X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Fanalysis%2FAAFrequency.java;h=61f3b7f3b391024ce2b5355975b143871ad60f3f;hb=990ed4ffbaa7a95b2ebb6bf6ab0440310f6e83ab;hp=132abc0e3a6c7e5fc092b1a1fd79dcc08152275c;hpb=55fe3787a3eda4a3fec091c3b6435012ad7e6bf3;p=jalview.git
diff --git a/src/jalview/analysis/AAFrequency.java b/src/jalview/analysis/AAFrequency.java
index 132abc0..61f3b7f 100755
--- a/src/jalview/analysis/AAFrequency.java
+++ b/src/jalview/analysis/AAFrequency.java
@@ -20,10 +20,6 @@
*/
package jalview.analysis;
-import java.util.Arrays;
-import java.util.Hashtable;
-import java.util.List;
-
import jalview.datamodel.AlignedCodonFrame;
import jalview.datamodel.AlignmentAnnotation;
import jalview.datamodel.AlignmentI;
@@ -41,6 +37,11 @@ import jalview.util.Format;
import jalview.util.MappingUtils;
import jalview.util.QuickSort;
+import java.awt.Color;
+import java.util.Arrays;
+import java.util.Hashtable;
+import java.util.List;
+
/**
* Takes in a vector or array of sequences and column start and column end and
* returns a new Hashtable[] of size maxSeqLength, if Hashtable not supplied.
@@ -146,14 +147,13 @@ public class AAFrequency
{
if (sequences[row] == null)
{
- System.err
- .println("WARNING: Consensus skipping null sequence - possible race condition.");
+ System.err.println(
+ "WARNING: Consensus skipping null sequence - possible race condition.");
continue;
}
- char[] seq = sequences[row].getSequence();
- if (seq.length > column)
+ if (sequences[row].getLength() > column)
{
- char c = seq[column];
+ char c = sequences[row].getCharAt(column);
residueCounts.add(c);
if (Comparison.isNucleotide(c))
{
@@ -291,7 +291,7 @@ public class AAFrequency
/**
* Derive the gap count annotation row.
*
- * @param consensus
+ * @param gaprow
* the annotation row to add annotations to
* @param profiles
* the source consensus data
@@ -300,11 +300,11 @@ public class AAFrequency
* @param endCol
* end column (exclusive)
*/
- public static void completeGapAnnot(AlignmentAnnotation consensus,
+ public static void completeGapAnnot(AlignmentAnnotation gaprow,
ProfilesI profiles, int startCol, int endCol, long nseq)
{
- if (consensus == null || consensus.annotations == null
- || consensus.annotations.length < endCol)
+ if (gaprow == null || gaprow.annotations == null
+ || gaprow.annotations.length < endCol)
{
/*
* called with a bad alignment annotation row
@@ -313,8 +313,9 @@ public class AAFrequency
return;
}
// always set ranges again
- consensus.graphMax = nseq;
- consensus.graphMin = 0;
+ gaprow.graphMax = nseq;
+ gaprow.graphMin = 0;
+ double scale = 0.8 / nseq;
for (int i = startCol; i < endCol; i++)
{
ProfileI profile = profiles.get(i);
@@ -324,16 +325,17 @@ public class AAFrequency
* happens if sequences calculated over were
* shorter than alignment width
*/
- consensus.annotations[i] = null;
+ gaprow.annotations[i] = null;
return;
}
final int gapped = profile.getNonGapped();
- String description = String.valueOf(gapped);
+ String description = "" + gapped;
- consensus.annotations[i] = new Annotation("", description, '0',
- gapped);
+ gaprow.annotations[i] = new Annotation("", description, '\0', gapped,
+ jalview.util.ColorUtils.bleachColour(Color.DARK_GRAY,
+ (float) scale * gapped));
}
}
@@ -342,7 +344,8 @@ public class AAFrequency
*
* - the full profile (percentages of all residues present), if
* showSequenceLogo is true, or
- * - just the modal (most common) residue(s), if showSequenceLogo is false
+ * - just the modal (most common) residue(s), if showSequenceLogo is
+ * false
*
* Percentages are as a fraction of all sequence, or only ungapped sequences
* if ignoreGaps is true.
@@ -363,8 +366,8 @@ public class AAFrequency
String description = null;
if (counts != null && showSequenceLogo)
{
- int normaliseBy = ignoreGaps ? profile.getNonGapped() : profile
- .getHeight();
+ int normaliseBy = ignoreGaps ? profile.getNonGapped()
+ : profile.getHeight();
description = counts.getTooltip(normaliseBy, dp);
}
else
@@ -395,7 +398,7 @@ public class AAFrequency
* contains
*
*
- * [profileType, numberOfValues, nonGapCount, charValue1, percentage1, charValue2, percentage2, ...]
+ * [profileType, numberOfValues, totalPercent, charValue1, percentage1, charValue2, percentage2, ...]
* in descending order of percentage value
*
*
@@ -408,7 +411,6 @@ public class AAFrequency
*/
public static int[] extractProfile(ProfileI profile, boolean ignoreGaps)
{
- int[] rtnval = new int[64];
ResidueCount counts = profile.getCounts();
if (counts == null)
{
@@ -419,29 +421,51 @@ public class AAFrequency
char[] symbols = symbolCounts.symbols;
int[] values = symbolCounts.values;
QuickSort.sort(values, symbols);
- int nextArrayPos = 2;
int totalPercentage = 0;
- final int divisor = ignoreGaps ? profile.getNonGapped() : profile
- .getHeight();
+ final int divisor = ignoreGaps ? profile.getNonGapped()
+ : profile.getHeight();
/*
* traverse the arrays in reverse order (highest counts first)
*/
+ int[] result = new int[3 + 2 * symbols.length];
+ int nextArrayPos = 3;
+ int nonZeroCount = 0;
+
for (int i = symbols.length - 1; i >= 0; i--)
{
int theChar = symbols[i];
int charCount = values[i];
-
- rtnval[nextArrayPos++] = theChar;
final int percentage = (charCount * 100) / divisor;
- rtnval[nextArrayPos++] = percentage;
+ if (percentage == 0)
+ {
+ /*
+ * this count (and any remaining) round down to 0% - discard
+ */
+ break;
+ }
+ nonZeroCount++;
+ result[nextArrayPos++] = theChar;
+ result[nextArrayPos++] = percentage;
totalPercentage += percentage;
}
- rtnval[0] = symbols.length;
- rtnval[1] = totalPercentage;
- int[] result = new int[rtnval.length + 1];
+
+ /*
+ * truncate array if any zero values were discarded
+ */
+ if (nonZeroCount < symbols.length)
+ {
+ int[] tmp = new int[3 + 2 * nonZeroCount];
+ System.arraycopy(result, 0, tmp, 0, tmp.length);
+ result = tmp;
+ }
+
+ /*
+ * fill in 'header' values
+ */
result[0] = AlignmentAnnotation.SEQUENCE_PROFILE;
- System.arraycopy(rtnval, 0, result, 1, rtnval.length);
+ result[1] = nonZeroCount;
+ result[2] = totalPercentage;
return result;
}
@@ -451,14 +475,15 @@ public class AAFrequency
* contains
*
*
- * [profileType, numberOfValues, totalCount, charValue1, percentage1, charValue2, percentage2, ...]
+ * [profileType, numberOfValues, totalPercentage, charValue1, percentage1, charValue2, percentage2, ...]
* in descending order of percentage value, where the character values encode codon triplets
*
*
* @param hashtable
* @return
*/
- public static int[] extractCdnaProfile(Hashtable hashtable,
+ public static int[] extractCdnaProfile(
+ Hashtable hashtable,
boolean ignoreGaps)
{
// this holds #seqs, #ungapped, and then codon count, indexed by encoded
@@ -489,9 +514,16 @@ public class AAFrequency
{
break; // nothing else of interest here
}
+ final int percentage = codonCount * 100 / divisor;
+ if (percentage == 0)
+ {
+ /*
+ * this (and any remaining) values rounded down to 0 - discard
+ */
+ break;
+ }
distinctValuesCount++;
result[j++] = codons[i];
- final int percentage = codonCount * 100 / divisor;
result[j++] = percentage;
totalPercentage += percentage;
}
@@ -515,7 +547,7 @@ public class AAFrequency
* the consensus data stores to be populated (one per column)
*/
public static void calculateCdna(AlignmentI alignment,
- Hashtable[] hconsensus)
+ Hashtable[] hconsensus)
{
final char gapCharacter = alignment.getGapCharacter();
List mappings = alignment.getCodonFrames();
@@ -528,7 +560,7 @@ public class AAFrequency
for (int col = 0; col < cols; col++)
{
// todo would prefer a Java bean for consensus data
- Hashtable columnHash = new Hashtable();
+ Hashtable columnHash = new Hashtable<>();
// #seqs, #ungapped seqs, counts indexed by (codon encoded + 1)
int[] codonCounts = new int[66];
codonCounts[0] = alignment.getSequences().size();
@@ -539,8 +571,8 @@ public class AAFrequency
{
continue;
}
- List codons = MappingUtils
- .findCodonsFor(seq, col, mappings);
+ List codons = MappingUtils.findCodonsFor(seq, col,
+ mappings);
for (char[] codon : codons)
{
int codonEncoded = CodingUtils.encodeCodon(codon);
@@ -573,7 +605,8 @@ public class AAFrequency
*/
public static void completeCdnaConsensus(
AlignmentAnnotation consensusAnnotation,
- Hashtable[] consensusData, boolean showProfileLogo, int nseqs)
+ Hashtable[] consensusData, boolean showProfileLogo,
+ int nseqs)
{
if (consensusAnnotation == null
|| consensusAnnotation.annotations == null
@@ -588,7 +621,7 @@ public class AAFrequency
consensusAnnotation.scaleColLabel = true;
for (int col = 0; col < consensusData.length; col++)
{
- Hashtable hci = consensusData[col];
+ Hashtable hci = consensusData[col];
if (hci == null)
{
// gapped protein column?
@@ -620,10 +653,10 @@ public class AAFrequency
int modalCodonEncoded = codons[codons.length - 1];
int modalCodonCount = sortedCodonCounts[codons.length - 1];
- String modalCodon = String.valueOf(CodingUtils
- .decodeCodon(modalCodonEncoded));
- if (sortedCodonCounts.length > 1
- && sortedCodonCounts[codons.length - 2] == sortedCodonCounts[codons.length - 1])
+ String modalCodon = String
+ .valueOf(CodingUtils.decodeCodon(modalCodonEncoded));
+ if (sortedCodonCounts.length > 1 && sortedCodonCounts[codons.length
+ - 2] == sortedCodonCounts[codons.length - 1])
{
/*
* two or more codons share the modal count
@@ -682,8 +715,8 @@ public class AAFrequency
{
if (samePercent.length() > 0)
{
- mouseOver.append(samePercent).append(": ")
- .append(lastPercent).append("% ");
+ mouseOver.append(samePercent).append(": ").append(lastPercent)
+ .append("% ");
}
samePercent.setLength(0);
samePercent.append(codon);