Coverage Report
File FeatureDistanceModel.java
Coverage histogram
12% of files have more coverage
Classes
| Class | Line # | Actions | |||
|---|---|---|---|---|---|
| FeatureDistanceModel | 40 | 55 | 25 |
Contributing tests
This file is covered by 7 tests.
.
Source view
| 1 | /* | |
| 2 | * Jalview - A Sequence Alignment Editor and Viewer ($$Version-Rel$$) | |
| 3 | * Copyright (C) $$Year-Rel$$ The Jalview Authors | |
| 4 | * | |
| 5 | * This file is part of Jalview. | |
| 6 | * | |
| 7 | * Jalview is free software: you can redistribute it and/or | |
| 8 | * modify it under the terms of the GNU General Public License | |
| 9 | * as published by the Free Software Foundation, either version 3 | |
| 10 | * of the License, or (at your option) any later version. | |
| 11 | * | |
| 12 | * Jalview is distributed in the hope that it will be useful, but | |
| 13 | * WITHOUT ANY WARRANTY; without even the implied warranty | |
| 14 | * of MERCHANTABILITY or FITNESS FOR A PARTICULAR | |
| 15 | * PURPOSE. See the GNU General Public License for more details. | |
| 16 | * | |
| 17 | * You should have received a copy of the GNU General Public License | |
| 18 | * along with Jalview. If not, see <http://www.gnu.org/licenses/>. | |
| 19 | * The Jalview Authors are detailed in the 'AUTHORS' file. | |
| 20 | */ | |
| 21 | package jalview.analysis.scoremodels; | |
| 22 | ||
| 23 | import jalview.api.AlignmentViewPanel; | |
| 24 | import jalview.api.FeatureRenderer; | |
| 25 | import jalview.api.analysis.ScoreModelI; | |
| 26 | import jalview.api.analysis.SimilarityParamsI; | |
| 27 | import jalview.datamodel.AlignmentView; | |
| 28 | import jalview.datamodel.SeqCigar; | |
| 29 | import jalview.datamodel.SequenceFeature; | |
| 30 | import jalview.math.Matrix; | |
| 31 | import jalview.math.MatrixI; | |
| 32 | import jalview.util.SetUtils; | |
| 33 | ||
| 34 | import java.util.HashMap; | |
| 35 | import java.util.HashSet; | |
| 36 | import java.util.List; | |
| 37 | import java.util.Map; | |
| 38 | import java.util.Set; | |
| 39 | ||
| 40 | public class FeatureDistanceModel extends DistanceScoreModel | |
| 41 | { | |
| 42 | private static final String NAME = "Sequence Feature Similarity"; | |
| 43 | ||
| 44 | private String description; | |
| 45 | ||
| 46 | FeatureRenderer fr; | |
| 47 | ||
| 48 | /** | |
| 49 | * Constructor | |
| 50 | */ | |
| 51 | 48 |  public FeatureDistanceModel() |
| 52 | { | |
| 53 | } | |
| 54 | ||
| 55 | 5 | @Override |
| 56 | public ScoreModelI getInstance(AlignmentViewPanel view) | |
| 57 | { | |
| 58 | 5 | FeatureDistanceModel instance; |
| 59 | 5 | try |
| 60 | { | |
| 61 | 5 | instance = this.getClass().getDeclaredConstructor().newInstance(); |
| 62 | 5 | instance.configureFromAlignmentView(view); |
| 63 | 5 | return instance; |
| 64 | } catch (InstantiationException | IllegalAccessException e) | |
| 65 | { | |
| 66 | 0 | System.err.println("Error in " + getClass().getName() |
| 67 | + ".getInstance(): " + e.getMessage()); | |
| 68 | 0 | return null; |
| 69 | } catch (ReflectiveOperationException roe) | |
| 70 | { | |
| 71 | 0 | return null; |
| 72 | } | |
| 73 | } | |
| 74 | ||
| 75 | 5 | boolean configureFromAlignmentView(AlignmentViewPanel view) |
| 76 | ||
| 77 | { | |
| 78 | 5 | fr = view.cloneFeatureRenderer(); |
| 79 | 5 | return true; |
| 80 | } | |
| 81 | ||
| 82 | /** | |
| 83 | * Calculates a distance measure [i][j] between each pair of sequences as the | |
| 84 | * average number of features they have but do not share. That is, find the | |
| 85 | * features each sequence pair has at each column, ignore feature types they | |
| 86 | * have in common, and count the rest. The totals are normalised by the number | |
| 87 | * of columns processed. | |
| 88 | * <p> | |
| 89 | * The parameters argument provides settings for treatment of gap-residue | |
| 90 | * aligned positions, and whether the score is over the longer or shorter of | |
| 91 | * each pair of sequences | |
| 92 | * | |
| 93 | * @param seqData | |
| 94 | * @param params | |
| 95 | */ | |
| 96 | 6 | @Override |
| 97 | public MatrixI findDistances(AlignmentView seqData, | |
| 98 | SimilarityParamsI params) | |
| 99 | { | |
| 100 | 6 | SeqCigar[] seqs = seqData.getSequences(); |
| 101 | 6 | int noseqs = seqs.length; |
| 102 | 6 | int cpwidth = 0;// = seqData.getWidth(); |
| 103 | 6 | double[][] distances = new double[noseqs][noseqs]; |
| 104 | 6 | List<String> dft = null; |
| 105 | 6 | if (fr != null) |
| 106 | { | |
| 107 | 6 | dft = fr.getDisplayedFeatureTypes(); |
| 108 | } | |
| 109 | 6 | if (dft == null || dft.isEmpty()) |
| 110 | { | |
| 111 | 0 | return new Matrix(distances); |
| 112 | } | |
| 113 | ||
| 114 | // need to get real position for view position | |
| 115 | 6 | int[] viscont = seqData.getVisibleContigs(); |
| 116 | ||
| 117 | /* | |
| 118 | * scan each column, compute and add to each distance[i, j] | |
| 119 | * the number of feature types that seqi and seqj do not share | |
| 120 | */ | |
| 121 | 12 | for (int vc = 0; vc < viscont.length; vc += 2) |
| 122 | { | |
| 123 | 32 | for (int cpos = viscont[vc]; cpos <= viscont[vc + 1]; cpos++) |
| 124 | { | |
| 125 | 26 | cpwidth++; |
| 126 | ||
| 127 | /* | |
| 128 | * first record feature types in this column for each sequence | |
| 129 | */ | |
| 130 | 26 | Map<SeqCigar, Set<String>> sfap = findFeatureTypesAtColumn(seqs, |
| 131 | cpos); | |
| 132 | ||
| 133 | /* | |
| 134 | * count feature types on either i'th or j'th sequence but not both | |
| 135 | * and add this 'distance' measure to the total for [i, j] for j > i | |
| 136 | */ | |
| 137 | 70 | for (int i = 0; i < (noseqs - 1); i++) |
| 138 | { | |
| 139 | 115 | for (int j = i + 1; j < noseqs; j++) |
| 140 | { | |
| 141 | 71 | SeqCigar sc1 = seqs[i]; |
| 142 | 71 | SeqCigar sc2 = seqs[j]; |
| 143 | 71 | Set<String> set1 = sfap.get(sc1); |
| 144 | 71 | Set<String> set2 = sfap.get(sc2); |
| 145 | 71 | boolean gap1 = set1 == null; |
| 146 | 71 | boolean gap2 = set2 == null; |
| 147 | ||
| 148 | /* | |
| 149 | * gap-gap always scores zero | |
| 150 | * residue-residue is always scored | |
| 151 | * include gap-residue score if params say to do so | |
| 152 | */ | |
| 153 | 71 | if ((!gap1 && !gap2) || params.includeGaps()) |
| 154 | { | |
| 155 | 67 | int seqDistance = SetUtils.countDisjunction(set1, set2); |
| 156 | 67 | distances[i][j] += seqDistance; |
| 157 | } | |
| 158 | } | |
| 159 | } | |
| 160 | } | |
| 161 | } | |
| 162 | ||
| 163 | /* | |
| 164 | * normalise the distance scores (summed over columns) by the | |
| 165 | * number of visible columns used in the calculation | |
| 166 | * and fill in the bottom half of the matrix | |
| 167 | */ | |
| 168 | // TODO JAL-2424 cpwidth may be out by 1 - affects scores but not tree shape | |
| 169 | 24 | for (int i = 0; i < noseqs; i++) |
| 170 | { | |
| 171 | 39 | for (int j = i + 1; j < noseqs; j++) |
| 172 | { | |
| 173 | 21 | distances[i][j] /= cpwidth; |
| 174 | 21 | distances[j][i] = distances[i][j]; |
| 175 | } | |
| 176 | } | |
| 177 | 6 | return new Matrix(distances); |
| 178 | } | |
| 179 | ||
| 180 | /** | |
| 181 | * Builds and returns a map containing a (possibly empty) list (one per | |
| 182 | * SeqCigar) of visible feature types at the given column position. The map | |
| 183 | * does not include entries for features which straddle a gapped column | |
| 184 | * positions. | |
| 185 | * | |
| 186 | * @param seqs | |
| 187 | * @param columnPosition | |
| 188 | * (0..) | |
| 189 | * @return | |
| 190 | */ | |
| 191 | 26 | protected Map<SeqCigar, Set<String>> findFeatureTypesAtColumn( |
| 192 | SeqCigar[] seqs, int columnPosition) | |
| 193 | { | |
| 194 | 26 | Map<SeqCigar, Set<String>> sfap = new HashMap<>(); |
| 195 | 26 | for (SeqCigar seq : seqs) |
| 196 | { | |
| 197 | 70 | int spos = seq.findPosition(columnPosition); |
| 198 | 70 | if (spos != -1) |
| 199 | { | |
| 200 | /* | |
| 201 | * position is not a gap | |
| 202 | */ | |
| 203 | 60 | Set<String> types = new HashSet<>(); |
| 204 | 60 | List<SequenceFeature> sfs = fr.findFeaturesAtResidue( |
| 205 | seq.getRefSeq(), spos, spos); | |
| 206 | 60 | for (SequenceFeature sf : sfs) |
| 207 | { | |
| 208 | 70 | types.add(sf.getType()); |
| 209 | } | |
| 210 | 60 | sfap.put(seq, types); |
| 211 | } | |
| 212 | } | |
| 213 | 26 | return sfap; |
| 214 | } | |
| 215 | ||
| 216 | 88 | @Override |
| 217 | public String getName() | |
| 218 | { | |
| 219 | 88 | return NAME; |
| 220 | } | |
| 221 | ||
| 222 | 0 | @Override |
| 223 | public String getDescription() | |
| 224 | { | |
| 225 | 0 | return description; |
| 226 | } | |
| 227 | ||
| 228 | 4 | @Override |
| 229 | public boolean isDNA() | |
| 230 | { | |
| 231 | 4 | return true; |
| 232 | } | |
| 233 | ||
| 234 | 2 | @Override |
| 235 | public boolean isProtein() | |
| 236 | { | |
| 237 | 2 | return true; |
| 238 | } | |
| 239 | ||
| 240 | 0 | @Override |
| 241 | public String toString() | |
| 242 | { | |
| 243 | 0 | return "Score between sequences based on hamming distance between binary vectors marking features displayed at each column"; |
| 244 | } | |
| 245 | } |