = 0; i < seqs.length; i++)
{
SequenceFeature[] sf = seqs[i].getSequenceFeatures();
- if (sf == null && seqs[i].getDatasetSequence() != null)
- {
- sf = seqs[i].getDatasetSequence().getSequenceFeatures();
- }
if (sf == null)
{
sf = new SequenceFeature[0];
for (int f = 0; f < sf.length; f++)
{
// filter for selection criteria
- if (
- // ignore features outwith alignment start-stop positions.
- (sf[f].end < sstart || sf[f].begin > 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))))
+ SequenceFeature feature = sf[f];
+
+ /*
+ * double-check feature overlaps columns (JAL-2544)
+ * (could avoid this with a findPositions(fromCol, toCol) method)
+ * findIndex returns base 1 column values, startCol/endCol are base 0
+ */
+ boolean noOverlap = seqs[i].findIndex(feature.getBegin()) > stop + 1
+ || seqs[i].findIndex(feature.getEnd()) < start + 1;
+ boolean skipFeatureType = featureLabels != null && !AlignmentSorter
+ .containsIgnoreCase(feature.type, featureLabels);
+ boolean skipFeatureGroup = groupLabels != null
+ && (feature.getFeatureGroup() != null
+ && !AlignmentSorter.containsIgnoreCase(
+ feature.getFeatureGroup(), groupLabels));
+ if (noOverlap || skipFeatureType || skipFeatureGroup)
{
// forget about this feature
sf[f] = null;
else
{
// or, also take a look at the scores if necessary.
- if (!ignoreScore && sf[f].getScore() != Float.NaN)
+ if (!ignoreScore && !Float.isNaN(feature.getScore()))
{
if (seqScores[i] == 0)
{
}
seqScores[i]++;
hasScore[i] = true;
- scores[i] += sf[f].getScore(); // take the first instance of this
+ scores[i] += feature.getScore(); // take the first instance of this
// score.
}
}
String[] labs = new String[fs.length];
for (int l = 0; l < labs.length; l++)
{
- labs[l] = (fs[l].getDescription() != null ? fs[l]
- .getDescription() : fs[l].getType());
+ labs[l] = (fs[l].getDescription() != null
+ ? fs[l].getDescription()
+ : fs[l].getType());
}
- jalview.util.QuickSort.sort(labs, ((Object[]) feats[i]));
+ QuickSort.sort(labs, ((Object[]) feats[i]));
}
}
if (hasScore[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]);
+ // 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);
+ QuickSort.sortByDouble(scores, seqs, sortByFeatureScoreAscending);
}
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 < seqs.length; i++)
{
- double nf;
- scores[i] = (0.05 + fr * i)
- + (nf = ((feats[i] == null) ? 0.0
- : 1.0 * ((SequenceFeature[]) feats[i]).length));
+ int featureCount = feats[i] == null ? 0
+ : ((SequenceFeature[]) feats[i]).length;
+ scores[i] = featureCount;
// System.err.println("Sorting on Density: seq "+seqs[i].getName()+
- // " Feats: "+nf+" Score : "+scores[i]);
+ // " Feats: "+featureCount+" Score : "+scores[i]);
}
- jalview.util.QuickSort.sort(scores, seqs);
+ QuickSort.sortByDouble(scores, seqs, sortByFeatureScoreAscending);
}
else
{
if (method == FEATURE_LABEL)
{
- throw new Error("Not yet implemented.");
+ throw new Error(
+ MessageManager.getString("error.not_yet_implemented"));
}
}
- if (lastSortByFeatureScore == null
- || !scoreLabel.toString().equals(lastSortByFeatureScore))
- {
- sortByFeatureScoreAscending = true;
- }
- else
- {
- sortByFeatureScoreAscending = !sortByFeatureScoreAscending;
- }
- if (sortByFeatureScoreAscending)
- {
- setOrder(alignment, seqs);
- }
- else
- {
- setReverseOrder(alignment, seqs);
- }
- lastSortByFeatureScore = scoreLabel.toString();
+
+ setOrder(alignment, seqs);
}
}