From 535702de73ec9ea47e55ca5015822ad0a8fe323a Mon Sep 17 00:00:00 2001 From: gmungoc Date: Wed, 28 Mar 2018 15:11:15 +0100 Subject: [PATCH] JAL-1265 output all dbrefs in Stockholm (+tests) (+some refactoring) --- src/jalview/io/StockholmFile.java | 397 ++++++++++++++++---------------- test/jalview/io/StockholmFileTest.java | 172 +++++++++++++- 2 files changed, 363 insertions(+), 206 deletions(-) diff --git a/src/jalview/io/StockholmFile.java b/src/jalview/io/StockholmFile.java index afbdf16..09c5b25 100644 --- a/src/jalview/io/StockholmFile.java +++ b/src/jalview/io/StockholmFile.java @@ -28,6 +28,7 @@ import jalview.datamodel.AlignmentAnnotation; import jalview.datamodel.AlignmentI; import jalview.datamodel.Annotation; import jalview.datamodel.DBRefEntry; +import jalview.datamodel.DBRefSource; import jalview.datamodel.Mapping; import jalview.datamodel.Sequence; import jalview.datamodel.SequenceFeature; @@ -42,10 +43,12 @@ import java.io.FileReader; import java.io.IOException; import java.util.ArrayList; import java.util.Enumeration; +import java.util.HashMap; import java.util.Hashtable; import java.util.LinkedHashMap; import java.util.List; import java.util.Map; +import java.util.Map.Entry; import java.util.Vector; import com.stevesoft.pat.Regex; @@ -76,14 +79,37 @@ public class StockholmFile extends AlignFile { private static final String ANNOTATION = "annotation"; - private static final Regex OPEN_PAREN = new Regex("(<|\\[)", "("); - - private static final Regex CLOSE_PAREN = new Regex("(>|\\])", ")"); + // private static final Regex OPEN_PAREN = new Regex("(<|\\[)", "("); + // private static final Regex CLOSE_PAREN = new Regex("(>|\\])", ")"); public static final Regex DETECT_BRACKETS = new Regex( "(<|>|\\[|\\]|\\(|\\)|\\{|\\})"); - StringBuffer out; // output buffer + /* + * lookup table of Stockholm 'feature' (annotation) types + * see http://sonnhammer.sbc.su.se/Stockholm.html + */ + private static Map featureTypes = null; + + static + { + featureTypes = new HashMap<>(); + featureTypes.put("SS", "Secondary Structure"); + featureTypes.put("SA", "Surface Accessibility"); + featureTypes.put("TM", "transmembrane"); + featureTypes.put("PP", "Posterior Probability"); + featureTypes.put("LI", "ligand binding"); + featureTypes.put("AS", "active site"); + featureTypes.put("IN", "intron"); + featureTypes.put("IR", "interacting residue"); + featureTypes.put("AC", "accession"); + featureTypes.put("OS", "organism"); + featureTypes.put("CL", "class"); + featureTypes.put("DE", "description"); + featureTypes.put("DR", "reference"); + featureTypes.put("LO", "look"); + featureTypes.put("RF", "Reference Positions"); + } AlignmentI al; @@ -110,6 +136,47 @@ public class StockholmFile extends AlignFile super(source); } + /** + * Answers the readable description for a (case-sensitive) annotation type + * code, for example "Reference Positions" for "RF". Returns the type code if + * no description is found. + * + * @param id + * @return + */ + public static String typeToDescription(String id) + { + if (featureTypes.containsKey(id)) + { + return featureTypes.get(id); + } + System.err.println( + "Warning : Unknown Stockholm annotation type code " + id); + return id; + } + + /** + * Answers the annotation type code for a (non-case-sensitive) readable + * description, for example "RF" for "Reference Positions" (or null if not + * found) + * + * @param description + * @return + */ + public static String descriptionToType(String description) + { + for (Entry entry : featureTypes.entrySet()) + { + if (entry.getValue().equalsIgnoreCase(description)) + { + return entry.getKey(); + } + } + System.err.println( + "Warning : Unknown Stockholm annotation type: " + description); + return null; + } + @Override public void initData() { @@ -219,7 +286,7 @@ public class StockholmFile extends AlignFile // logger.debug("Stockholm version: " + version); } - // We define some Regexes here that will be used regularily later + // We define some Regexes here that will be used regularly later rend = new Regex("^\\s*\\/\\/"); // Find the end of an alignment p = new Regex("(\\S+)\\/(\\d+)\\-(\\d+)"); // split sequence id in // id/from/to @@ -232,7 +299,7 @@ public class StockholmFile extends AlignFile Regex closeparen = new Regex("(>|\\])", ")"); // Detect if file is RNA by looking for bracket types - Regex detectbrackets = new Regex("(<|>|\\[|\\]|\\(|\\))"); + // Regex detectbrackets = new Regex("(<|>|\\[|\\]|\\(|\\))"); rend.optimize(); p.optimize(); @@ -366,7 +433,7 @@ public class StockholmFile extends AlignFile Hashtable content = (Hashtable) features.remove(type); // add alignment annotation for this feature - String key = type2id(type); + String key = descriptionToType(type); /* * have we added annotation rows for this type ? @@ -624,17 +691,18 @@ public class StockholmFile extends AlignFile } Hashtable content; - if (features.containsKey(this.id2type(type))) + if (features.containsKey(StockholmFile.typeToDescription(type))) { // logger.debug("Found content for " + this.id2type(type)); - content = (Hashtable) features.get(this.id2type(type)); + content = (Hashtable) features + .get(StockholmFile.typeToDescription(type)); } else { // logger.debug("Creating new content holder for " + // this.id2type(type)); content = new Hashtable(); - features.put(this.id2type(type), content); + features.put(StockholmFile.typeToDescription(type), content); } String ns = (String) content.get(ANNOTATION); @@ -809,10 +877,8 @@ public class StockholmFile extends AlignFile Vector annotation, String label, String annots) { - String convert1, convert2 = null; - - // convert1 = OPEN_PAREN.replaceAll(annots); - // convert2 = CLOSE_PAREN.replaceAll(convert1); + // String convert1 = OPEN_PAREN.replaceAll(annots); + // String convert2 = CLOSE_PAREN.replaceAll(convert1); // annots = convert2; String type = label; @@ -823,7 +889,7 @@ public class StockholmFile extends AlignFile : label; } boolean ss = false, posterior = false; - type = id2type(type); + type = typeToDescription(type); if (type.equalsIgnoreCase("secondary structure")) { ss = true; @@ -918,144 +984,123 @@ public class StockholmFile extends AlignFile } @Override - public String print(SequenceI[] s, boolean jvSuffix) + public String print(final SequenceI[] sequences, boolean jvSuffix) { - out = new StringBuffer(); + StringBuilder out = new StringBuilder(); out.append("# STOCKHOLM 1.0"); out.append(newline); - // find max length of id - int max = 0; - int maxid = 0; - int in = 0; - Hashtable dataRef = null; - while ((in < s.length) && (s[in] != null)) + int maxIdWidth = 0; + for (SequenceI seq : sequences) { - String tmp = printId(s[in], jvSuffix); - max = Math.max(max, s[in].getLength()); - - if (tmp.length() > maxid) + if (seq != null) { - maxid = tmp.length(); + String formattedId = printId(seq, jvSuffix); + maxIdWidth = Math.max(maxIdWidth, formattedId.length()); } - if (s[in].getDBRefs() != null) - { - for (int idb = 0; idb < s[in].getDBRefs().length; idb++) - { - if (dataRef == null) - { - dataRef = new Hashtable(); - } - - String datAs1 = s[in].getDBRefs()[idb].getSource().toString() - + " ; " - + s[in].getDBRefs()[idb].getAccessionId().toString(); - dataRef.put(tmp, datAs1); - } - } - in++; } - maxid += 9; - int i = 0; + maxIdWidth += 9; - // output database type - if (al.getProperties() != null) + /* + * generic alignment properties + */ + Hashtable props = al.getProperties(); + if (props != null) { - if (!al.getProperties().isEmpty()) + for (Object key : props.keySet()) { - Enumeration key = al.getProperties().keys(); - Enumeration val = al.getProperties().elements(); - while (key.hasMoreElements()) - { - out.append("#=GF " + key.nextElement() + " " + val.nextElement()); - out.append(newline); - } + out.append(String.format("#=GF %s %s", key.toString(), + props.get(key).toString())); + out.append(newline); } } - // output database accessions - if (dataRef != null) + /* + * output database accessions as #=GS (per sequence annotation) + * PFAM or RFAM are output as AC + * others are output as DR ; + */ + Format formatter = new Format("%-" + (maxIdWidth - 2) + "s"); + for (SequenceI seq : sequences) { - Enumeration en = dataRef.keys(); - while (en.hasMoreElements()) + if (seq != null) { - Object idd = en.nextElement(); - String type = (String) dataRef.remove(idd); - out.append(new Format("%-" + (maxid - 2) + "s") - .form("#=GS " + idd.toString() + " ")); - if (type.contains("PFAM") || type.contains("RFAM")) + DBRefEntry[] dbRefs = seq.getDBRefs(); + if (dbRefs != null) { - - out.append(" AC " + type.substring(type.indexOf(";") + 1)); - } - else - { - out.append(" DR " + type + " "); + String idField = formatter + .form("#=GS " + printId(seq, jvSuffix) + " "); + for (DBRefEntry dbRef : dbRefs) + { + out.append(idField); + printDbRef(out, dbRef); + } } - out.append(newline); } } - - // output annotations - while (i < s.length && s[i] != null) + /* + * output annotations + */ + for (SequenceI seq : sequences) { - AlignmentAnnotation[] alAnot = s[i].getAnnotation(); - if (alAnot != null) + if (seq != null) { - Annotation[] ann; - for (int j = 0; j < alAnot.length; j++) + AlignmentAnnotation[] alAnot = seq.getAnnotation(); + if (alAnot != null) { - - String key = type2id(alAnot[j].label); - boolean isrna = alAnot[j].isValidStruc(); - if (isrna) - { - // hardwire to secondary structure if there is RNA secondary - // structure on the annotation - key = "SS"; - } - if (key == null) + for (int j = 0; j < alAnot.length; j++) { + AlignmentAnnotation ann = alAnot[j]; + String key = descriptionToType(ann.label); + boolean isrna = ann.isValidStruc(); + if (isrna) + { + /* + * output as secondary structure if there is + * RNA secondary structure on the annotation + */ + key = "SS"; + } + if (key == null) + { + continue; + } - continue; + out.append(new Format("%-" + maxIdWidth + "s").form( + "#=GR " + printId(seq, jvSuffix) + " " + key + " ")); + Annotation[] anns = ann.annotations; + StringBuilder seqString = new StringBuilder(); + for (int k = 0; k < anns.length; k++) + { + seqString + .append(getAnnotationCharacter(key, k, anns[k], seq)); + } + out.append(seqString.toString()); + out.append(newline); } - - // out.append("#=GR "); - out.append(new Format("%-" + maxid + "s").form( - "#=GR " + printId(s[i], jvSuffix) + " " + key + " ")); - ann = alAnot[j].annotations; - String seq = ""; - for (int k = 0; k < ann.length; k++) - { - seq += outputCharacter(key, k, isrna, ann, s[i]); - - } - out.append(seq); - out.append(newline); } - } - out.append(new Format("%-" + maxid + "s") - .form(printId(s[i], jvSuffix) + " ")); - out.append(s[i].getSequenceAsString()); - out.append(newline); - i++; + out.append(new Format("%-" + maxIdWidth + "s") + .form(printId(seq, jvSuffix) + " ")); + out.append(seq.getSequenceAsString()); + out.append(newline); + } } - // alignment annotation - AlignmentAnnotation aa; + /* + * output alignment annotation (but not auto-calculated or sequence-related) + */ if (al.getAlignmentAnnotation() != null) { for (int ia = 0; ia < al.getAlignmentAnnotation().length; ia++) { - aa = al.getAlignmentAnnotation()[ia]; + AlignmentAnnotation aa = al.getAlignmentAnnotation()[ia]; if (aa.autoCalculated || !aa.visible || aa.sequenceRef != null) { continue; } - String seq = ""; - String label; + String label = aa.label; String key = ""; if (aa.label.equals("seq")) { @@ -1063,39 +1108,31 @@ public class StockholmFile extends AlignFile } else { - key = type2id(aa.label.toLowerCase()); - if (key == null) - { - label = aa.label; - } - else if ("RF".equals(key)) + key = descriptionToType(aa.label); + if ("RF".equals(key)) { label = key; } - else + else if (key != null) { label = key + "_cons"; } } - if (label == null) - { - label = aa.label; - } label = label.replace(" ", "_"); out.append( - new Format("%-" + maxid + "s").form("#=GC " + label + " ")); - boolean isrna = aa.isValidStruc(); + new Format("%-" + maxIdWidth + "s") + .form("#=GC " + label + " ")); + StringBuilder sb = new StringBuilder(aa.annotations.length); for (int j = 0; j < aa.annotations.length; j++) { - seq += outputCharacter(key, j, isrna, aa.annotations, null); + sb.append( + getAnnotationCharacter(key, j, aa.annotations[j], null)); } - out.append(seq); + out.append(sb.toString()); out.append(newline); } } - - out.append("//"); out.append(newline); @@ -1104,20 +1141,40 @@ public class StockholmFile extends AlignFile } /** - * add an annotation character to the output row + * A helper method that appends a formatted dbref to the output buffer + * + * @param out + * @param dbRef + */ + protected void printDbRef(StringBuilder out, DBRefEntry dbRef) + { + String db = dbRef.getSource(); + String acc = dbRef.getAccessionId(); + if (DBRefSource.PFAM.equalsIgnoreCase(db) + || DBRefSource.RFAM.equalsIgnoreCase(db)) + { + out.append(" AC " + acc); + } + else + { + out.append(" DR " + db + " ; " + acc); + } + out.append(newline); + } + + /** + * Returns an annotation character to add to the output row * * @param seq * @param key * @param k - * @param isrna * @param ann * @param sequenceI */ - private char outputCharacter(String key, int k, boolean isrna, - Annotation[] ann, SequenceI sequenceI) + private char getAnnotationCharacter(String key, int k, Annotation annot, + SequenceI sequenceI) { char seq = ' '; - Annotation annot = ann[k]; String ch = (annot == null) ? ((sequenceI == null) ? "-" : Character.toString(sequenceI.getCharAt(k))) @@ -1154,74 +1211,6 @@ public class StockholmFile extends AlignFile return seq; } - public String print() - { - out = new StringBuffer(); - out.append("# STOCKHOLM 1.0"); - out.append(newline); - print(getSeqsAsArray(), false); - return out.toString(); - } - - private static Hashtable typeIds = null; - - static - { - if (typeIds == null) - { - typeIds = new Hashtable(); - typeIds.put("SS", "Secondary Structure"); - typeIds.put("SA", "Surface Accessibility"); - typeIds.put("TM", "transmembrane"); - typeIds.put("PP", "Posterior Probability"); - typeIds.put("LI", "ligand binding"); - typeIds.put("AS", "active site"); - typeIds.put("IN", "intron"); - typeIds.put("IR", "interacting residue"); - typeIds.put("AC", "accession"); - typeIds.put("OS", "organism"); - typeIds.put("CL", "class"); - typeIds.put("DE", "description"); - typeIds.put("DR", "reference"); - typeIds.put("LO", "look"); - typeIds.put("RF", "Reference Positions"); - - } - } - - protected static String id2type(String id) - { - if (typeIds.containsKey(id)) - { - return (String) typeIds.get(id); - } - System.err.println( - "Warning : Unknown Stockholm annotation type code " + id); - return id; - } - - protected static String type2id(String type) - { - String key = null; - Enumeration e = typeIds.keys(); - while (e.hasMoreElements()) - { - Object ll = e.nextElement(); - if (typeIds.get(ll).toString().equalsIgnoreCase(type)) - { - key = (String) ll; - break; - } - } - if (key != null) - { - return key; - } - System.err.println( - "Warning : Unknown Stockholm annotation type: " + type); - return key; - } - /** * make a friendly ID string. * diff --git a/test/jalview/io/StockholmFileTest.java b/test/jalview/io/StockholmFileTest.java index 4273e6c..14050e8 100644 --- a/test/jalview/io/StockholmFileTest.java +++ b/test/jalview/io/StockholmFileTest.java @@ -22,12 +22,16 @@ package jalview.io; import static org.testng.AssertJUnit.assertEquals; import static org.testng.AssertJUnit.assertNotNull; +import static org.testng.AssertJUnit.assertNull; import static org.testng.AssertJUnit.assertTrue; import static org.testng.AssertJUnit.fail; +import jalview.datamodel.Alignment; import jalview.datamodel.AlignmentAnnotation; import jalview.datamodel.AlignmentI; import jalview.datamodel.Annotation; +import jalview.datamodel.DBRefEntry; +import jalview.datamodel.Sequence; import jalview.datamodel.SequenceFeature; import jalview.datamodel.SequenceI; import jalview.gui.JvOptionPane; @@ -230,8 +234,8 @@ public class StockholmFileTest // we might want to revise this in future int aa_new_size = (aa_new == null ? 0 : aa_new.length); int aa_original_size = (aa_original == null ? 0 : aa_original.length); - Map orig_groups = new HashMap(); - Map new_groups = new HashMap(); + Map orig_groups = new HashMap<>(); + Map new_groups = new HashMap<>(); if (aa_new != null && aa_original != null) { @@ -654,4 +658,168 @@ public class StockholmFileTest testAlignmentEquivalence(al, newAl, true, true, true); } + + @Test(groups = { "Functional" }) + public void testTypeToDescription() + { + assertEquals("Secondary Structure", + StockholmFile.typeToDescription("SS")); + assertEquals("Surface Accessibility", + StockholmFile.typeToDescription("SA")); + assertEquals("transmembrane", StockholmFile.typeToDescription("TM")); + assertEquals("Posterior Probability", + StockholmFile.typeToDescription("PP")); + assertEquals("ligand binding", StockholmFile.typeToDescription("LI")); + assertEquals("active site", StockholmFile.typeToDescription("AS")); + assertEquals("intron", StockholmFile.typeToDescription("IN")); + assertEquals("interacting residue", + StockholmFile.typeToDescription("IR")); + assertEquals("accession", StockholmFile.typeToDescription("AC")); + assertEquals("organism", StockholmFile.typeToDescription("OS")); + assertEquals("class", StockholmFile.typeToDescription("CL")); + assertEquals("description", StockholmFile.typeToDescription("DE")); + assertEquals("reference", StockholmFile.typeToDescription("DR")); + assertEquals("look", StockholmFile.typeToDescription("LO")); + assertEquals("Reference Positions", + StockholmFile.typeToDescription("RF")); + + // case-sensitive: + assertEquals("Rf", StockholmFile.typeToDescription("Rf")); + assertEquals("junk", StockholmFile.typeToDescription("junk")); + assertEquals("", StockholmFile.typeToDescription("")); + assertNull(StockholmFile.typeToDescription(null)); + } + + @Test(groups = { "Functional" }) + public void testDescriptionToType() + { + assertEquals("SS", + StockholmFile.descriptionToType("Secondary Structure")); + assertEquals("SA", + StockholmFile.descriptionToType("Surface Accessibility")); + assertEquals("TM", StockholmFile.descriptionToType("transmembrane")); + + // test is not case-sensitive: + assertEquals("SS", + StockholmFile.descriptionToType("secondary structure")); + + // test is white-space sensitive: + assertNull(StockholmFile.descriptionToType("secondary structure ")); + + assertNull(StockholmFile.descriptionToType("any old junk")); + assertNull(StockholmFile.descriptionToType("")); + assertNull(StockholmFile.descriptionToType(null)); + } + + @Test(groups = { "Functional" }) + public void testPrint() + { + SequenceI seq1 = new Sequence("seq1", "LKMF-RS-Q"); + SequenceI seq2 = new Sequence("seq2/10-15", "RRS-LIP-"); + SequenceI[] seqs = new SequenceI[] { seq1, seq2 }; + AlignmentI al = new Alignment(seqs); + + StockholmFile testee = new StockholmFile(al); + + /* + * basic output (sequences only): + * sequence ids are padded with 9 spaces more than the widest id + */ + String output = testee.print(seqs, true); + String expected = "# STOCKHOLM 1.0\n" + "seq1/1-7 LKMF-RS-Q\n" + + "seq2/10-15 RRS-LIP-\n//\n"; + assertEquals(expected, output); + + /* + * add some dbrefs + */ + seq1.addDBRef(new DBRefEntry("PFAM", "1", "PF00111")); + seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "P83527")); + seq2.addDBRef(new DBRefEntry("RFAM", "1", "AY119185.1")); + seq2.addDBRef(new DBRefEntry("EMBL", "1", "AF125575")); + output = testee.print(seqs, true); + // PFAM and RFAM dbrefs should be output as AC, others as DR + expected = "# STOCKHOLM 1.0\n" + "#=GS seq1/1-7 AC PF00111\n" + + "#=GS seq1/1-7 DR UNIPROT ; P83527\n" + + "#=GS seq2/10-15 AC AY119185.1\n" + + "#=GS seq2/10-15 DR EMBL ; AF125575\n" + + "seq1/1-7 LKMF-RS-Q\n" + + "seq2/10-15 RRS-LIP-\n//\n"; + assertEquals(expected, output); + + /* + * add some sequence and alignment annotation + */ + Annotation[] anns = new Annotation[5]; + for (int i = 0; i < anns.length; i++) + { + anns[i] = new Annotation(String.valueOf((char) ('B' + i)), + "Desc " + i, + (char) ('C' + i), i + 3); + } + + // expect "secondary structure" to be output as #=GR seqid SS + // using the secondary structure character (CDEFG) not display char (BCDEF) + AlignmentAnnotation aa1 = new AlignmentAnnotation("secondary structure", + "ssdesc", anns); + aa1.sequenceRef = seq1; + seq1.addAlignmentAnnotation(aa1); + al.addAnnotation(aa1); + + // "sec structure" should not be output as no corresponding feature id + AlignmentAnnotation aa2 = new AlignmentAnnotation("sec structure", + "ssdesc", anns); + aa2.sequenceRef = seq2; + seq2.addAlignmentAnnotation(aa2); + al.addAnnotation(aa2); + + // alignment annotation for Reference Positions: output as #=GC RF + AlignmentAnnotation aa3 = new AlignmentAnnotation("reference positions", + "refpos", anns); + al.addAnnotation(aa3); + + // 'seq' annotation: output as seq_cons + AlignmentAnnotation aa4 = new AlignmentAnnotation("seq", "seqdesc", + anns); + al.addAnnotation(aa4); + + // 'intron' annotation: output as IN_cons + AlignmentAnnotation aa5 = new AlignmentAnnotation("intron", + "introndesc", anns); + al.addAnnotation(aa5); + + // 'binding site' annotation: output as binding_site + AlignmentAnnotation aa6 = new AlignmentAnnotation("binding site", + "bindingdesc", anns); + al.addAnnotation(aa6); + + // 'autocalc' annotation should not be output + AlignmentAnnotation aa7 = new AlignmentAnnotation("Consensus", + "consensusdesc", anns); + aa7.autoCalculated = true; + al.addAnnotation(aa7); + + // hidden annotation should not be output + AlignmentAnnotation aa8 = new AlignmentAnnotation("domains", + "domaindesc", anns); + aa8.visible = false; + al.addAnnotation(aa8); + + output = testee.print(seqs, true); + //@formatter:off + expected = + "# STOCKHOLM 1.0\n" + + "#=GS seq1/1-7 AC PF00111\n" + + "#=GS seq1/1-7 DR UNIPROT ; P83527\n" + + "#=GS seq2/10-15 AC AY119185.1\n" + + "#=GS seq2/10-15 DR EMBL ; AF125575\n" + + "#=GR seq1/1-7 SS CDEFG\n" + + "seq1/1-7 LKMF-RS-Q\n" + + "seq2/10-15 RRS-LIP-\n" + + "#=GC RF BCDEF\n" + "#=GC seq_cons BCDEF\n" + + "#=GC IN_cons BCDEF\n" + "#=GC binding_site BCDEF\n" + + "//\n"; + //@formatter:on + assertEquals(expected, output); + } } -- 1.7.10.2