X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Fanalysis%2FStructureFrequency.java;h=3c8274e3ba190dfcf28a687e9c9b4b8f7eaca7a7;hb=d07fe6a0891b57a9d707e356f7769395ce94b0da;hp=5095ce1e5ffe3f8fe5fe3a9fb738842c31a5bac4;hpb=d7c00f48b00e3ede57c46ae4daf6a9203b9b6059;p=jalview.git
diff --git a/src/jalview/analysis/StructureFrequency.java b/src/jalview/analysis/StructureFrequency.java
index 5095ce1..3c8274e 100644
--- a/src/jalview/analysis/StructureFrequency.java
+++ b/src/jalview/analysis/StructureFrequency.java
@@ -1,38 +1,47 @@
/*
- * 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.Format;
-import jalview.datamodel.*;
+import java.util.ArrayList;
+import java.util.Hashtable;
/**
* 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.
* This class is used extensively in calculating alignment colourschemes that
* depend on the amount of conservation in each alignment column.
- *
+ *
* @author $author$
* @version $Revision$
*/
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";
@@ -49,7 +58,7 @@ public class StructureFrequency
/**
* Returns the 3' position of a base pair
- *
+ *
* @param pairs
* Secondary structure annotation
* @param indice
@@ -58,15 +67,15 @@ public class StructureFrequency
*/
public static int findPair(SequenceFeature[] pairs, int indice)
{
- System.out.print("indice"+indice+" ");
+
for (int i = 0; i < pairs.length; i++)
{
if (pairs[i].getBegin() == indice)
-
+
{
- System.out.println(pairs[i].getEnd());
+
return pairs[i].getEnd();
-
+
}
}
return -1;
@@ -75,7 +84,7 @@ public class StructureFrequency
/**
* Method to calculate a 'base pair consensus row', very similar to nucleotide
* consensus but takes into account a given structure
- *
+ *
* @param sequences
* @param start
* @param end
@@ -87,19 +96,11 @@ public class StructureFrequency
int end, Hashtable[] result, boolean profile,
AlignmentAnnotation rnaStruc)
{
-// System.out.println("longueur="+sequences.length);
-// for(int l=0;l<=(sequences.length-1);l++){
-// System.out.println("sequences "+l+":"+sequences[l].getSequenceAsString());
-// }
-// System.out.println("start="+start);
- System.out.println("end="+end);
-// System.out.println("result="+result.length);
-//
-// System.out.println("profile="+profile);
-// System.out.println("rnaStruc="+rnaStruc);
+
Hashtable residueHash;
String maxResidue;
char[] struc = rnaStruc.getRNAStruc().toCharArray();
+
SequenceFeature[] rna = rnaStruc._rnasecstr;
char c, s, cEnd;
int count = 0, nonGap = 0, i, bpEnd = -1, j, jSize = sequences.length;
@@ -107,7 +108,6 @@ public class StructureFrequency
int[][] pairs;
float percentage;
boolean wooble = true;
-
for (i = start; i < end; i++) // foreach column
{
residueHash = new Hashtable();
@@ -115,11 +115,11 @@ public class StructureFrequency
values = new int[255];
pairs = new int[255][255];
bpEnd = -1;
- //System.out.println("s="+struc[i]);
+ // System.out.println("s="+struc[i]);
if (i < struc.length)
{
s = struc[i];
-
+
}
else
{
@@ -139,23 +139,21 @@ public class StructureFrequency
}
else
{
-
-
+
bpEnd = findPair(rna, i);
-
- if (bpEnd>-1)
- {
- for (j = 0; j < jSize; j++) // foreach row
+
+ if (bpEnd > -1)
{
- if (sequences[j] == null)
+ for (j = 0; j < jSize; j++) // foreach row
{
- System.err
- .println("WARNING: Consensus skipping null sequence - possible race condition.");
- continue;
- }
- c = sequences[j].getCharAt(i);
- //System.out.println("c="+c);
-
+ if (sequences[j] == null)
+ {
+ System.err
+ .println("WARNING: Consensus skipping null sequence - possible race condition.");
+ continue;
+ }
+ c = sequences[j].getCharAt(i);
+ // System.out.println("c="+c);
// standard representation for gaps in sequence and structure
if (c == '.' || c == ' ')
@@ -169,49 +167,46 @@ public class StructureFrequency
continue;
}
cEnd = sequences[j].getCharAt(bpEnd);
-
-
- System.out.println("pairs ="+c+","+cEnd);
- if (checkBpType(c, cEnd)==true)
+
+ // System.out.println("pairs ="+c+","+cEnd);
+ if (checkBpType(c, cEnd) == true)
{
values['(']++; // H means it's a helix (structured)
maxResidue = "(";
- wooble=true;
- System.out.println("It's a pair wc");
-
+ wooble = true;
+ // System.out.println("It's a pair wc");
+
}
- if (checkBpType(c, cEnd)==false)
+ if (checkBpType(c, cEnd) == false)
{
- wooble =false;
+ wooble = false;
values['[']++; // H means it's a helix (structured)
maxResidue = "[";
- System.out.println("It's an pair non canonic");
- System.out.println(sequences[j].getRNA());
- System.out.println(rnaStruc.getRNAStruc().charAt(i));
+
}
- pairs[c][cEnd]++;
-
-
- }
+ pairs[c][cEnd]++;
+
+ }
}
// nonGap++;
}
// UPDATE this for new values
if (profile)
{
+ // 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);
}
- if (wooble==true)
+ if (wooble == true)
{
- count = values['('];
+ count = values['('];
}
- if (wooble==false)
+ if (wooble == false)
{
- count = values['['];
+ count = values['['];
}
residueHash.put(MAXCOUNT, new Integer(count));
residueHash.put(MAXRESIDUE, maxResidue);
@@ -227,18 +222,20 @@ public class StructureFrequency
}
if (bpEnd > 0)
{
- values[')'] = values['('];
+ values[')'] = values['('];
values[']'] = values['['];
values['('] = 0;
values['['] = 0;
residueHash = new Hashtable();
- if (wooble==true){
- System.out.println(maxResidue+","+wooble);
- maxResidue = ")";
+ if (wooble == true)
+ {
+ // System.out.println(maxResidue+","+wooble);
+ maxResidue = ")";
}
- if(wooble==false){
- System.out.println(maxResidue+","+wooble);
- maxResidue = "]";
+ if (wooble == false)
+ {
+ // System.out.println(maxResidue+","+wooble);
+ maxResidue = "]";
}
if (profile)
{
@@ -256,14 +253,14 @@ public class StructureFrequency
residueHash.put(PID_GAPS, new Float(percentage));
result[bpEnd] = residueHash;
-
+
}
}
}
/**
* Method to check if a base-pair is a canonical or a wobble bp
- *
+ *
* @param up
* 5' base
* @param down
@@ -335,7 +332,7 @@ public class StructureFrequency
/**
* Compute all or part of the annotation row from the given consensus
* hashtable
- *
+ *
* @param consensus
* - pre-allocated annotation row
* @param hconsensus
@@ -347,7 +344,7 @@ public class StructureFrequency
public static void completeConsensus(AlignmentAnnotation consensus,
Hashtable[] hconsensus, int iStart, int width,
boolean ignoreGapsInConsensusCalculation,
- boolean includeAllConsSymbols)
+ boolean includeAllConsSymbols, long nseq)
{
float tval, value;
if (consensus == null || consensus.annotations == null
@@ -357,10 +354,23 @@ 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;
- if (i >= hconsensus.length || ((hci=hconsensus[i])==null))
+ if (i >= hconsensus.length || ((hci = hconsensus[i]) == null))
{
// happens if sequences calculated over were shorter than alignment
// width
@@ -371,32 +381,28 @@ public class StructureFrequency
Float fv;
if (ignoreGapsInConsensusCalculation)
{
- fv =(Float) hci.get(StructureFrequency.PID_NOGAPS);
+ fv = (Float) hci.get(StructureFrequency.PID_NOGAPS);
}
else
{
fv = (Float) hci.get(StructureFrequency.PID_GAPS);
}
- if (fv==null)
+ if (fv == null)
{
consensus.annotations[i] = null;
// data has changed below us .. give up and
continue;
}
value = fv.floatValue();
- String maxRes = hci.get(StructureFrequency.MAXRESIDUE)
- .toString();
- String mouseOver = hci.get(StructureFrequency.MAXRESIDUE)
- + " ";
+ String maxRes = hci.get(StructureFrequency.MAXRESIDUE).toString();
+ String mouseOver = hci.get(StructureFrequency.MAXRESIDUE) + " ";
if (maxRes.length() > 1)
{
mouseOver = "[" + maxRes + "] ";
maxRes = "+";
}
- int[][] profile = (int[][]) hci
- .get(StructureFrequency.PROFILE);
- int[][] pairs = (int[][]) hci
- .get(StructureFrequency.PAIRPROFILE);
+ int[][] profile = (int[][]) hci.get(StructureFrequency.PROFILE);
+ int[][] pairs = (int[][]) hci.get(StructureFrequency.PAIRPROFILE);
if (pairs != null && includeAllConsSymbols) // Just responsible for the
// tooltip
@@ -404,15 +410,17 @@ public class StructureFrequency
{
mouseOver = "";
- /* TODO It's not sure what is the purpose of the alphabet and wheter it is useful for structure?
- *
+ /*
+ * TODO It's not sure what is the purpose of the alphabet and wheter it
+ * is useful for structure?
+ *
* if (alphabet != null) { for (int c = 0; c < alphabet.length; c++) {
* tval = ((float) profile[0][alphabet[c]]) 100f / (float)
* profile[1][ignoreGapsInConsensusCalculation ? 1 : 0]; mouseOver +=
* ((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++)
@@ -428,14 +436,19 @@ 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++;
}
@@ -445,7 +458,7 @@ public class StructureFrequency
}
else
{
- mouseOver += ((int) value + "%");
+ mouseOver += (fmt.form(value) + "%");
}
consensus.annotations[i] = new Annotation(maxRes, mouseOver, ' ',
value);
@@ -454,23 +467,26 @@ public class StructureFrequency
/**
* get the sorted base-pair profile for the given position of the consensus
- *
+ *
* @param hconsensus
* @return profile of the given column
*/
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++)
@@ -485,21 +501,28 @@ public class StructureFrequency
}
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;
+ // 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;
}
public static void main(String args[])