package jalview.analysis.scoremodels;
import static org.testng.Assert.assertEquals;
+import static org.testng.Assert.assertNotNull;
+import static org.testng.Assert.assertNotSame;
+import static org.testng.Assert.assertTrue;
+import static org.testng.internal.junit.ArrayAsserts.assertArrayEquals;
+
+import jalview.api.analysis.SimilarityParamsI;
+import jalview.io.DataSourceType;
+import jalview.io.FileParse;
+import jalview.io.ScoreMatrixFile;
+import jalview.math.MatrixI;
+import jalview.schemes.ResidueProperties;
+
+import java.io.IOException;
+import java.net.MalformedURLException;
+import java.util.Arrays;
import org.testng.annotations.Test;
public class ScoreMatrixTest
{
@Test(groups = "Functional")
+ public void testConstructor()
+ {
+ // note score matrix does not have to be symmetric (though it should be!)
+ float[][] scores = new float[3][];
+ scores[0] = new float[] { 1f, 2f, 3f };
+ scores[1] = new float[] { 4f, 5f, 6f };
+ scores[2] = new float[] { 7f, 8f, 9f };
+ ScoreMatrix sm = new ScoreMatrix("Test", "ABC".toCharArray(), scores);
+ assertEquals(sm.getSize(), 3);
+ assertArrayEquals(scores, sm.getMatrix());
+ assertEquals(sm.getPairwiseScore('A', 'a'), 1f);
+ assertEquals(sm.getPairwiseScore('b', 'c'), 6f);
+ assertEquals(sm.getPairwiseScore('c', 'b'), 8f);
+ assertEquals(sm.getPairwiseScore('A', 'D'), 0f);
+ assertEquals(sm.getMatrixIndex('c'), 2);
+ assertEquals(sm.getMatrixIndex(' '), -1);
+ }
+
+ @Test(
+ groups = "Functional",
+ expectedExceptions = { IllegalArgumentException.class })
+ public void testConstructor_matrixTooSmall()
+ {
+ float[][] scores = new float[2][];
+ scores[0] = new float[] { 1f, 2f };
+ scores[1] = new float[] { 3f, 4f };
+ new ScoreMatrix("Test", "ABC".toCharArray(), scores);
+ }
+
+ @Test(
+ groups = "Functional",
+ expectedExceptions = { IllegalArgumentException.class })
+ public void testConstructor_matrixTooBig()
+ {
+ float[][] scores = new float[2][];
+ scores[0] = new float[] { 1f, 2f };
+ scores[1] = new float[] { 3f, 4f };
+ new ScoreMatrix("Test", "A".toCharArray(), scores);
+ }
+
+ @Test(
+ groups = "Functional",
+ expectedExceptions = { IllegalArgumentException.class })
+ public void testConstructor_matrixNotSquare()
+ {
+ float[][] scores = new float[2][];
+ scores[0] = new float[] { 1f, 2f };
+ scores[1] = new float[] { 3f };
+ new ScoreMatrix("Test", "AB".toCharArray(), scores);
+ }
+
+ @Test(groups = "Functional")
public void testBuildSymbolIndex()
{
short[] index = ScoreMatrix.buildSymbolIndex("AX-. yxYp".toCharArray());
}
}
}
+
+ @Test(groups = "Functional")
+ public void testGetMatrix()
+ {
+ ScoreMatrix sm = ScoreModels.getInstance().getBlosum62();
+ float[][] m = sm.getMatrix();
+ assertEquals(m.length, sm.getSize());
+ assertEquals(m[2][4], -3f);
+ // verify a defensive copy is returned
+ float[][] m2 = sm.getMatrix();
+ assertNotSame(m, m2);
+ assertTrue(Arrays.deepEquals(m, m2));
+ }
+
+ @Test(groups = "Functional")
+ public void testGetMatrixIndex()
+ {
+ ScoreMatrix sm = ScoreModels.getInstance().getBlosum62();
+ assertEquals(sm.getMatrixIndex('A'), 0);
+ assertEquals(sm.getMatrixIndex('R'), 1);
+ assertEquals(sm.getMatrixIndex('r'), 1);
+ assertEquals(sm.getMatrixIndex('N'), 2);
+ assertEquals(sm.getMatrixIndex('D'), 3);
+ assertEquals(sm.getMatrixIndex('X'), 22);
+ assertEquals(sm.getMatrixIndex('x'), 22);
+ assertEquals(sm.getMatrixIndex('-'), 23);
+ assertEquals(sm.getMatrixIndex('*'), 24);
+ assertEquals(sm.getMatrixIndex('.'), -1);
+ assertEquals(sm.getMatrixIndex(' '), -1);
+ assertEquals(sm.getMatrixIndex('?'), -1);
+ assertEquals(sm.getMatrixIndex((char) 128), -1);
+ }
+
+ @Test(groups = "Functional")
+ public void testGetSize()
+ {
+ ScoreMatrix sm = ScoreModels.getInstance().getBlosum62();
+ assertEquals(sm.getMatrix().length, sm.getSize());
+ }
+
+ @Test(groups = "Functional")
+ public void testComputePairwiseScores()
+ {
+ /*
+ * NB score matrix expects '-' for gap
+ */
+ String[] seqs = new String[] { "FKL", "R-D", "QIA", "GWC" };
+ ScoreMatrix sm = ScoreModels.getInstance().getBlosum62();
+
+ MatrixI pairwise = sm.findSimilarities(seqs, SimilarityParams.Jalview);
+
+ /*
+ * should be NxN where N = number of sequences
+ */
+ assertEquals(pairwise.height(), 4);
+ assertEquals(pairwise.width(), 4);
+
+ /*
+ * should be symmetrical (because BLOSUM62 is)
+ */
+ for (int i = 0; i < pairwise.height(); i++)
+ {
+ for (int j = i + 1; j < pairwise.width(); j++)
+ {
+ assertEquals(pairwise.getValue(i, j), pairwise.getValue(j, i),
+ String.format("Not symmetric at [%d, %d]", i, j));
+ }
+ }
+ /*
+ * verify expected BLOSUM dot product scores
+ */
+ // F.F + K.K + L.L = 6 + 5 + 4 = 15
+ assertEquals(pairwise.getValue(0, 0), 15d);
+ // R.R + -.- + D.D = 5 + 1 + 6 = 12
+ assertEquals(pairwise.getValue(1, 1), 12d);
+ // Q.Q + I.I + A.A = 5 + 4 + 4 = 13
+ assertEquals(pairwise.getValue(2, 2), 13d);
+ // G.G + W.W + C.C = 6 + 11 + 9 = 26
+ assertEquals(pairwise.getValue(3, 3), 26d);
+ // F.R + K.- + L.D = -3 + -4 + -4 = -11
+ assertEquals(pairwise.getValue(0, 1), -11d);
+ // F.Q + K.I + L.A = -3 + -3 + -1 = -7
+ assertEquals(pairwise.getValue(0, 2), -7d);
+ // F.G + K.W + L.C = -3 + -3 + -1 = -7
+ assertEquals(pairwise.getValue(0, 3), -7d);
+ // R.Q + -.I + D.A = 1 + -4 + -2 = -5
+ assertEquals(pairwise.getValue(1, 2), -5d);
+ // R.G + -.W + D.C = -2 + -4 + -3 = -9
+ assertEquals(pairwise.getValue(1, 3), -9d);
+ // Q.G + I.W + A.C = -2 + -3 + 0 = -5
+ assertEquals(pairwise.getValue(2, 3), -5d);
+ }
+
+ /**
+ * Test that the result of outputMatrix can be reparsed to give an identical
+ * ScoreMatrix
+ *
+ * @throws IOException
+ * @throws MalformedURLException
+ */
+ @Test(groups = "Functional")
+ public void testOutputMatrix_roundTrip() throws MalformedURLException,
+ IOException
+ {
+ ScoreMatrix sm = ScoreModels.getInstance().getBlosum62();
+ String output = sm.outputMatrix(false);
+ FileParse fp = new FileParse(output, DataSourceType.PASTE);
+ ScoreMatrixFile parser = new ScoreMatrixFile(fp);
+ ScoreMatrix sm2 = parser.parseMatrix();
+ assertNotNull(sm2);
+ assertTrue(sm2.equals(sm));
+ }
+
+ @Test(groups = "Functional")
+ public void testEqualsAndHashCode()
+ {
+ ScoreMatrix sm = ScoreModels.getInstance().getBlosum62();
+ ScoreMatrix sm2 = new ScoreMatrix(sm.getName(), sm.getSymbols()
+ .toCharArray(), sm.getMatrix());
+ assertTrue(sm.equals(sm2));
+ assertEquals(sm.hashCode(), sm2.hashCode());
+ }
+
+ /**
+ * Tests for scoring options where the longer length of two sequences is used
+ */
+ @Test(groups = "Functional")
+ public void testcomputeSimilarity_matchLongestSequence()
+ {
+ /*
+ * ScoreMatrix expects '-' for gaps
+ */
+ String s1 = "FR-K-S";
+ String s2 = "FS--L";
+ ScoreMatrix blosum = ScoreModels.getInstance().getBlosum62();
+
+ /*
+ * score gap-gap and gap-char
+ * shorter sequence treated as if with trailing gaps
+ * score = F^F + R^S + -^- + K^- + -^L + S^-
+ * = 6 + -1 + 1 + -4 + -4 + -4 = -6
+ */
+ SimilarityParamsI params = new SimilarityParams(true, true, true, false);
+ assertEquals(blosum.computeSimilarity(s1, s2, params), -6d);
+ // matchGap (arg2) is ignored:
+ params = new SimilarityParams(true, false, true, false);
+ assertEquals(blosum.computeSimilarity(s1, s2, params), -6d);
+
+ /*
+ * score gap-char but not gap-gap
+ * score = F^F + R^S + 0 + K^- + -^L + S^-
+ * = 6 + -1 + 0 + -4 + -4 + -4 = -7
+ */
+ params = new SimilarityParams(false, true, true, false);
+ assertEquals(blosum.computeSimilarity(s1, s2, params), -7d);
+ // matchGap (arg2) is ignored:
+ params = new SimilarityParams(false, false, true, false);
+ assertEquals(blosum.computeSimilarity(s1, s2, params), -7d);
+
+ /*
+ * score gap-gap but not gap-char
+ * score = F^F + R^S + -^- + 0 + 0 + 0
+ * = 6 + -1 + 1 = 6
+ */
+ params = new SimilarityParams(true, false, false, false);
+ assertEquals(blosum.computeSimilarity(s1, s2, params), 6d);
+ // matchGap (arg2) is ignored:
+ params = new SimilarityParams(true, true, false, false);
+ assertEquals(blosum.computeSimilarity(s1, s2, params), 6d);
+
+ /*
+ * score neither gap-gap nor gap-char
+ * score = F^F + R^S + 0 + 0 + 0 + 0
+ * = 6 + -1 = 5
+ */
+ params = new SimilarityParams(false, false, false, false);
+ assertEquals(blosum.computeSimilarity(s1, s2, params), 5d);
+ // matchGap (arg2) is ignored:
+ params = new SimilarityParams(false, true, false, false);
+ assertEquals(blosum.computeSimilarity(s1, s2, params), 5d);
+ }
+
+ /**
+ * Tests for scoring options where only the shorter length of two sequences is
+ * used
+ */
+ @Test(groups = "Functional")
+ public void testcomputeSimilarity_matchShortestSequence()
+ {
+ /*
+ * ScoreMatrix expects '-' for gaps
+ */
+ String s1 = "FR-K-S";
+ String s2 = "FS--L";
+ ScoreMatrix blosum = ScoreModels.getInstance().getBlosum62();
+
+ /*
+ * score gap-gap and gap-char
+ * match shorter sequence only
+ * score = F^F + R^S + -^- + K^- + -^L
+ * = 6 + -1 + 1 + -4 + -4 = -2
+ */
+ SimilarityParamsI params = new SimilarityParams(true, true, true, true);
+ assertEquals(blosum.computeSimilarity(s1, s2, params), -2d);
+ // matchGap (arg2) is ignored:
+ params = new SimilarityParams(true, false, true, true);
+ assertEquals(blosum.computeSimilarity(s1, s2, params), -2d);
+
+ /*
+ * score gap-char but not gap-gap
+ * score = F^F + R^S + 0 + K^- + -^L
+ * = 6 + -1 + 0 + -4 + -4 = -3
+ */
+ params = new SimilarityParams(false, true, true, true);
+ assertEquals(blosum.computeSimilarity(s1, s2, params), -3d);
+ // matchGap (arg2) is ignored:
+ params = new SimilarityParams(false, false, true, true);
+ assertEquals(blosum.computeSimilarity(s1, s2, params), -3d);
+
+ /*
+ * score gap-gap but not gap-char
+ * score = F^F + R^S + -^- + 0 + 0
+ * = 6 + -1 + 1 = 6
+ */
+ params = new SimilarityParams(true, false, false, true);
+ assertEquals(blosum.computeSimilarity(s1, s2, params), 6d);
+ // matchGap (arg2) is ignored:
+ params = new SimilarityParams(true, true, false, true);
+ assertEquals(blosum.computeSimilarity(s1, s2, params), 6d);
+
+ /*
+ * score neither gap-gap nor gap-char
+ * score = F^F + R^S + 0 + 0 + 0
+ * = 6 + -1 = 5
+ */
+ params = new SimilarityParams(false, false, false, true);
+ assertEquals(blosum.computeSimilarity(s1, s2, params), 5d);
+ // matchGap (arg2) is ignored:
+ params = new SimilarityParams(false, true, false, true);
+ assertEquals(blosum.computeSimilarity(s1, s2, params), 5d);
+ }
+
+ @Test(groups = "Functional")
+ public void testSymmetric()
+ {
+ verifySymmetric(ScoreModels.getInstance().getBlosum62());
+ verifySymmetric(ScoreModels.getInstance().getPam250());
+ verifySymmetric(ScoreModels.getInstance().getDefaultModel(false)); // dna
+ }
+
+ private void verifySymmetric(ScoreMatrix sm)
+ {
+ float[][] m = sm.getMatrix();
+ int rows = m.length;
+ for (int row = 0; row < rows; row++)
+ {
+ assertEquals(m[row].length, rows);
+ for (int col = 0; col < rows; col++)
+ {
+ assertEquals(m[row][col], m[col][row], String.format("%s [%s, %s]",
+ sm.getName(), ResidueProperties.aa[row],
+ ResidueProperties.aa[col]));
+ }
+ }
+ }
+
+ /**
+ * A test that just asserts the expected values in the Blosum62 score matrix
+ */
+ @Test(groups = "Functional")
+ public void testBlosum62_values()
+ {
+ ScoreMatrix sm = ScoreModels.getInstance().getBlosum62();
+
+ /*
+ * verify expected scores against ARNDCQEGHILKMFPSTWYVBZX
+ * scraped from https://www.ncbi.nlm.nih.gov/Class/FieldGuide/BLOSUM62.txt
+ */
+ verifyValues(sm, 'A', new float[] { 4, -1, -2, -2, 0, -1, -1, 0, -2,
+ -1,
+ -1, -1, -1, -2, -1, 1, 0, -3, -2, 0, -2, -1, 0 });
+ verifyValues(sm, 'R', new float[] { -1, 5, 0, -2, -3, 1, 0, -2, 0, -3,
+ -2, 2, -1, -3, -2, -1, -1, -3, -2, -3, -1, 0, -1 });
+ verifyValues(sm, 'N', new float[] { -2, 0, 6, 1, -3, 0, 0, 0, 1, -3,
+ -3,
+ 0, -2, -3, -2, 1, 0, -4, -2, -3, 3, 0, -1 });
+ verifyValues(sm, 'D', new float[] { -2, -2, 1, 6, -3, 0, 2, -1, -1, -3,
+ -4, -1, -3, -3, -1, 0, -1, -4, -3, -3, 4, 1, -1 });
+ verifyValues(sm, 'C', new float[] { 0, -3, -3, -3, 9, -3, -4, -3, -3,
+ -1,
+ -1, -3, -1, -2, -3, -1, -1, -2, -2, -1, -3, -3, -2 });
+ verifyValues(sm, 'Q', new float[] { -1, 1, 0, 0, -3, 5, 2, -2, 0, -3,
+ -2,
+ 1, 0, -3, -1, 0, -1, -2, -1, -2, 0, 3, -1 });
+ verifyValues(sm, 'E', new float[] { -1, 0, 0, 2, -4, 2, 5, -2, 0, -3,
+ -3,
+ 1, -2, -3, -1, 0, -1, -3, -2, -2, 1, 4, -1 });
+ verifyValues(sm, 'G', new float[] { 0, -2, 0, -1, -3, -2, -2, 6, -2,
+ -4,
+ -4, -2, -3, -3, -2, 0, -2, -2, -3, -3, -1, -2, -1 });
+ verifyValues(sm, 'H', new float[] { -2, 0, 1, -1, -3, 0, 0, -2, 8, -3,
+ -3, -1, -2, -1, -2, -1, -2, -2, 2, -3, 0, 0, -1 });
+ verifyValues(sm, 'I', new float[] { -1, -3, -3, -3, -1, -3, -3, -4, -3,
+ 4, 2, -3, 1, 0, -3, -2, -1, -3, -1, 3, -3, -3, -1 });
+ verifyValues(sm, 'L', new float[] { -1, -2, -3, -4, -1, -2, -3, -4, -3,
+ 2, 4, -2, 2, 0, -3, -2, -1, -2, -1, 1, -4, -3, -1 });
+ verifyValues(sm, 'K', new float[] { -1, 2, 0, -1, -3, 1, 1, -2, -1, -3,
+ -2, 5, -1, -3, -1, 0, -1, -3, -2, -2, 0, 1, -1 });
+ verifyValues(sm, 'M', new float[] { -1, -1, -2, -3, -1, 0, -2, -3, -2,
+ 1,
+ 2, -1, 5, 0, -2, -1, -1, -1, -1, 1, -3, -1, -1 });
+ verifyValues(sm, 'F', new float[] { -2, -3, -3, -3, -2, -3, -3, -3, -1,
+ 0, 0, -3, 0, 6, -4, -2, -2, 1, 3, -1, -3, -3, -1 });
+ verifyValues(sm, 'P', new float[] { -1, -2, -2, -1, -3, -1, -1, -2, -2,
+ -3, -3, -1, -2, -4, 7, -1, -1, -4, -3, -2, -2, -1, -2 });
+ verifyValues(sm, 'S', new float[] { 1, -1, 1, 0, -1, 0, 0, 0, -1, -2,
+ -2,
+ 0, -1, -2, -1, 4, 1, -3, -2, -2, 0, 0, 0 });
+ verifyValues(sm, 'T', new float[] { 0, -1, 0, -1, -1, -1, -1, -2, -2,
+ -1,
+ -1, -1, -1, -2, -1, 1, 5, -2, -2, 0, -1, -1, 0 });
+ verifyValues(sm, 'W', new float[] { -3, -3, -4, -4, -2, -2, -3, -2, -2,
+ -3, -2, -3, -1, 1, -4, -3, -2, 11, 2, -3, -4, -3, -2 });
+ verifyValues(sm, 'Y', new float[] { -2, -2, -2, -3, -2, -1, -2, -3, 2,
+ -1, -1, -2, -1, 3, -3, -2, -2, 2, 7, -1, -3, -2, -1 });
+ verifyValues(sm, 'V', new float[] { 0, -3, -3, -3, -1, -2, -2, -3, -3,
+ 3,
+ 1, -2, 1, -1, -2, -2, 0, -3, -1, 4, -3, -2, -1 });
+ verifyValues(sm, 'B', new float[] { -2, -1, 3, 4, -3, 0, 1, -1, 0, -3,
+ -4, 0, -3, -3, -2, 0, -1, -4, -3, -3, 4, 1, -1 });
+ verifyValues(sm, 'Z', new float[] { -1, 0, 0, 1, -3, 3, 4, -2, 0, -3,
+ -3,
+ 1, -1, -3, -1, 0, -1, -3, -2, -2, 1, 4, -1 });
+ verifyValues(sm, 'X', new float[] { 0, -1, -1, -1, -2, -1, -1, -1, -1,
+ -1, -1, -1, -1, -1, -2, 0, 0, -2, -1, -1, -1, -1, -1 });
+ }
+
+ /**
+ * Helper method to check pairwise scores for one residue
+ *
+ * @param sm
+ * @param res
+ * @param expected
+ * score values against 'res', in ResidueProperties.aaIndex order
+ */
+ private void verifyValues(ScoreMatrix sm, char res, float[] expected)
+ {
+ for (int j = 0; j < expected.length; j++)
+ {
+ char c2 = ResidueProperties.aa[j].charAt(0);
+ assertEquals(sm.getPairwiseScore(res, c2), expected[j],
+ String.format("%s->%s", res, c2));
+ }
+ }
}
git://source.jalview.org / jalview.git / blobdiff