X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Fanalysis%2FStructureFrequency.java;h=c04df6c9ac56d71b71adcb1aacd73a36b92c7bd7;hb=4f77328104498504339216829abf5ea87e2791ec;hp=33a08e679bf2e3d5e69937b087d9bf347941077d;hpb=b57a02c25e335d033c97f8a6bacd6b54f62bd2b6;p=jalview.git
diff --git a/src/jalview/analysis/StructureFrequency.java b/src/jalview/analysis/StructureFrequency.java
index 33a08e6..c04df6c 100644
--- a/src/jalview/analysis/StructureFrequency.java
+++ b/src/jalview/analysis/StructureFrequency.java
@@ -1,26 +1,33 @@
/*
- * Jalview - A Sequence Alignment Editor and Viewer (Version 2.7)
- * Copyright (C) 2011 J Procter, AM Waterhouse, J Engelhardt, LM Lui, G Barton, M Clamp, S Searle
- *
+ * Jalview - A Sequence Alignment Editor and Viewer ($$Version-Rel$$)
+ * Copyright (C) $$Year-Rel$$ 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
+ * 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 .
+ *
+ * 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.analysis;
-import java.util.*;
+import jalview.datamodel.AlignmentAnnotation;
+import jalview.datamodel.Annotation;
+import jalview.datamodel.SequenceFeature;
+import jalview.datamodel.SequenceI;
+import jalview.util.Comparison;
+import jalview.util.Format;
-import jalview.datamodel.*;
+import java.util.Hashtable;
/**
* Takes in a vector or array of sequences and column start and column end and
@@ -33,6 +40,8 @@ import jalview.datamodel.*;
*/
public class StructureFrequency
{
+ public static final int STRUCTURE_PROFILE_LENGTH = 74;
+
// No need to store 1000s of strings which are not
// visible to the user.
public static final String MAXCOUNT = "C";
@@ -58,11 +67,15 @@ public class StructureFrequency
*/
public static int findPair(SequenceFeature[] pairs, int indice)
{
+
for (int i = 0; i < pairs.length; i++)
{
if (pairs[i].getBegin() == indice)
+
{
+
return pairs[i].getEnd();
+
}
}
return -1;
@@ -80,22 +93,27 @@ public class StructureFrequency
* @param rnaStruc
*/
public static final void calculate(SequenceI[] sequences, int start,
- int end, Hashtable[] result, boolean profile,
+ int end, Hashtable[] result, boolean profile,
AlignmentAnnotation rnaStruc)
{
- Hashtable residueHash;
+
+ Hashtable residueHash;
String maxResidue;
- char[] seq, struc = rnaStruc.getRNAStruc().toCharArray();
+ char[] struc = rnaStruc.getRNAStruc().toCharArray();
+
SequenceFeature[] rna = rnaStruc._rnasecstr;
char c, s, cEnd;
- int count, nonGap = 0, i, bpEnd = -1, j, jSize = sequences.length;
+ int bpEnd = -1;
+ int jSize = sequences.length;
int[] values;
int[][] pairs;
float percentage;
- for (i = start; i < end; i++) // foreach column
+ for (int i = start; i < end; i++) // foreach column
{
- residueHash = new Hashtable();
+ int canonicalOrWobblePairCount = 0, canonical = 0;
+ int otherPairCount = 0;
+ int nongap = 0;
maxResidue = "-";
values = new int[255];
pairs = new int[255][255];
@@ -113,7 +131,7 @@ public class StructureFrequency
s = '-';
}
- if (s != '(')
+ if (!Rna.isOpeningParenthesis(s))
{
if (s == '-')
{
@@ -123,63 +141,100 @@ public class StructureFrequency
else
{
bpEnd = findPair(rna, i);
+
if (bpEnd > -1)
{
- for (j = 0; j < jSize; j++) // foreach row
+ for (int j = 0; j < jSize; j++) // foreach row
{
if (sequences[j] == null)
{
- System.err
- .println("WARNING: Consensus skipping null sequence - possible race condition.");
+ System.err.println(
+ "WARNING: Consensus skipping null sequence - possible race condition.");
continue;
}
- c = sequences[j].getCharAt(i);
- {
- // standard representation for gaps in sequence and structure
- if (c == '.' || c == ' ')
- {
- c = '-';
- }
+ c = sequences[j].getCharAt(i);
+ cEnd = sequences[j].getCharAt(bpEnd);
- if (c == '-')
- {
- values['-']++;
- continue;
- }
- cEnd = sequences[j].getCharAt(bpEnd);
- if (checkBpType(c, cEnd))
+ if (Comparison.isGap(c) || Comparison.isGap(cEnd))
+ {
+ values['-']++;
+ continue;
+ }
+ nongap++;
+ /*
+ * ensure upper-case for counting purposes
+ */
+ if ('a' <= c && 'z' >= c)
+ {
+ c += 'A' - 'a';
+ }
+ if ('a' <= cEnd && 'z' >= cEnd)
+ {
+ cEnd += 'A' - 'a';
+ }
+ if (Rna.isCanonicalOrWobblePair(c, cEnd))
+ {
+ canonicalOrWobblePairCount++;
+ if (Rna.isCanonicalPair(c, cEnd))
{
- values['(']++; // H means it's a helix (structured)
+ canonical++;
}
- pairs[c][cEnd]++;
-
- maxResidue = "(";
}
+ else
+ {
+ otherPairCount++;
+ }
+ pairs[c][cEnd]++;
}
}
- // nonGap++;
}
- // UPDATE this for new values
+
+ residueHash = new Hashtable<>();
if (profile)
{
- residueHash.put(PROFILE, new int[][]
- { values, new int[]
- { jSize, (jSize - values['-']) } });
+ // TODO 1-dim array with jsize in [0], nongapped in [1]; or Pojo
+ residueHash.put(PROFILE,
+ new int[][]
+ { values, new int[] { jSize, (jSize - values['-']) } });
residueHash.put(PAIRPROFILE, pairs);
}
-
- count = values['('];
-
- residueHash.put(MAXCOUNT, new Integer(count));
+ values['('] = canonicalOrWobblePairCount;
+ values['['] = canonical;
+ values['{'] = otherPairCount;
+ /*
+ * the count is the number of valid pairs (as a percentage, determines
+ * the relative size of the profile logo)
+ */
+ int count = canonicalOrWobblePairCount;
+
+ /*
+ * display '(' if most pairs are canonical, or as
+ * '[' if there are more wobble pairs.
+ */
+ if (canonicalOrWobblePairCount > 0 || otherPairCount > 0)
+ {
+ if (canonicalOrWobblePairCount >= otherPairCount)
+ {
+ maxResidue = (canonicalOrWobblePairCount - canonical) < canonical
+ ? "("
+ : "[";
+ }
+ else
+ {
+ maxResidue = "{";
+ }
+ }
+ residueHash.put(MAXCOUNT, Integer.valueOf(count));
residueHash.put(MAXRESIDUE, maxResidue);
percentage = ((float) count * 100) / jSize;
- residueHash.put(PID_GAPS, new Float(percentage));
+ residueHash.put(PID_GAPS, Float.valueOf(percentage));
+
+ percentage = ((float) count * 100) / nongap;
+ residueHash.put(PID_NOGAPS, Float.valueOf(percentage));
- // percentage = ((float) count * 100) / (float) nongap;
- // residueHash.put(PID_NOGAPS, new Float(percentage));
if (result[i] == null)
{
result[i] = residueHash;
@@ -187,100 +242,36 @@ public class StructureFrequency
if (bpEnd > 0)
{
values[')'] = values['('];
+ values[']'] = values['['];
+ values['}'] = values['{'];
values['('] = 0;
+ values['['] = 0;
+ values['{'] = 0;
+ maxResidue = maxResidue.equals("(") ? ")"
+ : maxResidue.equals("[") ? "]" : "}";
- residueHash = new Hashtable();
- maxResidue = ")";
-
+ residueHash = new Hashtable<>();
if (profile)
{
- residueHash.put(PROFILE, new int[][]
- { values, new int[]
- { jSize, (jSize - values['-']) } });
+ residueHash.put(PROFILE,
+ new int[][]
+ { values, new int[] { jSize, (jSize - values['-']) } });
residueHash.put(PAIRPROFILE, pairs);
}
- residueHash.put(MAXCOUNT, new Integer(count));
+ residueHash.put(MAXCOUNT, Integer.valueOf(count));
residueHash.put(MAXRESIDUE, maxResidue);
percentage = ((float) count * 100) / jSize;
- residueHash.put(PID_GAPS, new Float(percentage));
+ residueHash.put(PID_GAPS, Float.valueOf(percentage));
- result[bpEnd] = residueHash;
- }
- }
- }
-
- /**
- * Method to check if a base-pair is a canonical or a wobble bp
- *
- * @param up
- * 5' base
- * @param down
- * 3' base
- * @return True if it is a canonical/wobble bp
- */
- public static boolean checkBpType(char up, char down)
- {
- if (up > 'Z')
- {
- up -= 32;
- }
- if (down > 'Z')
- {
- down -= 32;
- }
+ percentage = ((float) count * 100) / nongap;
+ residueHash.put(PID_NOGAPS, Float.valueOf(percentage));
- switch (up)
- {
- case 'A':
- switch (down)
- {
- case 'T':
- return true;
- case 'U':
- return true;
- }
- break;
- case 'C':
- switch (down)
- {
- case 'G':
- return true;
- }
- break;
- case 'T':
- switch (down)
- {
- case 'A':
- return true;
- case 'G':
- return true;
- }
- break;
- case 'G':
- switch (down)
- {
- case 'C':
- return true;
- case 'T':
- return true;
- case 'U':
- return true;
- }
- break;
- case 'U':
- switch (down)
- {
- case 'A':
- return true;
- case 'G':
- return true;
+ result[bpEnd] = residueHash;
}
- break;
}
- return false;
}
/**
@@ -296,9 +287,9 @@ public class StructureFrequency
* @param includeAllConsSymbols
*/
public static void completeConsensus(AlignmentAnnotation consensus,
- Hashtable[] hconsensus, int iStart, int width,
+ Hashtable[] hconsensus, int iStart, int width,
boolean ignoreGapsInConsensusCalculation,
- boolean includeAllConsSymbols)
+ boolean includeAllConsSymbols, long nseq)
{
float tval, value;
if (consensus == null || consensus.annotations == null
@@ -308,9 +299,22 @@ public class StructureFrequency
// initialised properly
return;
}
+ String fmtstr = "%3.1f";
+ int precision = 2;
+ while (nseq > 100)
+ {
+ precision++;
+ nseq /= 10;
+ }
+ if (precision > 2)
+ {
+ fmtstr = "%" + (2 + precision) + "." + precision + "f";
+ }
+ Format fmt = new Format(fmtstr);
+
for (int i = iStart; i < width; i++)
{
- Hashtable hci;
+ Hashtable hci;
if (i >= hconsensus.length || ((hci = hconsensus[i]) == null))
{
// happens if sequences calculated over were shorter than alignment
@@ -361,15 +365,14 @@ public class StructureFrequency
* ((c == 0) ? "" : "; ") + alphabet[c] + " " + ((int) tval) + "%"; } }
* else {
*/
- Object[] ca = new Object[625];
+ int[][] ca = new int[625][];
float[] vl = new float[625];
int x = 0;
for (int c = 65; c < 90; c++)
{
for (int d = 65; d < 90; d++)
{
- ca[x] = new int[]
- { c, d };
+ ca[x] = new int[] { c, d };
vl[x] = pairs[c][d];
x++;
}
@@ -377,14 +380,18 @@ public class StructureFrequency
jalview.util.QuickSort.sort(vl, ca);
int p = 0;
+ /*
+ * profile[1] is {total, ungappedTotal}
+ */
+ final int divisor = profile[1][ignoreGapsInConsensusCalculation ? 1
+ : 0];
for (int c = 624; c > 0; c--)
{
if (vl[c] > 0)
{
- tval = (vl[c] * 100f / profile[1][ignoreGapsInConsensusCalculation ? 1
- : 0]);
- mouseOver += ((p == 0) ? "" : "; ") + (char) ((int[]) ca[c])[0]
- + (char) ((int[]) ca[c])[1] + " " + ((int) tval) + "%";
+ tval = (vl[c] * 100f / divisor);
+ mouseOver += ((p == 0) ? "" : "; ") + (char) ca[c][0]
+ + (char) ca[c][1] + " " + fmt.form(tval) + "%";
p++;
}
@@ -394,7 +401,7 @@ public class StructureFrequency
}
else
{
- mouseOver += ((int) value + "%");
+ mouseOver += (fmt.form(value) + "%");
}
consensus.annotations[i] = new Annotation(maxRes, mouseOver, ' ',
value);
@@ -407,66 +414,57 @@ public class StructureFrequency
* @param hconsensus
* @return profile of the given column
*/
- public static int[] extractProfile(Hashtable hconsensus,
+ public static int[] extractProfile(Hashtable hconsensus,
boolean ignoreGapsInConsensusCalculation)
{
- int[] rtnval = new int[74]; // 2*(5*5)+2
+ int[] rtnval = new int[STRUCTURE_PROFILE_LENGTH]; // 2*(5*5)+2
int[][] profile = (int[][]) hconsensus.get(StructureFrequency.PROFILE);
int[][] pairs = (int[][]) hconsensus
.get(StructureFrequency.PAIRPROFILE);
if (profile == null)
+ {
return null;
+ }
// TODO fix the object length, also do it in completeConsensus
- Object[] ca = new Object[625];
+ // Object[] ca = new Object[625];
+ int[][] ca = new int[625][];
float[] vl = new float[625];
int x = 0;
for (int c = 65; c < 90; c++)
{
for (int d = 65; d < 90; d++)
{
- ca[x] = new int[]
- { c, d };
+ ca[x] = new int[] { c, d };
vl[x] = pairs[c][d];
x++;
}
}
jalview.util.QuickSort.sort(vl, ca);
- rtnval[0] = 2;
+ int valuesCount = 0;
rtnval[1] = 0;
+ int offset = 2;
+ final int divisor = profile[1][ignoreGapsInConsensusCalculation ? 1
+ : 0];
for (int c = 624; c > 0; c--)
{
if (vl[c] > 0)
{
- rtnval[rtnval[0]++] = ((int[]) ca[c])[0];
- rtnval[rtnval[0]++] = ((int[]) ca[c])[1];
- rtnval[rtnval[0]] = (int) (vl[c] * 100f / profile[1][ignoreGapsInConsensusCalculation ? 1
- : 0]);
- rtnval[1] += rtnval[rtnval[0]++];
+ rtnval[offset++] = ca[c][0];
+ rtnval[offset++] = ca[c][1];
+ rtnval[offset] = (int) (vl[c] * 100f / divisor);
+ rtnval[1] += rtnval[offset++];
+ valuesCount++;
}
}
+ rtnval[0] = valuesCount;
- return rtnval;
- }
-
- public static void main(String args[])
- {
- // Short test to see if checkBpType works
- ArrayList test = new ArrayList();
- test.add("A");
- test.add("c");
- test.add("g");
- test.add("T");
- test.add("U");
- for (String i : test)
- {
- for (String j : test)
- {
- System.out.println(i + "-" + j + ": "
- + StructureFrequency.checkBpType(i.charAt(0), j.charAt(0)));
- }
- }
+ // insert profile type code in position 0
+ int[] result = new int[rtnval.length + 1];
+ result[0] = AlignmentAnnotation.STRUCTURE_PROFILE;
+ System.arraycopy(rtnval, 0, result, 1, rtnval.length);
+ return result;
}
}