X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Fio%2FStockholmFile.java;fp=src%2Fjalview%2Fio%2FStockholmFile.java;h=61f28a9361a454e8ec0ad916111f47bdfb4f39df;hb=b5667f39acdf309cd92881b73edfda591e0acaf4;hp=0e73af13cc025548f130c4db7f2cabbae6e8dab0;hpb=9e31a59b947882325ffe614bd71fb732d7952a23;p=jalview.git diff --git a/src/jalview/io/StockholmFile.java b/src/jalview/io/StockholmFile.java index 0e73af1..61f28a9 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; @@ -76,13 +79,20 @@ public class StockholmFile extends AlignFile { private static final String ANNOTATION = "annotation"; - private static final Regex OPEN_PAREN = new Regex("(<|\\[)", "("); + private static final char UNDERSCORE = '_'; - 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( "(<|>|\\[|\\]|\\(|\\)|\\{|\\})"); + /* + * 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"; @@ -93,14 +103,34 @@ public class StockholmFile extends AlignFile StringBuffer out; // output buffer - AlignmentI al; + 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() { } /** - * Creates a new StockholmFile object for output. + * Creates a new StockholmFile object for output */ public StockholmFile(AlignmentI al) { @@ -118,6 +148,47 @@ 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() { @@ -227,7 +298,7 @@ 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 @@ -240,7 +311,7 @@ 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(); @@ -374,7 +445,7 @@ 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 ? @@ -632,17 +703,18 @@ 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); @@ -817,10 +889,8 @@ 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; @@ -834,6 +904,7 @@ public class StockholmFile extends AlignFile type = id2type(type); boolean isrnass = false; + if (type.equalsIgnoreCase("secondary structure")) { ss = true; @@ -850,6 +921,10 @@ public class StockholmFile extends AlignFile for (int i = 0; i < annots.length(); i++) { String pos = annots.substring(i, i + 1); + if (UNDERSCORE == pos.charAt(0)) + { + pos = " "; + } Annotation ann; ann = new Annotation(pos, "", ' ', 0f); // 0f is 'valid' null - will not // be written out @@ -931,143 +1006,123 @@ 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")) - { - - out.append(" AC " + type.substring(type.indexOf(";") + 1)); - } - else + DBRefEntry[] dbRefs = seq.getDBRefs(); + if (dbRefs != null) { - 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) + for (int j = 0; j < alAnot.length; j++) { - // hardwire to secondary structure if there is RNA secondary - // structure on the annotation - key = "SS"; - } - if (key == null) - { - - continue; - } + 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("#=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(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(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")) { @@ -1075,30 +1130,28 @@ public class StockholmFile extends AlignFile } else { - key = type2id(aa.label.toLowerCase()); - if (key == null) + key = descriptionToType(aa.label); + if ("RF".equals(key)) { - label = aa.label; + 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); } } @@ -1110,20 +1163,40 @@ 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) + static 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))) @@ -1140,9 +1213,9 @@ public class StockholmFile extends AlignFile boolean charset = false; if (annot == null) { - // sensible gap character - ssannotchar = ' '; - charset = true; + // TODO: TVA: Check against develop for correct behaviour! + // Stockholm format requires underscore, not space + return UNDERSCORE; } else { @@ -1176,77 +1249,6 @@ public class StockholmFile extends AlignFile : seq; } - 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; - } - /** * make a friendly ID string. *