X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Fws%2Fjws2%2FRNAalifoldClient.java;h=769100e28295ce326e3bce7ad8ff74c93eb84735;hb=17e77c3f2949a0729322b4a8d907f3f34b6a9914;hp=959214241209c0870cb94d6ddb878fb75d65b984;hpb=24c4dce749cbfccc4435e442090d04013e3e1b99;p=jalview.git
diff --git a/src/jalview/ws/jws2/RNAalifoldClient.java b/src/jalview/ws/jws2/RNAalifoldClient.java
index 9592142..769100e 100644
--- a/src/jalview/ws/jws2/RNAalifoldClient.java
+++ b/src/jalview/ws/jws2/RNAalifoldClient.java
@@ -1,186 +1,405 @@
+/*
+ * Jalview - A Sequence Alignment Editor and Viewer (Version 2.9)
+ * Copyright (C) 2015 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
+ * 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 .
+ * The Jalview Authors are detailed in the 'AUTHORS' file.
+ */
package jalview.ws.jws2;
import jalview.api.AlignCalcWorkerI;
import jalview.datamodel.AlignmentAnnotation;
import jalview.datamodel.Annotation;
import jalview.gui.AlignFrame;
-import jalview.schemes.NucleotideColourScheme;
+import jalview.util.MessageManager;
import jalview.ws.jws2.jabaws2.Jws2Instance;
import jalview.ws.params.WsParamSetI;
+import jalview.ws.uimodel.AlignAnalysisUIText;
+import java.text.MessageFormat;
import java.util.ArrayList;
-import java.util.Iterator;
+import java.util.LinkedHashMap;
import java.util.List;
-import java.util.Map;
-import java.util.TreeMap;
import java.util.TreeSet;
+import java.util.regex.Pattern;
-import compbio.data.msa.jaxws.Align;
+import compbio.data.sequence.FastaSequence;
+import compbio.data.sequence.RNAStructReader.AlifoldResult;
+import compbio.data.sequence.RNAStructScoreManager;
import compbio.data.sequence.Range;
import compbio.data.sequence.Score;
-import compbio.data.sequence.RNAStructScoreManager;
-import compbio.data.sequence.ScoreManager;
import compbio.metadata.Argument;
-public class RNAalifoldClient extends JabawsAlignCalcWorker implements
- AlignCalcWorkerI
+/**
+ * Client for the JABA RNA Alifold Service
+ *
+ * @author daluke - Daniel Barton
+ *
+ */
+
+public class RNAalifoldClient extends JabawsCalcWorker implements
+ AlignCalcWorkerI
{
-
- // test
-
- String methodName;
-
- AlignFrame af;
-
- public RNAalifoldClient(Jws2Instance sh, AlignFrame alignFrame,
- WsParamSetI preset, List paramset)
- {
- super(sh, alignFrame, preset, paramset);
-
- if (arguments == null) arguments = new ArrayList();
- arguments.add(sh.getRunnerConfig().getArgumentByOptionName("-p"));
-
-
- af = alignFrame;
- methodName = sh.serviceType;
-
- // defult false. Which one here?
- // submitGaps = true;
- nucleotidesAllowed = true;
- proteinAllowed = false;
-
- arguments.add(sh.getRunnerConfig().getArgumentByOptionName("-p"));
- }
-
- @Override
- public String getServiceActionText()
- {
- return "Submitting RNA alignment for Secondary Structure prediction using "
- + "RNAalifold Service";
- }
-
- @Override
- public void updateResultAnnotation(boolean immediate)
- {
-
-
- if (immediate || !calcMan.isWorking(this) && scoremanager != null)
- {
-
- List ourAnnot = new ArrayList();
- // ourAnnots = new ArrayList();
-
- // So I don't have to do any more casting
- List structs = ((RNAStructScoreManager) scoremanager).getStructs();
- List> data = ((RNAStructScoreManager) scoremanager).getData();
-
-
- // I think this will never find an annotation at the moment. It will always create...
- AlignmentAnnotation annotation = alignViewport.getAlignment()
- .findOrCreateAnnotation("Consensus Structure", getCalcId(), false, null , null);
-
- // construct Annotation from scoremanager
-
- // Deal with the consensus structure and (?)BasePair Probabilities
- Annotation[] anns = new Annotation[structs.get(1).length()];
-
- // check if the first Score object is populated with base pair probabilities
- Score fscore = data.get(0).first();
- boolean BPScores = (fscore.getScores().size() > 0
- && fscore.getRanges() != null);
-
- TreeMap basePairs = null;
- if (BPScores) {
- // The base pair probabilities are stored in a set in scoremanager... we want a map
- basePairs = new TreeMap();
- for (Score score : data.get(0)) {
- // The Score objects contain a set of size one containing the range and
- // an ArrayList of size one containing the probabilty
- basePairs.put(score.getRanges().first(), new Float(score.getScores().get(0)));
- }
- }
-
- // ignoring the Consensus alignemnt for now, get the Structure and make an AlignmentAnnotation
- String struct = structs.get(1); // get(1)
- for (int i = 0; i < struct.length(); i++) {
-
- if (BPScores) {
- // Return all the contacts associated with position i
- List contacts = isContact(basePairs, i+1);
-
- if (contacts.size() == 0) {
- anns[i] = new Annotation(struct.substring(i, i+1), "", struct.charAt(i), 0f);
- }
- else if (contacts.size() == 1) {
- // There is only one contact associated with this base
- float prob = basePairs.get(contacts.get(0));
- anns[i] = new Annotation(struct.substring(i, i+1), "", struct.charAt(i), prob);
- }
- else if (contacts.size() > 1) {
- // For now we will simply deal with alternate contact information by mentioning its
- // existance in the description
- float prob = basePairs.get(contacts.get(0));
- anns[i] = new Annotation(struct.substring(i, i+1), "This base has alternate contacts",
- struct.charAt(i), prob);
- }
- }
- else {
- // Same as the first if from the previous block
- anns[i] = new Annotation(struct.substring(i, i+1), "", struct.charAt(i), 0f);
- }
- }
-
- System.out.println("size of anns: " + anns.length);
-
- // Set the annotation to the AlignmentAnnotation object
- annotation.annotations = anns;
-
- // Set the probability
- annotation.setScore(data.get(1).first().getScores().get(0));
-
-
-
- // old
-// AlignmentAnnotation annot = new AlignmentAnnotation("Consensus Structure", "Free Energy", anns);
-
-
-
-
- System.out.println("RNAalifoldClient - annotation:\n");
- for (Annotation ann : annotation.annotations) {
- System.out.print(ann.toString()+"|");
- }
- System.out.println();
-
- // Instead of this look at existing methods for creating annotations
- ourAnnot.add(annotation);
-
-
- if (ourAnnot.size() > 0) {
- updateOurAnnots(ourAnnot);
- }
-
- }
- }
-
- // Check whether, at position i there is a base contact and return all the
- // contacts at this position. Should be in order of descending probability.
- private List isContact(TreeMap basePairs, int i) {
-
- List contacts = new ArrayList();
-
- for (Range contact : basePairs.keySet()) {
- // finds the contacts associtated with position i ordered by the natural
- // ordering of the Scores TreeSet in ScoreManager which is, descending probability
- if (contact.from == i || contact.to == i) contacts.add(contact);
- }
-
- return contacts;
- }
-
- public String getCalcId()
+
+ String methodName;
+
+ AlignFrame af;
+
+ // keeps track of whether the RNAalifold result includes base contact
+ // probabilities
+ boolean bpScores;
+
+ public RNAalifoldClient(Jws2Instance sh, AlignFrame alignFrame,
+ WsParamSetI preset, List paramset)
+ {
+ super(sh, alignFrame, preset, paramset);
+ af = alignFrame;
+ methodName = sh.serviceType;
+ alignedSeqs = true;
+ submitGaps = true;
+ nucleotidesAllowed = true;
+ proteinAllowed = false;
+ initViewportParams();
+ }
+
+ public String getCalcId()
+ {
+ return CALC_ID;
+ }
+
+ private static String CALC_ID = "jalview.ws.jws2.RNAalifoldClient";
+
+ public static AlignAnalysisUIText getAlignAnalysisUITest()
+ {
+ return new AlignAnalysisUIText(
+ compbio.ws.client.Services.RNAalifoldWS.toString(),
+ jalview.ws.jws2.RNAalifoldClient.class, CALC_ID, true, false,
+ true, MessageManager.getString("label.rnalifold_calculations"),
+ MessageManager.getString("tooltip.rnalifold_calculations"),
+ MessageManager.getString("label.rnalifold_settings"),
+ MessageManager.getString("tooltip.rnalifold_settings"));
+ }
+
+ @Override
+ public String getServiceActionText()
+ {
+ return "Submitting RNA alignment for Secondary Structure prediction using "
+ + "RNAalifold Service";
+ }
+
+ @Override
+ boolean checkValidInputSeqs(boolean dynamic, List seqs)
+ {
+ return (seqs.size() > 1);
+ }
+
+ @Override
+ public void updateResultAnnotation(boolean immediate)
+ {
+
+ if (immediate || !calcMan.isWorking(this) && scoremanager != null)
+ {
+
+ List ourAnnot = new ArrayList();
+
+ // Unpack the ScoreManager
+ List structs = ((RNAStructScoreManager) scoremanager)
+ .getStructs();
+ List> data = ((RNAStructScoreManager) scoremanager)
+ .getData();
+
+ // test to see if this data object contains base pair contacts
+ Score fscore = data.get(0).first();
+ this.bpScores = (fscore.getMethod()
+ .equals(AlifoldResult.contactProbabilities.toString()));
+
+ // add annotation for the consensus sequence alignment
+ createAnnotationRowforScoreHolder(ourAnnot, getCalcId(),
+ structs.get(0), null, null);
+
+ // Add annotations for the mfe Structure
+ createAnnotationRowforScoreHolder(ourAnnot, getCalcId(),
+ structs.get(1), data.get(1), null);
+
+ // decide whether to add base pair contact probability histogram
+ int count = 2;
+ if (bpScores)
+ {
+ createAnnotationRowforScoreHolder(ourAnnot, getCalcId(),
+ structs.get(2), data.get(0), data.get(2));
+ count++;
+ }
+
+ // Now loop for the rest of the Annotations (if there it isn't stochastic
+ // output
+ // only the centroid and MEA structures remain anyway)
+ for (int i = count; i < structs.size(); i++)
+ {
+ // The ensemble values should be displayed in the description of the
+ // first (or all?) Stochastic Backtrack Structures.
+ if (!data.get(i).first().getMethod()
+ .equals(AlifoldResult.ensembleValues.toString()))
+ {
+
+ createAnnotationRowforScoreHolder(ourAnnot, getCalcId(),
+ structs.get(i), data.get(i), null);
+ }
+ }
+
+ if (ourAnnot.size() > 0)
+ {
+
+ updateOurAnnots(ourAnnot);
+ ap.adjustAnnotationHeight();
+ }
+ }
+ }
+
+ protected void createAnnotationRowforScoreHolder(
+ List ourAnnot, String calcId, String struct,
+ TreeSet data, TreeSet descriptionData)
+ {
+ /*
+ * If contactProbability information is returned from RNAalifold it is
+ * stored in the first TreeSet object corresponding to the String Id
+ * which holds the consensus alignment. The method enumeration is then
+ * updated to AlifoldResult.contactProbabilties. This line recreates the
+ * same data object as was overwritten with the contact probabilites data.
+ */
+ if (data == null)
+ {
+ data = compbio.data.sequence.RNAStructReader
+ .newEmptyScore(AlifoldResult.consensusAlignment);
+ }
+
+ if (descriptionData == null)
+ {
+ descriptionData = data;
+ }
+
+ String[] typenameAndDescription = constructTypenameAndDescription(descriptionData
+ .first());
+ String typename = typenameAndDescription[0];
+ String description = typenameAndDescription[1];
+
+ AlignmentAnnotation annotation = alignViewport.getAlignment()
+ .findOrCreateAnnotation(typename, calcId, false, null, null);
+
+ constructAnnotationFromScoreHolder(annotation, struct, data);
+
+ /*
+ * update annotation description with the free Energy, frequency in ensemble
+ * or other data where appropriate.
+ *
+ * Doesnt deal with AlifoldResult.ensembleValues, the free energy of
+ * ensemble and frequency of mfe structure in ensemble. How to deal with
+ * these?
+ */
+ annotation.description = description;
+
+ annotation.belowAlignment = false;
+ // annotation.showAllColLabels = true;
+
+ alignViewport.getAlignment().validateAnnotation(annotation);
+ af.setMenusForViewport();
+
+ ourAnnot.add(annotation);
+ }
+
+ private AlignmentAnnotation constructAnnotationFromScoreHolder(
+ AlignmentAnnotation annotation, String struct, TreeSet data)
+ {
+ Annotation[] anns = new Annotation[gapMap != null ? gapMap.length + 1
+ : struct.length()];
+
+ if (data != null
+ && data.size() > 1
+ && data.first().getMethod()
+ .equals(AlifoldResult.contactProbabilities.toString()))
+ {
+
+ // The base pair probabilities are stored in a set in scoreholder. we want
+ // a map
+ LinkedHashMap basePairs = new LinkedHashMap();
+ for (Score score : data)
+ {
+ // The Score objects contain a set of size one containing the range and
+ // an ArrayList of size one containing the probabilty
+ basePairs.put(score.getRanges().first(), new Float(score
+ .getScores().get(0)));
+ }
+
+ for (int i = 0, ri = 0, iEnd = struct.length(); i < iEnd; i++, ri++)
+ {
+ if (gapMap != null)
+ {
+ // skip any gapped columns in the input data
+ while (!gapMap[ri])
+ {
+ ri++;
+ }
+ }
+ // Return all the contacts associated with position i
+ LinkedHashMap contacts = isContact(basePairs, i + 1);
+
+ String description = "";
+ float prob = 0f;
+
+ if (contacts.size() == 0)
+ {
+ description = "No Data";
+ }
+ else
+ {
+ for (Range contact : contacts.keySet())
+ {
+ float t = contacts.get(contact);
+ if (t > prob)
+ {
+ prob = t;
+ }
+ description += Integer.toString(contact.from) + "->"
+ + Integer.toString(contact.to) + ": "
+ + Float.toString(t) + "% | ";
+ }
+ }
+
+ anns[ri] = new Annotation(struct.substring(i, i + 1), description,
+ isSS(struct.charAt(i)), prob);
+ }
+ }
+ else if (data == null || data.size() == 1)
+ {
+ for (int i = 0, ri = 0, iEnd = struct.length(); i < iEnd; i++, ri++)
+ {
+ if (gapMap != null)
+ {
+ // skip any gapped columns in the input data
+ while (!gapMap[ri] && ri < gapMap.length)
+ {
+ ri++;
+ }
+ if (ri == gapMap.length)
+ {
+ break;
+ }
+ }
+ anns[ri] = new Annotation(struct.substring(i, i + 1), "",
+ isSS(struct.charAt(i)), Float.NaN);
+ }
+
+ annotation.graph = 0; // No graph
+ }
+
+ annotation.annotations = anns;
+
+ return annotation;
+ }
+
+ private String[] constructTypenameAndDescription(Score score)
+ {
+ String description = "";
+ String typename = "";
+ String datatype = score.getMethod();
+
+ // Look up java switch syntax and use one here
+ if (datatype.equals(AlifoldResult.mfeStructure.toString()))
+ {
+
+ description = MessageFormat.format(
+ "Minimum Free Energy Structure. Energy: {0} = {1} + {2}",
+ score.getScores().get(0), score.getScores().get(1), score
+ .getScores().get(2));
+ typename = "MFE Structure";
+ }
+ else if (datatype.equals(AlifoldResult.contactProbabilityStructure
+ .toString()))
+ {
+ description = MessageFormat
+ .format("Base Pair Contact Probabilities. "
+ + "Energy of Ensemble: {0} Frequency of Ensemble: {1}",
+ score.getScores().get(0), score.getScores().get(1));
+ typename = "Contact Probabilities";
+ }
+ else if (datatype.equals(AlifoldResult.centroidStructure.toString()))
+ {
+ description = MessageFormat.format(
+ "Centroid Structure. Energy: {0} = {1} + {2}", score
+ .getScores().get(0), score.getScores().get(1), score
+ .getScores().get(2));
+ typename = "Centroid Structure";
+ }
+ else if (datatype.equals(AlifoldResult.stochBTStructure.toString()))
+ {
+ if (score.getScores().size() > 0)
+ {
+ description = MessageFormat.format("Probability: {0} Energy: {1}",
+ score.getScores().get(0), score.getScores().get(1));
+ }
+ else
+ {
+ description = "Stochastic Backtrack Structure";
+ }
+ }
+ else if (datatype.equals(AlifoldResult.MEAStucture.toString()))
+ {
+ description = MessageFormat.format(
+ "Maximum Expected Accuracy Values: '{' {0} MEA={1} '}", score
+ .getScores().get(0), score.getScores().get(1));
+ typename = "MEA Structure";
+ }
+ else if (datatype.equals(AlifoldResult.consensusAlignment.toString()))
+ {
+ typename = "RNAalifold Consensus";
+ description = "Consensus Alignment Produced by RNAalifold";
+ }
+ else
+ {
+ typename = datatype;
+ description = typename;
+ }
+
+ return new String[] { typename, description };
+ }
+
+ // Check whether, at position i there is a base contact and return all the
+ // contacts at this position. Should be in order of descending probability.
+ private LinkedHashMap isContact(
+ LinkedHashMap basePairs, int i)
+ {
+ LinkedHashMap contacts = new LinkedHashMap();
+
+ for (Range contact : basePairs.keySet())
+ {
+ // finds the contacts associtated with position i ordered by the natural
+ // ordering of the Scores TreeSet in ScoreManager which is, descending
+ // probability
+ if (contact.from == i || contact.to == i)
+ {
+ contacts.put(contact, basePairs.get(contact));
+ }
+ }
+
+ return contacts;
+ }
+
+ private char isSS(char chr)
{
- return SequenceAnnotationWSClient.AAConCalcId;
+ String regex = "\\(|\\)|\\{|\\}|\\[|\\]";
+ char ss = (Pattern.matches(regex, Character.toString(chr))) ? 'S' : ' ';
+ return ss;
}
-
}