import jalview.datamodel.ProfilesI;
import jalview.datamodel.ResidueCount;
import jalview.datamodel.ResidueCount.SymbolCounts;
+import jalview.datamodel.SecondaryStructureCount;
import jalview.datamodel.SequenceI;
import jalview.ext.android.SparseIntArray;
import jalview.util.Comparison;
public class AAFrequency
{
public static final String PROFILE = "P";
+ private static final String SS_ANNOTATION_LABEL = "Secondary Structure";
+ private static final char COIL = 'C';
/*
* Quick look-up of String value of char 'A' to 'Z'
// jalview.bin.Console.outPrintln(elapsed);
}
+
+ public static final ProfilesI calculateSS(List<SequenceI> list, int start,
+ int end)
+ {
+ return calculateSS(list, start, end, false);
+ }
+
+ public static final ProfilesI calculateSS(List<SequenceI> sequences,
+ int start, int end, boolean profile)
+ {
+ SequenceI[] seqs = new SequenceI[sequences.size()];
+ int width = 0;
+ synchronized (sequences)
+ {
+ for (int i = 0; i < sequences.size(); i++)
+ {
+ seqs[i] = sequences.get(i);
+ int length = seqs[i].getLength();
+ if (length > width)
+ {
+ width = length;
+ }
+ }
+
+ if (end >= width)
+ {
+ end = width;
+ }
+
+ ProfilesI reply = calculateSS(seqs, width, start, end, profile);
+ return reply;
+ }
+ }
+
+ public static final ProfilesI calculateSS(final SequenceI[] sequences,
+ int width, int start, int end, boolean saveFullProfile)
+ {
+ // long now = System.currentTimeMillis();
+ int seqCount = sequences.length;
+
+ ProfileI[] result = new ProfileI[width];
+
+ for (int column = start; column < end; column++)
+ {
+ /*
+ * Apply a heuristic to detect nucleotide data (which can
+ * be counted in more compact arrays); here we test for
+ * more than 90% nucleotide; recheck every 10 columns in case
+ * of misleading data e.g. highly conserved Alanine in peptide!
+ * Mistakenly guessing nucleotide has a small performance cost,
+ * as it will result in counting in sparse arrays.
+ * Mistakenly guessing peptide has a small space cost,
+ * as it will use a larger than necessary array to hold counts.
+ */
+
+ int ssCount = 0;
+
+ SecondaryStructureCount ssCounts = new SecondaryStructureCount();
+
+ for (int row = 0; row < seqCount; row++)
+ {
+ if (sequences[row] == null)
+ {
+ jalview.bin.Console.errPrintln(
+ "WARNING: Consensus skipping null sequence - possible race condition.");
+ continue;
+ }
+
+ char c = sequences[row].getCharAt(column);
+
+ if (sequences[row].getLength() > column && !Comparison.isGap(c))
+ {
+
+ AlignmentAnnotation[] aa = sequences[row].getAnnotation(SS_ANNOTATION_LABEL);
+ if(aa == null) {
+ continue;
+ }
+ int seqPosition = sequences[row].findPosition(column);
+ char ss;
+ if (aa[0].getAnnotationForPosition(seqPosition) != null) {
+ ss = aa[0].getAnnotationForPosition(seqPosition).secondaryStructure;
+
+ //There is no representation for coil and it can be either ' ' or null.
+ if (ss == ' ') {
+ ss = COIL;
+ }
+ }
+ else {
+ ss = COIL;
+ }
+
+ //secondaryStructures[row][column] = ss;
+
+ ssCounts.add(ss);
+ ssCount++;
+
+ }
+ else
+ {
+ /*
+ * count a gap if the sequence doesn't reach this column
+ */
+ ssCounts.addGap();
+ }
+ }
+
+ int maxSSCount = ssCounts.getModalCount();
+ String maxSS = ssCounts.getSSForCount(maxSSCount);
+ int gapCount = ssCounts.getGapCount();
+ ProfileI profile = new Profile(maxSS, ssCount, gapCount,
+ maxSSCount);
+
+ if (saveFullProfile)
+ {
+ profile.setSSCounts(ssCounts);
+ }
+
+ result[column] = profile;
+ }
+ return new Profiles(result);
+ // long elapsed = System.currentTimeMillis() - now;
+ // jalview.bin.Console.outPrintln(elapsed);
+ }
+
/**
* Make an estimate of the profile size we are going to compute i.e. how many
* different characters may be present in it. Overestimating has a cost of
// long elapsed = System.currentTimeMillis() - now;
// jalview.bin.Console.outPrintln(-elapsed);
}
+
+
+ public static void completeSSConsensus(AlignmentAnnotation ssConsensus,
+ ProfilesI profiles, int startCol, int endCol, boolean ignoreGaps,
+ boolean showSequenceLogo, long nseq)
+ {
+ // long now = System.currentTimeMillis();
+ if (ssConsensus == null || ssConsensus.annotations == null
+ || ssConsensus.annotations.length < endCol)
+ {
+ /*
+ * called with a bad alignment annotation row
+ * wait for it to be initialised properly
+ */
+ return;
+ }
+
+ for (int i = startCol; i < endCol; i++)
+ {
+ ProfileI profile = profiles.get(i);
+ if (profile == null)
+ {
+ /*
+ * happens if sequences calculated over were
+ * shorter than alignment width
+ */
+ ssConsensus.annotations[i] = null;
+ return;
+ }
+
+ final int dp = getPercentageDp(nseq);
+
+ float value = profile.getSSPercentageIdentity(ignoreGaps);
+
+ String description = getSSTooltip(profile, value, showSequenceLogo,
+ ignoreGaps, dp);
+
+ String modalSS = profile.getModalSS();
+ if ("".equals(modalSS))
+ {
+ modalSS = "-";
+ }
+ else if (modalSS.length() > 1)
+ {
+ modalSS = "+";
+ }
+ ssConsensus.annotations[i] = new Annotation(modalSS, description,
+ ' ', value);
+ }
+ // long elapsed = System.currentTimeMillis() - now;
+ // jalview.bin.Console.outPrintln(-elapsed);
+ }
/**
* Derive the gap count annotation row.
}
return description;
}
+
+ static String getSSTooltip(ProfileI profile, float pid,
+ boolean showSequenceLogo, boolean ignoreGaps, int dp)
+ {
+ SecondaryStructureCount counts = profile.getSSCounts();
+
+ String description = null;
+ if (counts != null && showSequenceLogo)
+ {
+ int normaliseBy = ignoreGaps ? profile.getNonGapped()
+ : profile.getHeight();
+ description = counts.getTooltip(normaliseBy, dp);
+ }
+ else
+ {
+ StringBuilder sb = new StringBuilder(64);
+ String maxSS = profile.getModalSS();
+ if (maxSS.length() > 1)
+ {
+ sb.append("[").append(maxSS).append("]");
+ }
+ else
+ {
+ sb.append(maxSS);
+ }
+ if (maxSS.length() > 0)
+ {
+ sb.append(" ");
+ Format.appendPercentage(sb, pid, dp);
+ sb.append("%");
+ }
+ description = sb.toString();
+ }
+ return description;
+ }
/**
* Returns the sorted profile for the given consensus data. The returned array
*/
public static int[] extractProfile(ProfileI profile, boolean ignoreGaps)
{
- ResidueCount counts = profile.getCounts();
- if (counts == null)
+ char[] symbols;
+ int[] values;
+
+ if (profile.getCounts() != null)
{
+ ResidueCount counts = profile.getCounts();
+ SymbolCounts symbolCounts = counts.getSymbolCounts();
+ symbols = symbolCounts.symbols;
+ values = symbolCounts.values;
+
+ }
+ else if(profile.getSSCounts() != null)
+ {
+ SecondaryStructureCount counts = profile.getSSCounts();
+ // to do
+ SecondaryStructureCount.SymbolCounts symbolCounts = counts.getSymbolCounts();
+ symbols = symbolCounts.symbols;
+ values = symbolCounts.values;
+ }
+ else {
return null;
}
+
- SymbolCounts symbolCounts = counts.getSymbolCounts();
- char[] symbols = symbolCounts.symbols;
- int[] values = symbolCounts.values;
QuickSort.sort(values, symbols);
int totalPercentage = 0;
final int divisor = ignoreGaps ? profile.getNonGapped()