/* * Jalview - A Sequence Alignment Editor and Viewer (Version 2.9.0b2) * Copyright (C) 2015 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.datamodel; import static org.testng.AssertJUnit.assertEquals; import static org.testng.AssertJUnit.assertFalse; import static org.testng.AssertJUnit.assertTrue; import jalview.io.AppletFormatAdapter; import jalview.io.FormatAdapter; import jalview.util.MapList; import java.io.IOException; import java.util.Iterator; import org.testng.annotations.BeforeMethod; import org.testng.annotations.Test; /** * Unit tests for Alignment datamodel. * * @author gmcarstairs * */ public class AlignmentTest { // @formatter:off private static final String TEST_DATA = "# STOCKHOLM 1.0\n" + "#=GS D.melanogaster.1 AC AY119185.1/838-902\n" + "#=GS D.melanogaster.2 AC AC092237.1/57223-57161\n" + "#=GS D.melanogaster.3 AC AY060611.1/560-627\n" + "D.melanogaster.1 G.AGCC.CU...AUGAUCGA\n" + "#=GR D.melanogaster.1 SS ................((((\n" + "D.melanogaster.2 C.AUUCAACU.UAUGAGGAU\n" + "#=GR D.melanogaster.2 SS ................((((\n" + "D.melanogaster.3 G.UGGCGCU..UAUGACGCA\n" + "#=GR D.melanogaster.3 SS (.(((...(....(((((((\n" + "//"; private static final String AA_SEQS_1 = ">Seq1Name/5-8\n" + "K-QY--L\n" + ">Seq2Name/12-15\n" + "-R-FP-W-\n"; private static final String CDNA_SEQS_1 = ">Seq1Name/100-111\n" + "AC-GG--CUC-CAA-CT\n" + ">Seq2Name/200-211\n" + "-CG-TTA--ACG---AAGT\n"; private static final String CDNA_SEQS_2 = ">Seq1Name/50-61\n" + "GCTCGUCGTACT\n" + ">Seq2Name/60-71\n" + "GGGTCAGGCAGT\n"; // @formatter:on private AlignmentI al; /** * Helper method to load an alignment and ensure dataset sequences are set up. * * @param data * @param format * TODO * @return * @throws IOException */ protected AlignmentI loadAlignment(final String data, String format) throws IOException { AlignmentI a = new FormatAdapter().readFile(data, AppletFormatAdapter.PASTE, format); a.setDataset(null); return a; } /* * Read in Stockholm format test data including secondary structure * annotations. */ @BeforeMethod(alwaysRun = true) public void setUp() throws IOException { al = loadAlignment(TEST_DATA, "STH"); int i = 0; for (AlignmentAnnotation ann : al.getAlignmentAnnotation()) { ann.setCalcId("CalcIdFor" + al.getSequenceAt(i).getName()); i++; } } /** * Test method that returns annotations that match on calcId. */ @Test(groups = { "Functional" }) public void testFindAnnotation_byCalcId() { Iterable anns = al .findAnnotation("CalcIdForD.melanogaster.2"); Iterator iter = anns.iterator(); assertTrue(iter.hasNext()); AlignmentAnnotation ann = iter.next(); assertEquals("D.melanogaster.2", ann.sequenceRef.getName()); assertFalse(iter.hasNext()); } @Test(groups = { "Functional" }) public void testDeleteAllAnnotations_includingAutocalculated() { AlignmentAnnotation aa = new AlignmentAnnotation("Consensus", "Consensus", 0.5); aa.autoCalculated = true; al.addAnnotation(aa); AlignmentAnnotation[] anns = al.getAlignmentAnnotation(); assertEquals("Wrong number of annotations before deleting", 4, anns.length); al.deleteAllAnnotations(true); assertEquals("Not all deleted", 0, al.getAlignmentAnnotation().length); } @Test(groups = { "Functional" }) public void testDeleteAllAnnotations_excludingAutocalculated() { AlignmentAnnotation aa = new AlignmentAnnotation("Consensus", "Consensus", 0.5); aa.autoCalculated = true; al.addAnnotation(aa); AlignmentAnnotation[] anns = al.getAlignmentAnnotation(); assertEquals("Wrong number of annotations before deleting", 4, anns.length); al.deleteAllAnnotations(false); assertEquals("Not just one annotation left", 1, al.getAlignmentAnnotation().length); } /** * Tests for realigning as per a supplied alignment: Dna as Dna. * * Note: AlignedCodonFrame's state variables are named for protein-to-cDNA * mapping, but can be exploited for a general 'sequence-to-sequence' mapping * as here. * * @throws IOException */ @Test(groups = { "Functional" }) public void testAlignAs_dnaAsDna() throws IOException { // aligned cDNA: AlignmentI al1 = loadAlignment(CDNA_SEQS_1, "FASTA"); // unaligned cDNA: AlignmentI al2 = loadAlignment(CDNA_SEQS_2, "FASTA"); /* * Make mappings between sequences. The 'aligned cDNA' is playing the role * of what would normally be protein here. */ makeMappings(al2, al1); ((Alignment) al2).alignAs(al1, false, true); assertEquals("GC-TC--GUC-GTA-CT", al2.getSequenceAt(0) .getSequenceAsString()); assertEquals("-GG-GTC--AGG---CAGT", al2.getSequenceAt(1) .getSequenceAsString()); } /** * Aligning protein from cDNA. * * @throws IOException */ @Test(groups = { "Functional" }) public void testAlignAs_proteinAsCdna() throws IOException { // see also AlignmentUtilsTests AlignmentI al1 = loadAlignment(CDNA_SEQS_1, "FASTA"); AlignmentI al2 = loadAlignment(AA_SEQS_1, "FASTA"); makeMappings(al1, al2); ((Alignment) al2).alignAs(al1, false, true); assertEquals("K-Q-Y-L-", al2.getSequenceAt(0).getSequenceAsString()); assertEquals("-R-F-P-W", al2.getSequenceAt(1).getSequenceAsString()); } /** * Aligning protein from cDNA for a single sequence. This is the 'simple' case * (as there is no need to compute codon 'alignments') but worth testing * before tackling the multiple sequence case. * * @throws IOException */ @Test(groups = { "Functional" }) public void testAlignAs_proteinAsCdna_singleSequence() throws IOException { /* * simplest case remove all gaps */ verifyAlignAs(">protein\n-Q-K-\n", ">dna\nCAAaaa\n", "QK"); /* * with sequence offsets */ verifyAlignAs(">protein/12-13\n-Q-K-\n", ">dna/20-25\nCAAaaa\n", "QK"); } /** * Test aligning cdna as per protein alignment. * * @throws IOException */ @Test(groups = { "Functional" }) public void testAlignAs_cdnaAsProtein() throws IOException { /* * Load alignments and add mappings for cDNA to protein */ AlignmentI al1 = loadAlignment(CDNA_SEQS_1, "FASTA"); AlignmentI al2 = loadAlignment(AA_SEQS_1, "FASTA"); makeMappings(al1, al2); /* * Realign DNA; currently keeping existing gaps in introns only */ ((Alignment) al1).alignAs(al2, false, true); assertEquals("ACG---GCUCCA------ACT", al1.getSequenceAt(0) .getSequenceAsString()); assertEquals("---CGT---TAACGA---AGT", al1.getSequenceAt(1) .getSequenceAsString()); } /** * Test aligning cdna as per protein - single sequences * * @throws IOException */ @Test(groups = { "Functional" }) public void testAlignAs_cdnaAsProtein_singleSequence() throws IOException { /* * simple case insert one gap */ verifyAlignAs(">dna\nCAAaaa\n", ">protein\nQ-K\n", "CAA---aaa"); /* * simple case but with sequence offsets */ verifyAlignAs(">dna/5-10\nCAAaaa\n", ">protein/20-21\nQ-K\n", "CAA---aaa"); /* * insert gaps as per protein, drop gaps within codons */ verifyAlignAs(">dna/10-18\nCA-Aa-aa--AGA\n", ">aa/6-8\n-Q-K--R\n", "---CAA---aaa------AGA"); } /** * Helper method that makes mappings and then aligns the first alignment as * the second * * @param fromSeqs * @param toSeqs * @param expected * @throws IOException */ public void verifyAlignAs(String fromSeqs, String toSeqs, String expected) throws IOException { /* * Load alignments and add mappings from nucleotide to protein (or from * first to second if both the same type) */ AlignmentI al1 = loadAlignment(fromSeqs, "FASTA"); AlignmentI al2 = loadAlignment(toSeqs, "FASTA"); makeMappings(al1, al2); /* * Realign DNA; currently keeping existing gaps in introns only */ ((Alignment) al1).alignAs(al2, false, true); assertEquals(expected, al1.getSequenceAt(0).getSequenceAsString()); } /** * Helper method to make mappings from protein to dna sequences, and add the * mappings to the protein alignment * * @param alFrom * @param alTo */ public void makeMappings(AlignmentI alFrom, AlignmentI alTo) { AlignmentI prot = !alFrom.isNucleotide() ? alFrom : alTo; AlignmentI nuc = alFrom == prot ? alTo : alFrom; int ratio = (alFrom.isNucleotide() == alTo.isNucleotide() ? 1 : 3); AlignedCodonFrame acf = new AlignedCodonFrame(); for (int i = 0; i < nuc.getHeight(); i++) { SequenceI seqFrom = nuc.getSequenceAt(i); SequenceI seqTo = prot.getSequenceAt(i); MapList ml = new MapList(new int[] { seqFrom.getStart(), seqFrom.getEnd() }, new int[] { seqTo.getStart(), seqTo.getEnd() }, ratio, 1); acf.addMap(seqFrom, seqTo, ml); } prot.addCodonFrame(acf); } /** * Test aligning dna as per protein alignment, for the case where there are * introns (i.e. some dna sites have no mapping from a peptide). * * @throws IOException */ @Test(groups = { "Functional" }) public void testAlignAs_dnaAsProtein_withIntrons() throws IOException { /* * Load alignments and add mappings for cDNA to protein */ String dna1 = "A-Aa-gG-GCC-cT-TT"; String dna2 = "c--CCGgg-TT--T-AA-A"; AlignmentI al1 = loadAlignment(">Seq1/6-17\n" + dna1 + "\n>Seq2/20-31\n" + dna2 + "\n", "FASTA"); AlignmentI al2 = loadAlignment( ">Seq1/7-9\n-P--YK\n>Seq2/11-13\nG-T--F\n", "FASTA"); AlignedCodonFrame acf = new AlignedCodonFrame(); // Seq1 has intron at dna positions 3,4,9 so splice is AAG GCC TTT // Seq2 has intron at dna positions 1,5,6 so splice is CCG TTT AAA // TODO sequence offsets MapList ml1 = new MapList(new int[] { 6, 7, 10, 13, 15, 17 }, new int[] { 7, 9 }, 3, 1); acf.addMap(al1.getSequenceAt(0), al2.getSequenceAt(0), ml1); MapList ml2 = new MapList(new int[] { 21, 23, 26, 31 }, new int[] { 11, 13 }, 3, 1); acf.addMap(al1.getSequenceAt(1), al2.getSequenceAt(1), ml2); al2.addCodonFrame(acf); /* * Align ignoring gaps in dna introns and exons */ ((Alignment) al1).alignAs(al2, false, false); assertEquals("---AAagG------GCCcTTT", al1.getSequenceAt(0) .getSequenceAsString()); // note 1 gap in protein corresponds to 'gg-' in DNA (3 positions) assertEquals("cCCGgg-TTT------AAA", al1.getSequenceAt(1) .getSequenceAsString()); /* * Reset and realign, preserving gaps in dna introns and exons */ al1.getSequenceAt(0).setSequence(dna1); al1.getSequenceAt(1).setSequence(dna2); ((Alignment) al1).alignAs(al2, true, true); // String dna1 = "A-Aa-gG-GCC-cT-TT"; // String dna2 = "c--CCGgg-TT--T-AA-A"; // assumption: we include 'the greater of' protein/dna gap lengths, not both assertEquals("---A-Aa-gG------GCC-cT-TT", al1.getSequenceAt(0) .getSequenceAsString()); assertEquals("c--CCGgg-TT--T------AA-A", al1.getSequenceAt(1) .getSequenceAsString()); } }