Class |
Line # |
Actions |
|||
---|---|---|---|---|---|
RNAalifoldClient | 52 | 108 | 46 |
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.ws.jws2; | |
22 | ||
23 | import jalview.datamodel.AlignmentAnnotation; | |
24 | import jalview.datamodel.Annotation; | |
25 | import jalview.gui.AlignFrame; | |
26 | import jalview.util.MessageManager; | |
27 | import jalview.ws.jws2.jabaws2.Jws2Instance; | |
28 | import jalview.ws.params.WsParamSetI; | |
29 | import jalview.ws.uimodel.AlignAnalysisUIText; | |
30 | ||
31 | import java.text.MessageFormat; | |
32 | import java.util.ArrayList; | |
33 | import java.util.LinkedHashMap; | |
34 | import java.util.List; | |
35 | import java.util.TreeSet; | |
36 | import java.util.regex.Pattern; | |
37 | ||
38 | import compbio.data.sequence.FastaSequence; | |
39 | import compbio.data.sequence.RNAStructReader.AlifoldResult; | |
40 | import compbio.data.sequence.RNAStructScoreManager; | |
41 | import compbio.data.sequence.Range; | |
42 | import compbio.data.sequence.Score; | |
43 | import compbio.metadata.Argument; | |
44 | ||
45 | /** | |
46 | * Client for the JABA RNA Alifold Service | |
47 | * | |
48 | * @author daluke - Daniel Barton | |
49 | * | |
50 | */ | |
51 | ||
52 | public class RNAalifoldClient extends JabawsCalcWorker | |
53 | { | |
54 | ||
55 | String methodName; | |
56 | ||
57 | AlignFrame af; | |
58 | ||
59 | // keeps track of whether the RNAalifold result includes base contact | |
60 | // probabilities | |
61 | boolean bpScores; | |
62 | ||
63 | 0 | public RNAalifoldClient(Jws2Instance sh, AlignFrame alignFrame, |
64 | WsParamSetI preset, List<Argument> paramset) | |
65 | { | |
66 | 0 | super(sh, alignFrame, preset, paramset); |
67 | 0 | af = alignFrame; |
68 | 0 | methodName = sh.serviceType; |
69 | 0 | alignedSeqs = true; |
70 | 0 | submitGaps = true; |
71 | 0 | nucleotidesAllowed = true; |
72 | 0 | proteinAllowed = false; |
73 | 0 | initViewportParams(); |
74 | } | |
75 | ||
76 | 0 | @Override |
77 | public String getCalcId() | |
78 | { | |
79 | 0 | return CALC_ID; |
80 | } | |
81 | ||
82 | private static String CALC_ID = "jalview.ws.jws2.RNAalifoldClient"; | |
83 | ||
84 | 6 | public static AlignAnalysisUIText getAlignAnalysisUITest() |
85 | { | |
86 | 6 | return new AlignAnalysisUIText( |
87 | compbio.ws.client.Services.RNAalifoldWS.toString(), | |
88 | jalview.ws.jws2.RNAalifoldClient.class, CALC_ID, true, false, | |
89 | true, MessageManager.getString("label.rnalifold_calculations"), | |
90 | MessageManager.getString("tooltip.rnalifold_calculations"), | |
91 | MessageManager.getString("label.rnalifold_settings"), | |
92 | MessageManager.getString("tooltip.rnalifold_settings")); | |
93 | } | |
94 | ||
95 | 0 | @Override |
96 | public String getServiceActionText() | |
97 | { | |
98 | 0 | return "Submitting RNA alignment for Secondary Structure prediction using " |
99 | + "RNAalifold Service"; | |
100 | } | |
101 | ||
102 | 0 | @Override |
103 | boolean checkValidInputSeqs(boolean dynamic, List<FastaSequence> seqs) | |
104 | { | |
105 | 0 | return (seqs.size() > 1); |
106 | } | |
107 | ||
108 | 0 | @Override |
109 | public void updateResultAnnotation(boolean immediate) | |
110 | { | |
111 | ||
112 | 0 | if (immediate || !calcMan.isWorking(this) && scoremanager != null) |
113 | { | |
114 | ||
115 | 0 | List<AlignmentAnnotation> ourAnnot = new ArrayList<AlignmentAnnotation>(); |
116 | ||
117 | // Unpack the ScoreManager | |
118 | 0 | List<String> structs = ((RNAStructScoreManager) scoremanager) |
119 | .getStructs(); | |
120 | 0 | List<TreeSet<Score>> data = ((RNAStructScoreManager) scoremanager) |
121 | .getData(); | |
122 | ||
123 | // test to see if this data object contains base pair contacts | |
124 | 0 | Score fscore = data.get(0).first(); |
125 | 0 | this.bpScores = (fscore.getMethod() |
126 | .equals(AlifoldResult.contactProbabilities.toString())); | |
127 | ||
128 | // add annotation for the consensus sequence alignment | |
129 | 0 | createAnnotationRowforScoreHolder(ourAnnot, getCalcId(), |
130 | structs.get(0), null, null); | |
131 | ||
132 | // Add annotations for the mfe Structure | |
133 | 0 | createAnnotationRowforScoreHolder(ourAnnot, getCalcId(), |
134 | structs.get(1), data.get(1), null); | |
135 | ||
136 | // decide whether to add base pair contact probability histogram | |
137 | 0 | int count = 2; |
138 | 0 | if (bpScores) |
139 | { | |
140 | 0 | createAnnotationRowforScoreHolder(ourAnnot, getCalcId(), |
141 | structs.get(2), data.get(0), data.get(2)); | |
142 | 0 | count++; |
143 | } | |
144 | ||
145 | // Now loop for the rest of the Annotations (if there it isn't stochastic | |
146 | // output | |
147 | // only the centroid and MEA structures remain anyway) | |
148 | 0 | for (int i = count; i < structs.size(); i++) |
149 | { | |
150 | // The ensemble values should be displayed in the description of the | |
151 | // first (or all?) Stochastic Backtrack Structures. | |
152 | 0 | if (!data.get(i).first().getMethod() |
153 | .equals(AlifoldResult.ensembleValues.toString())) | |
154 | { | |
155 | ||
156 | 0 | createAnnotationRowforScoreHolder(ourAnnot, getCalcId(), |
157 | structs.get(i), data.get(i), null); | |
158 | } | |
159 | } | |
160 | ||
161 | 0 | if (ourAnnot.size() > 0) |
162 | { | |
163 | ||
164 | 0 | updateOurAnnots(ourAnnot); |
165 | 0 | ap.adjustAnnotationHeight(); |
166 | } | |
167 | } | |
168 | } | |
169 | ||
170 | 0 | protected void createAnnotationRowforScoreHolder( |
171 | List<AlignmentAnnotation> ourAnnot, String calcId, String struct, | |
172 | TreeSet<Score> data, TreeSet<Score> descriptionData) | |
173 | { | |
174 | /* | |
175 | * If contactProbability information is returned from RNAalifold it is | |
176 | * stored in the first TreeSet<Score> object corresponding to the String Id | |
177 | * which holds the consensus alignment. The method enumeration is then | |
178 | * updated to AlifoldResult.contactProbabilties. This line recreates the | |
179 | * same data object as was overwritten with the contact probabilites data. | |
180 | */ | |
181 | 0 | if (data == null) |
182 | { | |
183 | 0 | data = compbio.data.sequence.RNAStructReader |
184 | .newEmptyScore(AlifoldResult.consensusAlignment); | |
185 | } | |
186 | ||
187 | 0 | if (descriptionData == null) |
188 | { | |
189 | 0 | descriptionData = data; |
190 | } | |
191 | ||
192 | 0 | String[] typenameAndDescription = constructTypenameAndDescription( |
193 | descriptionData.first()); | |
194 | 0 | String typename = typenameAndDescription[0]; |
195 | 0 | String description = typenameAndDescription[1]; |
196 | ||
197 | 0 | AlignmentAnnotation annotation = alignViewport.getAlignment() |
198 | .findOrCreateAnnotation(typename, calcId, false, null, null); | |
199 | ||
200 | 0 | constructAnnotationFromScoreHolder(annotation, struct, data); |
201 | ||
202 | /* | |
203 | * update annotation description with the free Energy, frequency in ensemble | |
204 | * or other data where appropriate. | |
205 | * | |
206 | * Doesnt deal with AlifoldResult.ensembleValues, the free energy of | |
207 | * ensemble and frequency of mfe structure in ensemble. How to deal with | |
208 | * these? | |
209 | */ | |
210 | 0 | annotation.description = description; |
211 | ||
212 | 0 | annotation.belowAlignment = false; |
213 | // annotation.showAllColLabels = true; | |
214 | ||
215 | 0 | alignViewport.getAlignment().validateAnnotation(annotation); |
216 | 0 | af.setMenusForViewport(); |
217 | ||
218 | 0 | ourAnnot.add(annotation); |
219 | } | |
220 | ||
221 | 0 | private AlignmentAnnotation constructAnnotationFromScoreHolder( |
222 | AlignmentAnnotation annotation, String struct, | |
223 | TreeSet<Score> data) | |
224 | { | |
225 | 0 | Annotation[] anns = new Annotation[gapMap != null ? gapMap.length + 1 |
226 | : struct.length()]; | |
227 | ||
228 | 0 | if (data != null && data.size() > 1 && data.first().getMethod() |
229 | .equals(AlifoldResult.contactProbabilities.toString())) | |
230 | { | |
231 | ||
232 | // The base pair probabilities are stored in a set in scoreholder. we want | |
233 | // a map | |
234 | 0 | LinkedHashMap<Range, Float> basePairs = new LinkedHashMap<Range, Float>(); |
235 | 0 | for (Score score : data) |
236 | { | |
237 | // The Score objects contain a set of size one containing the range and | |
238 | // an ArrayList<float> of size one containing the probabilty | |
239 | 0 | basePairs.put(score.getRanges().first(), |
240 | Float.valueOf(score.getScores().get(0))); | |
241 | } | |
242 | ||
243 | 0 | for (int i = 0, ri = 0, iEnd = struct.length(); i < iEnd; i++, ri++) |
244 | { | |
245 | 0 | if (gapMap != null) |
246 | { | |
247 | // skip any gapped columns in the input data | |
248 | 0 | while (!gapMap[ri]) |
249 | { | |
250 | 0 | ri++; |
251 | } | |
252 | } | |
253 | // Return all the contacts associated with position i | |
254 | 0 | LinkedHashMap<Range, Float> contacts = isContact(basePairs, i + 1); |
255 | ||
256 | 0 | String description = ""; |
257 | 0 | float prob = 0f; |
258 | ||
259 | 0 | if (contacts.size() == 0) |
260 | { | |
261 | 0 | description = "No Data"; |
262 | } | |
263 | else | |
264 | { | |
265 | 0 | for (Range contact : contacts.keySet()) |
266 | { | |
267 | 0 | float t = contacts.get(contact); |
268 | 0 | if (t > prob) |
269 | { | |
270 | 0 | prob = t; |
271 | } | |
272 | 0 | description += Integer.toString(contact.from) + "->" |
273 | + Integer.toString(contact.to) + ": " | |
274 | + Float.toString(t) + "% | "; | |
275 | } | |
276 | } | |
277 | ||
278 | 0 | anns[ri] = new Annotation(struct.substring(i, i + 1), description, |
279 | isSS(struct.charAt(i)), prob); | |
280 | } | |
281 | } | |
282 | 0 | else if (data == null || data.size() == 1) |
283 | { | |
284 | 0 | for (int i = 0, ri = 0, iEnd = struct.length(); i < iEnd; i++, ri++) |
285 | { | |
286 | 0 | if (gapMap != null) |
287 | { | |
288 | // skip any gapped columns in the input data | |
289 | 0 | while (!gapMap[ri] && ri < gapMap.length) |
290 | { | |
291 | 0 | ri++; |
292 | } | |
293 | 0 | if (ri == gapMap.length) |
294 | { | |
295 | 0 | break; |
296 | } | |
297 | } | |
298 | 0 | anns[ri] = new Annotation(struct.substring(i, i + 1), "", |
299 | isSS(struct.charAt(i)), Float.NaN); | |
300 | } | |
301 | ||
302 | 0 | annotation.graph = 0; // No graph |
303 | } | |
304 | ||
305 | 0 | annotation.annotations = anns; |
306 | ||
307 | 0 | return annotation; |
308 | } | |
309 | ||
310 | 0 | private String[] constructTypenameAndDescription(Score score) |
311 | { | |
312 | 0 | String description = ""; |
313 | 0 | String typename = ""; |
314 | 0 | String datatype = score.getMethod(); |
315 | ||
316 | // Look up java switch syntax and use one here | |
317 | 0 | if (datatype.equals(AlifoldResult.mfeStructure.toString())) |
318 | { | |
319 | ||
320 | 0 | description = MessageFormat.format( |
321 | "Minimum Free Energy Structure. Energy: {0} = {1} + {2}", | |
322 | score.getScores().get(0), score.getScores().get(1), | |
323 | score.getScores().get(2)); | |
324 | 0 | typename = "MFE Structure"; |
325 | } | |
326 | 0 | else if (datatype |
327 | .equals(AlifoldResult.contactProbabilityStructure.toString())) | |
328 | { | |
329 | 0 | description = MessageFormat.format("Base Pair Contact Probabilities. " |
330 | + "Energy of Ensemble: {0} Frequency of Ensemble: {1}", | |
331 | score.getScores().get(0), score.getScores().get(1)); | |
332 | 0 | typename = "Contact Probabilities"; |
333 | } | |
334 | 0 | else if (datatype.equals(AlifoldResult.centroidStructure.toString())) |
335 | { | |
336 | 0 | description = MessageFormat.format( |
337 | "Centroid Structure. Energy: {0} = {1} + {2}", | |
338 | score.getScores().get(0), score.getScores().get(1), | |
339 | score.getScores().get(2)); | |
340 | 0 | typename = "Centroid Structure"; |
341 | } | |
342 | 0 | else if (datatype.equals(AlifoldResult.stochBTStructure.toString())) |
343 | { | |
344 | 0 | if (score.getScores().size() > 0) |
345 | { | |
346 | 0 | description = MessageFormat.format("Probability: {0} Energy: {1}", |
347 | score.getScores().get(0), score.getScores().get(1)); | |
348 | } | |
349 | else | |
350 | { | |
351 | 0 | description = "Stochastic Backtrack Structure"; |
352 | } | |
353 | } | |
354 | 0 | else if (datatype.equals(AlifoldResult.MEAStucture.toString())) |
355 | { | |
356 | 0 | description = MessageFormat.format( |
357 | "Maximum Expected Accuracy Values: '{' {0} MEA={1} '}", | |
358 | score.getScores().get(0), score.getScores().get(1)); | |
359 | 0 | typename = "MEA Structure"; |
360 | } | |
361 | 0 | else if (datatype.equals(AlifoldResult.consensusAlignment.toString())) |
362 | { | |
363 | 0 | typename = "RNAalifold Consensus"; |
364 | 0 | description = "Consensus Alignment Produced by RNAalifold"; |
365 | } | |
366 | else | |
367 | { | |
368 | 0 | typename = datatype; |
369 | 0 | description = typename; |
370 | } | |
371 | ||
372 | 0 | return new String[] { typename, description }; |
373 | } | |
374 | ||
375 | // Check whether, at position i there is a base contact and return all the | |
376 | // contacts at this position. Should be in order of descending probability. | |
377 | 0 | private LinkedHashMap<Range, Float> isContact( |
378 | LinkedHashMap<Range, Float> basePairs, int i) | |
379 | { | |
380 | 0 | LinkedHashMap<Range, Float> contacts = new LinkedHashMap<Range, Float>(); |
381 | ||
382 | 0 | for (Range contact : basePairs.keySet()) |
383 | { | |
384 | // finds the contacts associtated with position i ordered by the natural | |
385 | // ordering of the Scores TreeSet in ScoreManager which is, descending | |
386 | // probability | |
387 | 0 | if (contact.from == i || contact.to == i) |
388 | { | |
389 | 0 | contacts.put(contact, basePairs.get(contact)); |
390 | } | |
391 | } | |
392 | ||
393 | 0 | return contacts; |
394 | } | |
395 | ||
396 | 0 | private char isSS(char chr) |
397 | { | |
398 | 0 | String regex = "\\(|\\)|\\{|\\}|\\[|\\]"; |
399 | 0 | char ss = (Pattern.matches(regex, Character.toString(chr))) ? 'S' : ' '; |
400 | 0 | return ss; |
401 | } | |
402 | } |