3 import jalview.bin.Cache;
4 import jalview.datamodel.Alignment;
5 import jalview.datamodel.AlignmentAnnotation;
6 import jalview.datamodel.AlignmentI;
7 import jalview.datamodel.HiddenMarkovModel;
8 import jalview.datamodel.SequenceI;
9 import jalview.gui.AlignFrame;
10 import jalview.gui.Desktop;
11 import jalview.gui.JvOptionPane;
12 import jalview.io.DataSourceType;
13 import jalview.io.FileParse;
14 import jalview.io.StockholmFile;
15 import jalview.util.FileUtils;
16 import jalview.util.MessageManager;
17 import jalview.ws.params.ArgumentI;
18 import jalview.ws.params.simple.BooleanOption;
19 import jalview.ws.params.simple.Option;
21 import java.io.BufferedReader;
23 import java.io.FileReader;
24 import java.io.IOException;
25 import java.util.ArrayList;
26 import java.util.Collections;
27 import java.util.List;
28 import java.util.Scanner;
30 import javax.swing.JOptionPane;
32 public class HMMSearch extends HmmerCommand
34 static final String HMMSEARCH = "hmmsearch";
37 * constants for i18n lookup of passed parameter names
39 static final String DATABASE_KEY = "label.database";
41 static final String THIS_ALIGNMENT_KEY = "label.this_alignment";
43 static final String USE_ACCESSIONS_KEY = "label.use_accessions";
45 static final String AUTO_ALIGN_SEQS_KEY = "label.auto_align_seqs";
47 static final String NUMBER_OF_RESULTS_KEY = "label.number_of_results";
49 static final String TRIM_TERMINI_KEY = "label.trim_termini";
51 static final String REPORTING_CUTOFF_KEY = "label.reporting_cutoff";
53 static final String CUTOFF_NONE = "None";
55 static final String CUTOFF_SCORE = "Score";
57 static final String CUTOFF_EVALUE = "E-Value";
59 static final String SEQ_EVALUE_KEY = "label.seq_evalue";
61 static final String DOM_EVALUE_KEY = "label.dom_evalue";
63 static final String SEQ_SCORE_KEY = "label.seq_score";
65 static final String DOM_SCORE_KEY = "label.dom_score";
67 boolean realign = false;
71 int seqsToReturn = Integer.MAX_VALUE;
75 private String databaseName;
78 * Constructor for the HMMSearchThread
82 public HMMSearch(AlignFrame af, List<ArgumentI> args)
88 * Runs the HMMSearchThread: the data on the alignment or group is exported,
89 * then the command is executed in the command line and then the data is
90 * imported and displayed in a new frame. Call this method directly to execute
91 * synchronously, or via start() in a new Thread for asynchronously.
96 HiddenMarkovModel hmm = getHmmProfile();
99 // shouldn't happen if we got this far
100 Cache.log.error("Error: no hmm for hmmsearch");
104 SequenceI hmmSeq = hmm.getConsensusSequence();
105 long msgId = System.currentTimeMillis();
106 af.setProgressBar(MessageManager.getString("status.running_hmmsearch"),
111 File hmmFile = FileUtils.createTempFile("hmm", ".hmm");
112 File hitsAlignmentFile = FileUtils.createTempFile("hitAlignment",
114 File searchOutputFile = FileUtils.createTempFile("searchOutput",
117 exportHmm(hmm, hmmFile.getAbsoluteFile());
119 boolean ran = runCommand(searchOutputFile, hitsAlignmentFile, hmmFile);
122 JvOptionPane.showInternalMessageDialog(af, MessageManager
123 .formatMessage("warn.command_failed", "hmmsearch"));
127 importData(hmmSeq, hitsAlignmentFile, hmmFile, searchOutputFile);
128 // TODO make realignment of search results a step at this level
129 // and make it conditional on this.realign
130 } catch (IOException | InterruptedException e)
136 af.setProgressBar("", msgId);
141 * Executes an hmmsearch with the given hmm as input. The database to be
142 * searched is a local file as specified by the 'Database' parameter, or the
143 * current alignment (written to file) if none is specified.
145 * @param searchOutputFile
146 * @param hitsAlignmentFile
150 * @throws IOException
152 private boolean runCommand(File searchOutputFile, File hitsAlignmentFile,
153 File hmmFile) throws IOException
155 String command = getCommandPath(HMMSEARCH);
161 List<String> args = new ArrayList<>();
163 buildArguments(args, searchOutputFile, hitsAlignmentFile, hmmFile);
165 return runCommand(args);
169 * Appends command line arguments to the given list, to specify input and
170 * output files for the search, and any additional options that may have been
171 * passed from the parameters dialog
174 * @param searchOutputFile
175 * @param hitsAlignmentFile
177 * @throws IOException
179 protected void buildArguments(List<String> args, File searchOutputFile,
180 File hitsAlignmentFile, File hmmFile) throws IOException
183 args.add(getFilePath(searchOutputFile));
185 args.add(getFilePath(hitsAlignmentFile));
187 boolean dbFound = false;
189 File databaseFile = null;
191 boolean useEvalueCutoff = false;
192 boolean useScoreCutoff = false;
193 String seqEvalueCutoff = null;
194 String domEvalueCutoff = null;
195 String seqScoreCutoff = null;
196 String domScoreCutoff = null;
197 databaseName = "Alignment";
201 for (ArgumentI arg : params)
203 String name = arg.getName();
204 if (MessageManager.getString(NUMBER_OF_RESULTS_KEY)
207 seqsToReturn = Integer.parseInt(arg.getValue());
209 else if (MessageManager.getString(AUTO_ALIGN_SEQS_KEY)
214 else if (MessageManager.getString(USE_ACCESSIONS_KEY)
219 else if (MessageManager.getString(REPORTING_CUTOFF_KEY)
222 if (CUTOFF_EVALUE.equals(arg.getValue()))
224 useEvalueCutoff = true;
226 else if (CUTOFF_SCORE.equals(arg.getValue()))
228 useScoreCutoff = true;
231 else if (MessageManager.getString(SEQ_EVALUE_KEY).equals(name))
233 seqEvalueCutoff = arg.getValue();
235 else if (MessageManager.getString(SEQ_SCORE_KEY).equals(name))
237 seqScoreCutoff = arg.getValue();
239 else if (MessageManager.getString(DOM_EVALUE_KEY)
242 domEvalueCutoff = arg.getValue();
244 else if (MessageManager.getString(DOM_SCORE_KEY).equals(name))
246 domScoreCutoff = arg.getValue();
248 else if (MessageManager.getString(TRIM_TERMINI_KEY)
253 else if (MessageManager.getString(DATABASE_KEY).equals(name))
256 dbPath = arg.getValue();
257 if (!MessageManager.getString(THIS_ALIGNMENT_KEY)
260 int pos = dbPath.lastIndexOf(File.separator);
261 databaseName = dbPath.substring(pos + 1);
262 databaseFile = new File(dbPath);
271 args.add(seqEvalueCutoff);
273 args.add(domEvalueCutoff);
275 else if (useScoreCutoff)
278 args.add(seqScoreCutoff);
280 args.add(domScoreCutoff);
283 if (!dbFound || MessageManager.getString(THIS_ALIGNMENT_KEY)
287 * no external database specified for search, so
288 * export current alignment as 'database' to search,
289 * excluding any HMM consensus sequences it contains
291 databaseFile = FileUtils.createTempFile("database", ".sto");
292 AlignmentI al = af.getViewport().getAlignment();
293 AlignmentI copy = new Alignment(al);
294 List<SequenceI> hmms = copy.getHmmSequences();
295 for (SequenceI hmmSeq : hmms)
297 copy.deleteSequence(hmmSeq);
299 exportStockholm(copy.getSequencesArray(), databaseFile, null);
302 args.add(getFilePath(hmmFile));
303 args.add(getFilePath(databaseFile));
307 * Imports the data from the temporary file to which the output of hmmsearch
308 * was directed. The results are optionally realigned using hmmalign.
312 private void importData(SequenceI hmmSeq, File inputAlignmentTemp,
313 File hmmTemp, File searchOutputFile)
314 throws IOException, InterruptedException
316 BufferedReader br = new BufferedReader(
317 new FileReader(inputAlignmentTemp));
320 if (br.readLine() == null)
322 JOptionPane.showMessageDialog(af,
323 MessageManager.getString("label.no_sequences_found"));
326 StockholmFile file = new StockholmFile(new FileParse(
327 inputAlignmentTemp.getAbsolutePath(), DataSourceType.FILE));
328 seqs = file.getSeqsAsArray();
330 readTable(searchOutputFile);
332 int seqCount = Math.min(seqs.length, seqsToReturn);
333 SequenceI[] hmmAndSeqs = new SequenceI[seqCount + 1];
334 hmmAndSeqs[0] = hmmSeq;
335 System.arraycopy(seqs, 0, hmmAndSeqs, 1, seqCount);
339 realignResults(hmmAndSeqs);
343 AlignmentI al = new Alignment(hmmAndSeqs);
344 AlignFrame alignFrame = new AlignFrame(al, AlignFrame.DEFAULT_WIDTH,
345 AlignFrame.DEFAULT_HEIGHT);
346 String ttl = "hmmSearch of " + databaseName + " using "
348 Desktop.addInternalFrame(alignFrame, ttl, AlignFrame.DEFAULT_WIDTH,
349 AlignFrame.DEFAULT_HEIGHT);
353 inputAlignmentTemp.delete();
354 searchOutputFile.delete();
365 * Realigns the given sequences using hmmalign, to the HMM profile sequence
366 * which is the first in the array, and opens the results in a new frame
370 protected void realignResults(SequenceI[] hmmAndSeqs)
373 * and align the search results to the HMM profile
375 AlignmentI al = new Alignment(hmmAndSeqs);
376 AlignFrame frame = new AlignFrame(al, 1, 1);
377 List<ArgumentI> alignArgs = new ArrayList<>();
378 String alignTo = hmmAndSeqs[0].getName();
379 List<String> options = Collections.singletonList(alignTo);
380 Option option = new Option(MessageManager.getString("label.use_hmm"),
381 "", true, alignTo, alignTo, options, null);
382 alignArgs.add(option);
385 alignArgs.add(new BooleanOption(
386 MessageManager.getString(TRIM_TERMINI_KEY),
387 MessageManager.getString("label.trim_termini_desc"), true,
390 HmmerCommand hmmalign = new HMMAlign(frame, alignArgs);
395 * Reads in the scores table output by hmmsearch and adds annotation to
396 * sequences for E-value and bit score
398 * @param inputTableTemp
399 * @throws IOException
401 void readTable(File inputTableTemp) throws IOException
403 BufferedReader br = new BufferedReader(new FileReader(inputTableTemp));
405 while (!line.startsWith("Query:"))
407 line = br.readLine();
409 for (int i = 0; i < 5; i++)
411 line = br.readLine();
415 while (!" ------ inclusion threshold ------".equals(line)
418 SequenceI seq = seqs[index];
419 Scanner scanner = new Scanner(line);
420 String str = scanner.next();
421 addScoreAnnotation(str, seq, "hmmsearch E-value",
422 "Full sequence E-value");
423 str = scanner.next();
424 addScoreAnnotation(str, seq, "hmmsearch Score",
425 "Full sequence bit score");
427 line = br.readLine();
435 * A helper method that adds one score-only (non-positional) annotation to a
443 protected void addScoreAnnotation(String value, SequenceI seq,
444 String label, String description)
448 AlignmentAnnotation annot = new AlignmentAnnotation(label,
450 annot.setCalcId(HMMSEARCH);
451 double eValue = Double.parseDouble(value);
452 annot.setScore(eValue);
453 annot.setSequenceRef(seq);
454 seq.addAlignmentAnnotation(annot);
455 } catch (NumberFormatException e)
457 System.err.println("Error parsing " + label + " from " + value);