From 247737b13c78af1123f9e41edc44d9f16c1a2358 Mon Sep 17 00:00:00 2001 From: tva Date: Wed, 5 Jun 2019 12:06:31 +0100 Subject: [PATCH] JAL-3285 add methods and small changes to StockholmFile left out by git --- src/jalview/io/StockholmFile.java | 387 +++++++++++++++++++------------------ 1 file changed, 200 insertions(+), 187 deletions(-) diff --git a/src/jalview/io/StockholmFile.java b/src/jalview/io/StockholmFile.java index 61f28a9..3afb967 100644 --- a/src/jalview/io/StockholmFile.java +++ b/src/jalview/io/StockholmFile.java @@ -80,6 +80,10 @@ public class StockholmFile extends AlignFile private static final String ANNOTATION = "annotation"; private static final char UNDERSCORE = '_'; + + 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("(>|\\])", ")"); @@ -87,12 +91,6 @@ public class StockholmFile extends AlignFile public static final Regex DETECT_BRACKETS = new Regex( "(<|>|\\[|\\]|\\(|\\)|\\{|\\})"); - /* - * lookup table of Stockholm 'feature' (annotation) types - * see http://sonnhammer.sbc.su.se/Stockholm.html - */ - private static Map featureTypes = null; - // WUSS extended symbols. Avoid ambiguity with protein SS annotations by using NOT_RNASS first. public static final String RNASS_BRACKETS = "<>[](){}AaBbCcDdEeFfGgHhIiJjKkLlMmNnOoPpQqRrSsTtUuVvWwXxYyZz"; @@ -103,26 +101,6 @@ public class StockholmFile extends AlignFile StringBuffer out; // output buffer - 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"); - } - private AlignmentI al; public StockholmFile() @@ -148,47 +126,6 @@ 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() { @@ -311,7 +248,7 @@ 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(); @@ -445,7 +382,7 @@ public class StockholmFile extends AlignFile Hashtable content = (Hashtable) features.remove(type); // add alignment annotation for this feature - String key = descriptionToType(type); + String key = type2id(type); /* * have we added annotation rows for this type ? @@ -703,18 +640,18 @@ public class StockholmFile extends AlignFile } Hashtable content; - if (features.containsKey(StockholmFile.typeToDescription(type))) + if (features.containsKey(this.id2type(type))) { // logger.debug("Found content for " + this.id2type(type)); content = (Hashtable) features - .get(StockholmFile.typeToDescription(type)); + .get(this.id2type(type)); } else { // logger.debug("Creating new content holder for " + // this.id2type(type)); content = new Hashtable(); - features.put(StockholmFile.typeToDescription(type), content); + features.put(id2type(type), content); } String ns = (String) content.get(ANNOTATION); @@ -889,6 +826,7 @@ public class StockholmFile extends AlignFile Vector annotation, String label, String annots) { + String convert1, convert2 = null; // String convert1 = OPEN_PAREN.replaceAll(annots); // String convert2 = CLOSE_PAREN.replaceAll(convert1); // annots = convert2; @@ -1006,123 +944,143 @@ public class StockholmFile extends AlignFile } @Override - public String print(final SequenceI[] sequences, boolean jvSuffix) + public String print(SequenceI[] s, boolean jvSuffix) { - StringBuilder out = new StringBuilder(); + out = new StringBuffer(); out.append("# STOCKHOLM 1.0"); out.append(newline); - int maxIdWidth = 0; - for (SequenceI seq : sequences) + // find max length of id + int max = 0; + int maxid = 0; + int in = 0; + Hashtable dataRef = null; + while ((in < s.length) && (s[in] != null)) { - if (seq != null) + String tmp = printId(s[in], jvSuffix); + max = Math.max(max, s[in].getLength()); + + if (tmp.length() > maxid) + { + maxid = tmp.length(); + } + if (s[in].getDBRefs() != null) { - String formattedId = printId(seq, jvSuffix); - maxIdWidth = Math.max(maxIdWidth, formattedId.length()); + 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++; } - maxIdWidth += 9; + maxid += 9; + int i = 0; - /* - * generic alignment properties - */ - Hashtable props = al.getProperties(); - if (props != null) + // output database type + if (al.getProperties() != null) { - for (Object key : props.keySet()) + if (!al.getProperties().isEmpty()) { - out.append(String.format("#=GF %s %s", key.toString(), - props.get(key).toString())); - out.append(newline); + Enumeration key = al.getProperties().keys(); + Enumeration val = al.getProperties().elements(); + while (key.hasMoreElements()) + { + out.append("#=GF " + key.nextElement() + " " + val.nextElement()); + out.append(newline); + } } } - /* - * 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) + // output database accessions + if (dataRef != null) { - if (seq != null) + Enumeration en = dataRef.keys(); + while (en.hasMoreElements()) { - DBRefEntry[] dbRefs = seq.getDBRefs(); - if (dbRefs != 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")) { - String idField = formatter - .form("#=GS " + printId(seq, jvSuffix) + " "); - for (DBRefEntry dbRef : dbRefs) - { - out.append(idField); - printDbRef(out, dbRef); - } + + out.append(" AC " + type.substring(type.indexOf(";") + 1)); } + else + { + out.append(" DR " + type + " "); + } + out.append(newline); } } - /* - * output annotations - */ - for (SequenceI seq : sequences) + // output annotations + while (i < s.length && s[i] != null) { - if (seq != null) + AlignmentAnnotation[] alAnot = s[i].getAnnotation(); + if (alAnot != null) { - AlignmentAnnotation[] alAnot = seq.getAnnotation(); - if (alAnot != null) + Annotation[] ann; + for (int j = 0; j < alAnot.length; j++) { - for (int j = 0; j < alAnot.length; j++) + + 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) { - 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; - } - 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); + continue; } - } - out.append(new Format("%-" + maxIdWidth + "s") - .form(printId(seq, jvSuffix) + " ")); - out.append(seq.getSequenceAsString()); - 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++; } - /* - * output alignment annotation (but not auto-calculated or sequence-related) - */ + // alignment annotation + AlignmentAnnotation aa; if (al.getAlignmentAnnotation() != null) { for (int ia = 0; ia < al.getAlignmentAnnotation().length; ia++) { - AlignmentAnnotation aa = al.getAlignmentAnnotation()[ia]; + aa = al.getAlignmentAnnotation()[ia]; if (aa.autoCalculated || !aa.visible || aa.sequenceRef != null) { continue; } - String label = aa.label; + String seq = ""; + String label; String key = ""; if (aa.label.equals("seq")) { @@ -1130,28 +1088,30 @@ public class StockholmFile extends AlignFile } else { - key = descriptionToType(aa.label); - if ("RF".equals(key)) + key = type2id(aa.label.toLowerCase()); + if (key == null) { - label = key; + label = aa.label; } - else if (key != null) + else { label = key + "_cons"; } } + if (label == null) + { + label = aa.label; + } label = label.replace(" ", "_"); out.append( - new Format("%-" + maxIdWidth + "s") - .form("#=GC " + label + " ")); - StringBuilder sb = new StringBuilder(aa.annotations.length); + new Format("%-" + maxid + "s").form("#=GC " + label + " ")); + boolean isrna = aa.isValidStruc(); for (int j = 0; j < aa.annotations.length; j++) { - sb.append( - getAnnotationCharacter(key, j, aa.annotations[j], null)); + seq += outputCharacter(key, j, isrna, aa.annotations, null); } - out.append(sb.toString()); + out.append(seq); out.append(newline); } } @@ -1162,41 +1122,22 @@ public class StockholmFile extends AlignFile return out.toString(); } - /** - * 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 + * add an annotation character to the output row * * @param seq * @param key * @param k + * @param isrna * @param ann * @param sequenceI */ - static char getAnnotationCharacter(String key, int k, Annotation annot, - SequenceI sequenceI) + private char outputCharacter(String key, int k, boolean isrna, + Annotation[] ann, SequenceI sequenceI) { char seq = ' '; + Annotation annot = ann[k]; String ch = (annot == null) ? ((sequenceI == null) ? "-" : Character.toString(sequenceI.getCharAt(k))) @@ -1213,9 +1154,9 @@ public class StockholmFile extends AlignFile boolean charset = false; if (annot == null) { - // TODO: TVA: Check against develop for correct behaviour! - // Stockholm format requires underscore, not space - return UNDERSCORE; + // sensible gap character + ssannotchar = ' '; + charset = true; } else { @@ -1267,4 +1208,76 @@ public class StockholmFile extends AlignFile dataName = dataName.substring(1, e).trim(); return dataName; } + + + public String print() + { + out = new StringBuffer(); + out.append("# STOCKHOLM 1.0"); + out.append(newline); + print(getSeqsAsArray(), false); + + out.append("//"); + out.append(newline); + 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; + } } -- 1.7.10.2