From 09436561a5b39d37a51056a61b440adc0de95ae5 Mon Sep 17 00:00:00 2001
From: tpylak <tpylak>
Date: Fri, 17 Jul 2009 12:24:59 +0000
Subject: [PATCH] LMS-1028 break sequence into many shorter lines, mark
peptides on the sequence
SVN: 11813
---
.../client/web/client/application/Dict.java | 2 -
.../client/application/OccurrencesMarker.java | 139 ++++++++++--------
.../web/client/application/ProteinViewer.java | 40 ++---
.../application/OccurrencesMarkerTest.java | 64 ++++++--
4 files changed, 136 insertions(+), 109 deletions(-)
diff --git a/rtd_phosphonetx/source/java/ch/systemsx/cisd/openbis/plugin/phosphonetx/client/web/client/application/Dict.java b/rtd_phosphonetx/source/java/ch/systemsx/cisd/openbis/plugin/phosphonetx/client/web/client/application/Dict.java
index d882974f053..41abf3dcb77 100644
--- a/rtd_phosphonetx/source/java/ch/systemsx/cisd/openbis/plugin/phosphonetx/client/web/client/application/Dict.java
+++ b/rtd_phosphonetx/source/java/ch/systemsx/cisd/openbis/plugin/phosphonetx/client/web/client/application/Dict.java
@@ -52,8 +52,6 @@ public class Dict
public static final String DATA_SET_PERM_ID = "data_set_perm_id";
- public static final String PEPTIDE = "peptide";
-
public static final String PEPTIDE_COUNT = "peptide_count";
public static final String SEQUENCE_NAME = "sequence_name";
diff --git a/rtd_phosphonetx/source/java/ch/systemsx/cisd/openbis/plugin/phosphonetx/client/web/client/application/OccurrencesMarker.java b/rtd_phosphonetx/source/java/ch/systemsx/cisd/openbis/plugin/phosphonetx/client/web/client/application/OccurrencesMarker.java
index 27d0ba462b7..cdc2b47c85a 100644
--- a/rtd_phosphonetx/source/java/ch/systemsx/cisd/openbis/plugin/phosphonetx/client/web/client/application/OccurrencesMarker.java
+++ b/rtd_phosphonetx/source/java/ch/systemsx/cisd/openbis/plugin/phosphonetx/client/web/client/application/OccurrencesMarker.java
@@ -17,51 +17,74 @@
package ch.systemsx.cisd.openbis.plugin.phosphonetx.client.web.client.application;
import java.util.ArrayList;
-import java.util.Arrays;
import java.util.Collections;
import java.util.List;
/**
* This class is able to find all occurrences of the given set of words in the provided template and
- * mark the beginning and end of those words. Cases when occurrences overlap are handles properly.
+ * mark the beginning and end of those words. Note that when occurrences overlap the overlaps are
+ * merged and the border between them is not visualized.
*
* @author Tomasz Pylak
*/
public class OccurrencesMarker
{
- private final String startMarker;
+ private final char startMarker;
- private final String endMarker;
+ private final char endMarker;
/** Produces an HTML code whih all occurrences properly marked */
- public static String markOccurrencesWithHtml(String template, List<String> words)
+ public static String markOccurrencesWithHtml(String template, List<String> words, int lineLength)
+ {
+ char start = '(';
+ char end = ')';
+ OccurrencesMarker marker = new OccurrencesMarker(start, end);
+ String markedTemplate = marker.mark(template, words);
+ markedTemplate = breakLines(markedTemplate, lineLength, "<br>");
+ markedTemplate = marker.replaceTags(markedTemplate, "<font color='red'>", "</font>");
+ return markedTemplate;
+ }
+
+ // @Private
+ String replaceTags(String text, String startTag, String endTag)
+ {
+ String newText = text;
+ newText = newText.replaceAll("\\" + startMarker, startTag);
+ newText = newText.replaceAll("\\" + endMarker, endTag);
+ return newText;
+ }
+
+ /** split lines, so that each line has lineLength characters at most */
+ static String breakLines(String text, int lineLength, String endOfLine)
{
- String start = "(";
- String end = ")";
- List<String> markedTemplates = new OccurrencesMarker(start, end).mark(template, words);
StringBuffer sb = new StringBuffer();
- for (String markedTemplate : markedTemplates)
+ String textToBreak = text;
+ while (textToBreak.length() > lineLength)
{
- markedTemplate.replaceAll(start, "<font color='red'>");
- markedTemplate.replaceAll(end, "</font>");
- sb.append(markedTemplate);
- sb.append("<BR>");
+ String line = textToBreak.substring(0, lineLength);
+ sb.append(line);
+ sb.append(endOfLine);
+ textToBreak = textToBreak.substring(lineLength);
}
+ sb.append(textToBreak);
return sb.toString();
}
- public OccurrencesMarker(String startMarker, String endMarker)
+ public OccurrencesMarker(char startMarker, char endMarker)
{
this.startMarker = startMarker;
this.endMarker = endMarker;
}
/**
- * If the words occurrences do not overlap in the template, the returned list has only one
- * element. Otherwise there are several elements, each contains non-overlapping marks. skips
- * words which do not occur in the template.
+ * All letters which belong to the matching words are marked. If some words overlap, the markers
+ * are merged. E.g. for the template 'xabcx' and words 'ab' and 'bc', the result will be
+ * 'x(abc)x'.
+ * <p>
+ * Words which do not occur in the template are ignored.
+ * </p>
*/
- public List<String> mark(String template, List<String> words)
+ public String mark(String template, List<String> words)
{
List<Occurrence> occurrences = calcSortedOccurrences(template, words);
if (hasOverlapping(occurrences))
@@ -69,7 +92,7 @@ public class OccurrencesMarker
return markOverlapping(template, occurrences);
} else
{
- return Arrays.asList(markNonoverlapping(template, occurrences));
+ return markNonoverlapping(template, occurrences);
}
}
@@ -118,64 +141,57 @@ public class OccurrencesMarker
{
return getStartIndex() - o.getStartIndex();
}
- }
- private List<String> markOverlapping(String template, List<Occurrence> sortedOccurrences)
- {
- List<List<Occurrence>> distinctOccurrencesList = splitToDistinctGroups(sortedOccurrences);
- List<String> result = new ArrayList<String>();
- for (List<Occurrence> distinctOccurrences : distinctOccurrencesList)
+ @Override
+ public String toString()
{
- String marked = markNonoverlapping(template, distinctOccurrences);
- result.add(marked);
+ return "[" + word + "@" + startIndex + "]";
}
- return result;
}
- // Splits all occurrences into many groups in such a way, that occurrences in one group do not
- // overlap with each other.
- private static List<List<Occurrence>> splitToDistinctGroups(List<Occurrence> sortedOccurrences)
+ private String markOverlapping(String template, List<Occurrence> sortedOccurrences)
{
- setVisitedFlag(sortedOccurrences, false);
- List<List<Occurrence>> result = new ArrayList<List<Occurrence>>();
- int unvisited = sortedOccurrences.size();
- while (unvisited > 0)
- {
- List<Occurrence> distinctOccurrences =
- chooseDistinctUnvisitedOccurrences(sortedOccurrences);
- result.add(distinctOccurrences);
-
- setVisitedFlag(distinctOccurrences, true);
- unvisited -= distinctOccurrences.size();
- }
- return result;
+ List<Occurrence> mergedOccurrences = mergeOverlaps(template, sortedOccurrences);
+ return markNonoverlapping(template, mergedOccurrences);
}
- private static List<Occurrence> chooseDistinctUnvisitedOccurrences(
- List<Occurrence> sortedOccurrences)
+ private List<Occurrence> mergeOverlaps(String template, List<Occurrence> sortedOccurrences)
{
- List<Occurrence> distinctOccurrences = new ArrayList<Occurrence>();
- int lastIncludedCharIndex = -1;
+ List<Occurrence> result = new ArrayList<Occurrence>();
+ if (sortedOccurrences.size() == 0)
+ {
+ return result;
+ }
+ int startIndex = -1;
+ int endIndex = -1;
for (Occurrence occurrence : sortedOccurrences)
{
- if (occurrence.isVisited() == false)
- {
- if (occurrence.getStartIndex() > lastIncludedCharIndex)
- {
- distinctOccurrences.add(occurrence);
- lastIncludedCharIndex = occurrence.getEndIndex();
+ if (occurrence.getStartIndex() <= endIndex)
+ { // overlap
+ endIndex = Math.max(endIndex, occurrence.getEndIndex());
+ } else
+ { // current word does not overlap with the words browsed before
+ if (startIndex != -1)
+ { // create a new word from the words browsed before
+ Occurrence newOccurrence = createOccurence(template, startIndex, endIndex);
+ result.add(newOccurrence);
}
+ startIndex = occurrence.getStartIndex();
+ endIndex = occurrence.getEndIndex();
}
}
- return distinctOccurrences;
+ Occurrence newOccurrence = createOccurence(template, startIndex, endIndex);
+ result.add(newOccurrence);
+ return result;
}
- private static void setVisitedFlag(List<Occurrence> occurrences, boolean visited)
+ private static Occurrence createOccurence(String template, int startIndex, int endIndex)
{
- for (Occurrence occurrence : occurrences)
- {
- occurrence.setVisited(visited);
- }
+ assert startIndex != -1 : "start index should be initialized";
+ assert endIndex != -1 : "end index should be initialized";
+ String mergedWord = template.substring(startIndex, endIndex + 1);
+ Occurrence newOccurrence = new Occurrence(mergedWord, startIndex);
+ return newOccurrence;
}
// marks all occurrences in the template, assuming that all occurrences do not
@@ -207,7 +223,7 @@ public class OccurrencesMarker
{
return true;
}
- prevEndIndex = occurrence.getEndIndex();
+ prevEndIndex = Math.max(prevEndIndex, occurrence.getEndIndex());
}
return false;
}
@@ -240,5 +256,4 @@ public class OccurrencesMarker
}
return result;
}
-
}
diff --git a/rtd_phosphonetx/source/java/ch/systemsx/cisd/openbis/plugin/phosphonetx/client/web/client/application/ProteinViewer.java b/rtd_phosphonetx/source/java/ch/systemsx/cisd/openbis/plugin/phosphonetx/client/web/client/application/ProteinViewer.java
index 6f6174c7286..f6bde440444 100644
--- a/rtd_phosphonetx/source/java/ch/systemsx/cisd/openbis/plugin/phosphonetx/client/web/client/application/ProteinViewer.java
+++ b/rtd_phosphonetx/source/java/ch/systemsx/cisd/openbis/plugin/phosphonetx/client/web/client/application/ProteinViewer.java
@@ -17,7 +17,6 @@
package ch.systemsx.cisd.openbis.plugin.phosphonetx.client.web.client.application;
import java.util.ArrayList;
-import java.util.Arrays;
import java.util.LinkedHashMap;
import java.util.List;
import java.util.Map;
@@ -61,6 +60,8 @@ public class ProteinViewer extends
AbstractViewer<IPhosphoNetXClientServiceAsync, IEntityInformationHolder> implements
IDatabaseModificationObserver
{
+ private static final int AMINOACIDS_IN_SEQUENCE_PER_LINE = 60;
+
private static final String PREFIX = "protein-viewer_";
public static final String ID_PREFIX = GenericConstants.ID_PREFIX + PREFIX;
@@ -124,8 +125,6 @@ public class ProteinViewer extends
private void recreateUI(ProteinByExperiment protein)
{
- // fillDebugData(protein);
-
setLayout(new BorderLayout());
removeAll();
setScrollMode(Scroll.AUTO);
@@ -143,28 +142,6 @@ public class ProteinViewer extends
}
}
- // TODO 2009-07-17, Tomasz Pylak: remove me!!!!!!!!!!!!!!!!!!!!!!!!!
- private static void fillDebugData(ProteinByExperiment protein)
- {
- ProteinDetails details = new ProteinDetails();
- details.setDatabaseNameAndVersion("database");
- details
- .setSequence("ISDFHSJDGJHFGHHJKDGHJGBGBKJBGSHDFYUERGFUYEWBGCHJEBVHJERBCVUYERBFYUEWBCYUWERBCUYEBCUYEBR");
- details.setDataSetPermID("20090716105149429-34");
- details.setDataSetTechID(15L);
- details.setDataSetTypeCode("PROT_RESULT");
- details.setFalseDiscoveryRate(123);
-
- // - more than once
- // - overlapping
- List<Peptide> peptides =
- Arrays.asList(createPeptide("SJDGJHFGHHJ"), createPeptide("FUYEWBGCHJEBVH"),
- createPeptide("YERBFYUEWBCYU"),
- createPeptide("ERBCVUYERBFYUEWBCYUWERBCUYEB"));
- details.setPeptides(peptides);
- protein.setDetails(details);
- }
-
private static Peptide createPeptide(String sequence)
{
Peptide peptide = new Peptide();
@@ -223,12 +200,11 @@ public class ProteinViewer extends
{
properties.put(viewContext.getMessage(Dict.DATABASE_NAME_AND_VERSION), proteinDetails
.getDatabaseNameAndVersion());
- properties.put(viewContext.getMessage(Dict.SEQUENCE_NAME), proteinDetails.getSequence());
- properties.put(viewContext.getMessage(Dict.PEPTIDE_COUNT), proteinDetails.getPeptides()
- .size());
String markedSequence =
markPeptides(proteinDetails.getSequence(), proteinDetails.getPeptides());
- properties.put(viewContext.getMessage(Dict.PEPTIDE), markedSequence);
+ properties.put(viewContext.getMessage(Dict.SEQUENCE_NAME), markedSequence);
+ properties.put(viewContext.getMessage(Dict.PEPTIDE_COUNT), proteinDetails.getPeptides()
+ .size());
properties.put(viewContext.getMessage(Dict.FDR), proteinDetails.getFalseDiscoveryRate());
@@ -241,7 +217,11 @@ public class ProteinViewer extends
private static String markPeptides(String sequence, List<Peptide> peptides)
{
List<String> peptideSequences = extractSequences(peptides);
- return OccurrencesMarker.markOccurrencesWithHtml(sequence, peptideSequences);
+ String markedSequence =
+ OccurrencesMarker.markOccurrencesWithHtml(sequence, peptideSequences,
+ AMINOACIDS_IN_SEQUENCE_PER_LINE);
+ // the letters should have fixed width
+ return "<font style=\"font-family:monospace\">" + markedSequence + "</font>";
}
private static List<String> extractSequences(List<Peptide> peptides)
diff --git a/rtd_phosphonetx/sourceTest/java/ch/systemsx/cisd/openbis/plugin/phosphonetx/client/web/client/application/OccurrencesMarkerTest.java b/rtd_phosphonetx/sourceTest/java/ch/systemsx/cisd/openbis/plugin/phosphonetx/client/web/client/application/OccurrencesMarkerTest.java
index 1f58f89ae6b..0986301308f 100644
--- a/rtd_phosphonetx/sourceTest/java/ch/systemsx/cisd/openbis/plugin/phosphonetx/client/web/client/application/OccurrencesMarkerTest.java
+++ b/rtd_phosphonetx/sourceTest/java/ch/systemsx/cisd/openbis/plugin/phosphonetx/client/web/client/application/OccurrencesMarkerTest.java
@@ -17,43 +17,77 @@
package ch.systemsx.cisd.openbis.plugin.phosphonetx.client.web.client.application;
import java.util.Arrays;
-import java.util.List;
import org.testng.AssertJUnit;
import org.testng.annotations.Test;
+import ch.rinn.restrictions.Friend;
+
/**
* @author Tomasz Pylak
*/
+@Friend(toClasses = OccurrencesMarker.class)
public class OccurrencesMarkerTest extends AssertJUnit
{
@Test
public void testFindNonOverlappingOccurrences()
{
- OccurrencesMarker marker = new OccurrencesMarker("<", ">");
- List<String> marked =
- marker.mark("hello, my beautiful world!", Arrays.asList("hello", "world"));
- assertEquals(1, marked.size());
- assertEquals("<hello>, my beautiful <world>!", marked.get(0));
+ OccurrencesMarker marker = createMarker();
+ String marked = marker.mark("hello, my beautiful world!", Arrays.asList("hello", "world"));
+ assertEquals("<hello>, my beautiful <world>!", marked);
+ }
+
+ private OccurrencesMarker createMarker()
+ {
+ return new OccurrencesMarker('<', '>');
}
@Test
public void testFindOverlappingOccurrences()
{
- OccurrencesMarker marker = new OccurrencesMarker("<", ">");
- List<String> marked = marker.mark("aaaa", Arrays.asList("aa"));
- assertEquals(2, marked.size());
- assertEquals("<aa><aa>", marked.get(0));
- assertEquals("a<aa>a", marked.get(1));
+ OccurrencesMarker marker = createMarker();
+ String marked = marker.mark("aaaa", Arrays.asList("aa"));
+ assertEquals("<aaaa>", marked);
}
@Test
public void testFindNoOccurrences()
{
- OccurrencesMarker marker = new OccurrencesMarker("<", ">");
- List<String> marked = marker.mark("aaaa", Arrays.asList("x"));
- assertEquals(1, marked.size());
- assertEquals("aaaa", marked.get(0));
+ OccurrencesMarker marker = createMarker();
+ String marked = marker.mark("aaaa", Arrays.asList("x"));
+ assertEquals("aaaa", marked);
+ }
+
+ @Test
+ public void testFindContainingOccurrences()
+ {
+ OccurrencesMarker marker = createMarker();
+ String marked = marker.mark("xabcx", Arrays.asList("abc", "b"));
+ assertEquals("x<abc>x", marked);
+ }
+
+ @Test
+ public void testBreakLinesLastLineNotFull()
+ {
+ String lines = OccurrencesMarker.breakLines("1234567", 3, "x");
+ assertEquals("123x456x7", lines);
+ }
+
+ @Test
+ public void testBreakLinesLastLineFull()
+ {
+ String lines = OccurrencesMarker.breakLines("123456", 3, "x");
+ assertEquals("123x456", lines);
+ }
+
+ @Test
+ public void testReplaceTags()
+ {
+ String marked = new OccurrencesMarker('(', ')').replaceTags("a(a)a", "<", ">");
+ assertEquals("a<a>a", marked);
+
+ marked = new OccurrencesMarker('<', '>').replaceTags("a<a>a", "(", ")");
+ assertEquals("a(a)a", marked);
}
}
--
GitLab