Class |
Line # |
Actions |
|||
---|---|---|---|---|---|
EnsemblCds | 41 | 19 | 10 |
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.ext.ensembl; | |
22 | ||
23 | import jalview.datamodel.SequenceFeature; | |
24 | import jalview.datamodel.SequenceI; | |
25 | import jalview.io.gff.SequenceOntologyFactory; | |
26 | import jalview.io.gff.SequenceOntologyI; | |
27 | ||
28 | import java.util.ArrayList; | |
29 | import java.util.List; | |
30 | ||
31 | /** | |
32 | * A client for direct fetching of CDS sequences from Ensembl (i.e. that part of | |
33 | * the genomic sequence that is translated to protein) | |
34 | * | |
35 | * TODO: not currently used as CDS sequences are computed from CDS features on | |
36 | * transcripts - delete this class? | |
37 | * | |
38 | * @author gmcarstairs | |
39 | * | |
40 | */ | |
41 | public class EnsemblCds extends EnsemblSeqProxy | |
42 | { | |
43 | /* | |
44 | * fetch cds features on genomic sequence (to identify the CDS regions) | |
45 | * and exon and variation features (to retain for display) | |
46 | */ | |
47 | private static final EnsemblFeatureType[] FEATURES_TO_FETCH = { | |
48 | EnsemblFeatureType.cds, EnsemblFeatureType.exon, | |
49 | EnsemblFeatureType.variation }; | |
50 | ||
51 | /** | |
52 | * Default constructor (to use rest.ensembl.org) | |
53 | */ | |
54 | 4 | public EnsemblCds() |
55 | { | |
56 | 4 | super(); |
57 | } | |
58 | ||
59 | /** | |
60 | * Constructor given the target domain to fetch data from | |
61 | * | |
62 | * @param d | |
63 | */ | |
64 | 0 | public EnsemblCds(String d) |
65 | { | |
66 | 0 | super(d); |
67 | } | |
68 | ||
69 | 0 | @Override |
70 | public String getDbName() | |
71 | { | |
72 | 0 | return "ENSEMBL (CDS)"; |
73 | } | |
74 | ||
75 | 0 | @Override |
76 | protected EnsemblSeqType getSourceEnsemblType() | |
77 | { | |
78 | 0 | return EnsemblSeqType.CDS; |
79 | } | |
80 | ||
81 | 0 | @Override |
82 | protected EnsemblFeatureType[] getFeaturesToFetch() | |
83 | { | |
84 | 0 | return FEATURES_TO_FETCH; |
85 | } | |
86 | ||
87 | /** | |
88 | * Answers true unless the feature type is 'CDS' (or a sub-type of CDS in the | |
89 | * Sequence Ontology). CDS features are only retrieved in order to identify | |
90 | * the cds sequence range, and are redundant information on the cds sequence | |
91 | * itself. | |
92 | */ | |
93 | 5 | @Override |
94 | protected boolean retainFeature(SequenceFeature sf, String accessionId) | |
95 | { | |
96 | 5 | if (SequenceOntologyFactory.getInstance().isA(sf.getType(), |
97 | SequenceOntologyI.CDS)) | |
98 | { | |
99 | 2 | return false; |
100 | } | |
101 | 3 | return featureMayBelong(sf, accessionId); |
102 | } | |
103 | ||
104 | /** | |
105 | * Answers a list of sequence features (if any) whose type is 'CDS' (or a | |
106 | * subtype of CDS in the Sequence Ontology), and whose Parent is the | |
107 | * transcript we are retrieving | |
108 | */ | |
109 | 2 | @Override |
110 | protected List<SequenceFeature> getIdentifyingFeatures(SequenceI seq, | |
111 | String accId) | |
112 | { | |
113 | 2 | List<SequenceFeature> result = new ArrayList<>(); |
114 | 2 | List<SequenceFeature> sfs = seq.getFeatures() |
115 | .getFeaturesByOntology(SequenceOntologyI.CDS); | |
116 | 2 | for (SequenceFeature sf : sfs) |
117 | { | |
118 | 7 | String parentFeature = (String) sf.getValue(PARENT); |
119 | 7 | if (accId.equals(parentFeature)) |
120 | { | |
121 | 4 | result.add(sf); |
122 | } | |
123 | } | |
124 | 2 | return result; |
125 | } | |
126 | ||
127 | /** | |
128 | * Overrides this method to trivially return a range which is the whole of the | |
129 | * nucleotide sequence. This is both faster than scanning for CDS features, | |
130 | * and also means we don't need to keep CDS features on CDS sequence (where | |
131 | * they are redundant information). | |
132 | */ | |
133 | 0 | protected List<int[]> getCdsRanges(SequenceI dnaSeq) |
134 | { | |
135 | 0 | int len = dnaSeq.getLength(); |
136 | 0 | List<int[]> ranges = new ArrayList<>(); |
137 | 0 | ranges.add(new int[] { 1, len }); |
138 | 0 | return ranges; |
139 | } | |
140 | ||
141 | } |