*/
package jalview.analysis;
-import jalview.analysis.scoremodels.PIDModel;
-import jalview.api.analysis.DistanceScoreModelI;
import jalview.api.analysis.ScoreModelI;
import jalview.api.analysis.SimilarityParamsI;
-import jalview.api.analysis.SimilarityScoreModelI;
+import jalview.bin.Cache;
import jalview.datamodel.AlignmentView;
+import jalview.datamodel.Point;
import jalview.math.MatrixI;
import java.io.PrintStream;
*/
public class PCA implements Runnable
{
- MatrixI symm;
+ /*
+ * inputs
+ */
+ final private AlignmentView seqs;
- double[] eigenvalue;
+ final private ScoreModelI scoreModel;
- MatrixI eigenvector;
+ final private SimilarityParamsI similarityParams;
- StringBuilder details = new StringBuilder(1024);
+ /*
+ * outputs
+ */
+ private MatrixI pairwiseScores;
- private AlignmentView seqs;
+ private MatrixI tridiagonal;
- private ScoreModelI scoreModel;
-
- private SimilarityParamsI similarityParams;
+ private MatrixI eigenMatrix;
- public PCA(AlignmentView s, ScoreModelI sm, SimilarityParamsI options)
+ /**
+ * Constructor given the sequences to compute for, the similarity model to
+ * use, and a set of parameters for sequence comparison
+ *
+ * @param sequences
+ * @param sm
+ * @param options
+ */
+ public PCA(AlignmentView sequences, ScoreModelI sm, SimilarityParamsI options)
{
- this.seqs = s;
- this.similarityParams = options;
+ this.seqs = sequences;
this.scoreModel = sm;
-
- details.append("PCA calculation using " + sm.getName()
- + " sequence similarity matrix\n========\n\n");
+ this.similarityParams = options;
}
/**
*/
public double getEigenvalue(int i)
{
- return eigenvector.getD()[i];
+ return eigenMatrix.getD()[i];
}
/**
*
* @return DOCUMENT ME!
*/
- public float[][] getComponents(int l, int n, int mm, float factor)
+ public Point[] getComponents(int l, int n, int mm, float factor)
{
- float[][] out = new float[getHeight()][3];
+ Point[] out = new Point[getHeight()];
for (int i = 0; i < getHeight(); i++)
{
- out[i][0] = (float) component(i, l) * factor;
- out[i][1] = (float) component(i, n) * factor;
- out[i][2] = (float) component(i, mm) * factor;
+ float x = (float) component(i, l) * factor;
+ float y = (float) component(i, n) * factor;
+ float z = (float) component(i, mm) * factor;
+ out[i] = new Point(x, y, z);
}
return out;
{
double out = 0.0;
- for (int i = 0; i < symm.width(); i++)
+ for (int i = 0; i < pairwiseScores.width(); i++)
{
- out += (symm.getValue(row, i) * eigenvector.getValue(i, n));
+ out += (pairwiseScores.getValue(row, i) * eigenMatrix.getValue(i, n));
}
- return out / eigenvector.getD()[n];
+ return out / eigenMatrix.getD()[n];
}
+ /**
+ * Answers a formatted text report of the PCA calculation results (matrices
+ * and eigenvalues) suitable for display
+ *
+ * @return
+ */
public String getDetails()
{
- return details.toString();
+ StringBuilder sb = new StringBuilder(1024);
+ sb.append("PCA calculation using ").append(scoreModel.getName())
+ .append(" sequence similarity matrix\n========\n\n");
+ PrintStream ps = wrapOutputBuffer(sb);
+
+ /*
+ * pairwise similarity scores
+ */
+ sb.append(" --- OrigT * Orig ---- \n");
+ pairwiseScores.print(ps, "%8.2f");
+
+ /*
+ * tridiagonal matrix, with D and E vectors
+ */
+ sb.append(" ---Tridiag transform matrix ---\n");
+ sb.append(" --- D vector ---\n");
+ tridiagonal.printD(ps, "%15.4e");
+ ps.println();
+ sb.append("--- E vector ---\n");
+ tridiagonal.printE(ps, "%15.4e");
+ ps.println();
+
+ /*
+ * eigenvalues matrix, with D vector
+ */
+ sb.append(" --- New diagonalization matrix ---\n");
+ eigenMatrix.print(ps, "%8.2f");
+ sb.append(" --- Eigenvalues ---\n");
+ eigenMatrix.printD(ps, "%15.4e");
+ ps.println();
+
+ return sb.toString();
}
/**
- * DOCUMENT ME!
+ * Performs the PCA calculation
*/
@Override
public void run()
{
- PrintStream ps = new PrintStream(System.out)
- {
- @Override
- public void print(String x)
- {
- details.append(x);
- }
-
- @Override
- public void println()
- {
- details.append("\n");
- }
- };
-
- // long now = System.currentTimeMillis();
try
{
- eigenvector = computeSimilarity(seqs);
-
- details.append(" --- OrigT * Orig ---- \n");
- eigenvector.print(ps, "%8.2f");
-
- symm = eigenvector.copy();
-
- eigenvector.tred();
+ /*
+ * sequence pairwise similarity scores
+ */
+ pairwiseScores = scoreModel.findSimilarities(seqs, similarityParams);
- details.append(" ---Tridiag transform matrix ---\n");
- details.append(" --- D vector ---\n");
- eigenvector.printD(ps, "%15.4e");
- ps.println();
- details.append("--- E vector ---\n");
- eigenvector.printE(ps, "%15.4e");
- ps.println();
+ /*
+ * tridiagonal matrix
+ */
+ tridiagonal = pairwiseScores.copy();
+ tridiagonal.tred();
- // Now produce the diagonalization matrix
- eigenvector.tqli();
+ /*
+ * the diagonalization matrix
+ */
+ eigenMatrix = tridiagonal.copy();
+ eigenMatrix.tqli();
} catch (Exception q)
{
+ Cache.log.error("Error computing PCA: " + q.getMessage());
q.printStackTrace();
- details.append("\n*** Unexpected exception when performing PCA ***\n"
- + q.getLocalizedMessage());
- details.append("*** Matrices below may not be fully diagonalised. ***\n");
}
-
- details.append(" --- New diagonalization matrix ---\n");
- eigenvector.print(ps, "%8.2f");
- details.append(" --- Eigenvalues ---\n");
- eigenvector.printD(ps, "%15.4e");
- ps.println();
- /*
- * for (int seq=0;seq<symm.rows;seq++) { ps.print("\"Seq"+seq+"\""); for
- * (int ev=0;ev<symm.rows; ev++) {
- *
- * ps.print(","+component(seq, ev)); } ps.println(); }
- */
- // System.out.println(("PCA.run() took "
- // + (System.currentTimeMillis() - now) + "ms"));
}
/**
- * Computes a pairwise similarity matrix for the given sequence regions using
- * the configured score model. If the score model is a similarity model, then
- * it computes the result directly. If it is a distance model, then use it to
- * compute pairwise distances, and convert these to similarity scores.
+ * Returns a PrintStream that wraps (appends its output to) the given
+ * StringBuilder
*
- * @param av
+ * @param sb
* @return
*/
- MatrixI computeSimilarity(AlignmentView av)
+ protected PrintStream wrapOutputBuffer(StringBuilder sb)
{
- MatrixI result = null;
- if (scoreModel instanceof SimilarityScoreModelI)
+ PrintStream ps = new PrintStream(System.out)
{
- result = ((SimilarityScoreModelI) scoreModel).findSimilarities(av,
- similarityParams);
- if (scoreModel instanceof PIDModel)
+ @Override
+ public void print(String x)
{
- /*
- * scale % identities to width of alignment for backwards
- * compatibility with Jalview 2.10.1 SeqSpace PCA calculation
- */
- result.multiply(av.getWidth() / 100d);
+ sb.append(x);
}
- }
- else if (scoreModel instanceof DistanceScoreModelI)
- {
- /*
- * find distances and convert to similarity scores
- * reverseRange(false) preserves but reverses the min-max range
- */
- result = ((DistanceScoreModelI) scoreModel).findDistances(av,
- similarityParams);
- result.reverseRange(false);
- }
- else
- {
- System.err
- .println("Unexpected type of score model, cannot calculate similarity");
- }
- return result;
+ @Override
+ public void println()
+ {
+ sb.append("\n");
+ }
+ };
+ return ps;
}
/**
public int getHeight()
{
// TODO can any of seqs[] be null?
- return seqs.getSequences().length;
+ return pairwiseScores.height();// seqs.getSequences().length;
+ }
+
+ /**
+ * Answers the sequence pairwise similarity scores which were the first step
+ * of the PCA calculation
+ *
+ * @return
+ */
+ public MatrixI getPairwiseScores()
+ {
+ return pairwiseScores;
+ }
+
+ public void setPairwiseScores(MatrixI m)
+ {
+ pairwiseScores = m;
+ }
+
+ public MatrixI getEigenmatrix()
+ {
+ return eigenMatrix;
+ }
+
+ public void setEigenmatrix(MatrixI m)
+ {
+ eigenMatrix = m;
+ }
+
+ public MatrixI getTridiagonal()
+ {
+ return tridiagonal;
+ }
+
+ public void setTridiagonal(MatrixI tridiagonal)
+ {
+ this.tridiagonal = tridiagonal;
}
}