/*
* 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
* 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.analysis;
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 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";
public static final String MAXRESIDUE = "R";
public static final String PID_GAPS = "G";
public static final String PID_NOGAPS = "N";
public static final String PROFILE = "P";
public static final String PAIRPROFILE = "B";
/**
* Returns the 3' position of a base pair
*
* @param pairs
* Secondary structure annotation
* @param indice
* 5' position of a base pair
* @return 3' position of a base pair
*/
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;
}
/**
* 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
* @param result
* @param profile
* @param rnaStruc
*/
public static final void calculate(SequenceI[] sequences, int start,
int end, Hashtable[] result, boolean profile,
AlignmentAnnotation rnaStruc)
{
Hashtable residueHash;
String maxResidue;
char[] struc = rnaStruc.getRNAStruc().toCharArray();
SequenceFeature[] rna = rnaStruc._rnasecstr;
char c, s, cEnd;
int bpEnd = -1;
int jSize = sequences.length;
int[] values;
int[][] pairs;
float percentage;
for (int i = start; i < end; i++) // foreach column
{
int canonicalOrWobblePairCount = 0, canonical = 0;
int otherPairCount = 0;
int nongap = 0;
maxResidue = "-";
values = new int[255];
pairs = new int[255][255];
bpEnd = -1;
if (i < struc.length)
{
s = struc[i];
}
else
{
s = '-';
}
if (s == '.' || s == ' ')
{
s = '-';
}
if (!Rna.isOpeningParenthesis(s))
{
if (s == '-')
{
values['-']++;
}
}
else
{
bpEnd = findPair(rna, i);
if (bpEnd > -1)
{
for (int j = 0; j < jSize; j++) // foreach row
{
if (sequences[j] == null)
{
System.err.println(
"WARNING: Consensus skipping null sequence - possible race condition.");
continue;
}
c = sequences[j].getCharAt(i);
cEnd = sequences[j].getCharAt(bpEnd);
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))
{
canonical++;
}
}
else
{
otherPairCount++;
}
pairs[c][cEnd]++;
}
}
}
residueHash = new Hashtable<>();
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);
}
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, Float.valueOf(percentage));
percentage = ((float) count * 100) / nongap;
residueHash.put(PID_NOGAPS, Float.valueOf(percentage));
if (result[i] == null)
{
result[i] = residueHash;
}
if (bpEnd > 0)
{
values[')'] = values['('];
values[']'] = values['['];
values['}'] = values['{'];
values['('] = 0;
values['['] = 0;
values['{'] = 0;
maxResidue = maxResidue.equals("(") ? ")"
: maxResidue.equals("[") ? "]" : "}";
residueHash = new Hashtable<>();
if (profile)
{
residueHash.put(PROFILE,
new int[][]
{ values, new int[] { jSize, (jSize - values['-']) } });
residueHash.put(PAIRPROFILE, pairs);
}
residueHash.put(MAXCOUNT, Integer.valueOf(count));
residueHash.put(MAXRESIDUE, maxResidue);
percentage = ((float) count * 100) / jSize;
residueHash.put(PID_GAPS, Float.valueOf(percentage));
percentage = ((float) count * 100) / nongap;
residueHash.put(PID_NOGAPS, Float.valueOf(percentage));
result[bpEnd] = residueHash;
}
}
}
/**
* Compute all or part of the annotation row from the given consensus
* hashtable
*
* @param consensus
* - pre-allocated annotation row
* @param hconsensus
* @param iStart
* @param width
* @param ignoreGapsInConsensusCalculation
* @param includeAllConsSymbols
*/
public static void completeConsensus(AlignmentAnnotation consensus,
Hashtable[] hconsensus, int iStart, int width,
boolean ignoreGapsInConsensusCalculation,
boolean includeAllConsSymbols, long nseq)
{
float tval, value;
if (consensus == null || consensus.annotations == null
|| consensus.annotations.length < width)
{
// called with a bad alignment annotation row - wait for it to be
// 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))
{
// happens if sequences calculated over were shorter than alignment
// width
consensus.annotations[i] = null;
continue;
}
value = 0;
Float fv;
if (ignoreGapsInConsensusCalculation)
{
fv = (Float) hci.get(StructureFrequency.PID_NOGAPS);
}
else
{
fv = (Float) hci.get(StructureFrequency.PID_GAPS);
}
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) + " ";
if (maxRes.length() > 1)
{
mouseOver = "[" + maxRes + "] ";
maxRes = "+";
}
int[][] profile = (int[][]) hci.get(StructureFrequency.PROFILE);
int[][] pairs = (int[][]) hci.get(StructureFrequency.PAIRPROFILE);
if (pairs != null && includeAllConsSymbols) // Just responsible for the
// tooltip
// TODO Update tooltips for Structure row
{
mouseOver = "";
/*
* 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 {
*/
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 };
vl[x] = pairs[c][d];
x++;
}
}
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 / divisor);
mouseOver += ((p == 0) ? "" : "; ") + (char) ca[c][0]
+ (char) ca[c][1] + " " + fmt.form(tval) + "%";
p++;
}
}
// }
}
else
{
mouseOver += (fmt.form(value) + "%");
}
consensus.annotations[i] = new Annotation(maxRes, mouseOver, ' ',
value);
}
}
/**
* 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[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];
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 };
vl[x] = pairs[c][d];
x++;
}
}
jalview.util.QuickSort.sort(vl, ca);
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[offset++] = ca[c][0];
rtnval[offset++] = ca[c][1];
rtnval[offset] = (int) (vl[c] * 100f / divisor);
rtnval[1] += rtnval[offset++];
valuesCount++;
}
}
rtnval[0] = valuesCount;
// 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;
}
}