X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Fio%2FStockholmFile.java;h=58b171da24b96ef58fd6e1be6d2c4f9a676028f4;hb=d25ed8f9fb513fe5c33823afbeeb557c73a8c27b;hp=3b0dab41edeeb0b916f8920033dab51e5cb3baa7;hpb=6684899d56d5bfb61faf814560010d7cb7b8bd95;p=jalview.git diff --git a/src/jalview/io/StockholmFile.java b/src/jalview/io/StockholmFile.java index 3b0dab4..58b171d 100644 --- a/src/jalview/io/StockholmFile.java +++ b/src/jalview/io/StockholmFile.java @@ -28,11 +28,13 @@ 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; import jalview.datamodel.SequenceI; import jalview.schemes.ResidueProperties; +import jalview.util.Comparison; import jalview.util.Format; import jalview.util.MessageManager; @@ -41,11 +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.StringTokenizer; +import java.util.Map.Entry; import java.util.Vector; import com.stevesoft.pat.Regex; @@ -74,23 +77,50 @@ import fr.orsay.lri.varna.models.rna.RNA; */ public class StockholmFile extends AlignFile { - private static final Regex OPEN_PAREN = new Regex("(<|\\[)", "("); + private static final String ANNOTATION = "annotation"; - private static final Regex CLOSE_PAREN = new Regex("(>|\\])", ")"); + private static final char UNDERSCORE = '_'; - private static final Regex DETECT_BRACKETS = new Regex( - "(<|>|\\[|\\]|\\(|\\))"); + // private static final Regex OPEN_PAREN = new Regex("(<|\\[)", "("); + // private static final Regex CLOSE_PAREN = new Regex("(>|\\])", ")"); - StringBuffer out; // output buffer + public static final Regex DETECT_BRACKETS = new Regex( + "(<|>|\\[|\\]|\\(|\\)|\\{|\\})"); - AlignmentI al; + /* + * 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"); + } + + private AlignmentI al; public StockholmFile() { } /** - * Creates a new StockholmFile object for output. + * Creates a new StockholmFile object for output */ public StockholmFile(AlignmentI al) { @@ -108,6 +138,47 @@ 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() { @@ -163,8 +234,8 @@ public class StockholmFile extends AlignFile for (int k = 0; k < rna.length(); k++) { - ann[k] = new Annotation(annot[k], "", Rna.getRNASecStrucState( - annot[k]).charAt(0), 0f); + ann[k] = new Annotation(annot[k], "", + Rna.getRNASecStrucState(annot[k]).charAt(0), 0f); } AlignmentAnnotation align = new AlignmentAnnotation("Sec. str.", @@ -195,7 +266,7 @@ public class StockholmFile extends AlignFile String version; // String id; Hashtable seqAnn = new Hashtable(); // Sequence related annotations - LinkedHashMap seqs = new LinkedHashMap(); + LinkedHashMap seqs = new LinkedHashMap<>(); Regex p, r, rend, s, x; // Temporary line for processing RNA annotation // String RNAannot = ""; @@ -207,9 +278,8 @@ public class StockholmFile extends AlignFile r = new Regex("# STOCKHOLM ([\\d\\.]+)"); if (!r.search(nextLine())) { - throw new IOException( - MessageManager - .getString("exception.stockholm_invalid_format")); + throw new IOException(MessageManager + .getString("exception.stockholm_invalid_format")); } else { @@ -218,7 +288,7 @@ 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 @@ -231,7 +301,7 @@ 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(); @@ -365,7 +435,12 @@ 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 ? + */ + boolean annotsAdded = false; if (key != null) { if (accAnnotations != null @@ -374,6 +449,7 @@ public class StockholmFile extends AlignFile Vector vv = (Vector) accAnnotations.get(key); for (int ii = 0; ii < vv.size(); ii++) { + annotsAdded = true; AlignmentAnnotation an = (AlignmentAnnotation) vv .elementAt(ii); seqO.addAlignmentAnnotation(an); @@ -386,6 +462,11 @@ public class StockholmFile extends AlignFile while (j.hasMoreElements()) { String desc = j.nextElement().toString(); + if (ANNOTATION.equals(desc) && annotsAdded) + { + // don't add features if we already added an annotation row + continue; + } String ns = content.get(desc).toString(); char[] byChar = ns.toCharArray(); for (int k = 0; k < byChar.length; k++) @@ -401,7 +482,7 @@ public class StockholmFile extends AlignFile int new_pos = posmap[k]; // look up nearest seqeunce // position to this column SequenceFeature feat = new SequenceFeature(type, desc, - new_pos, new_pos, 0f, null); + new_pos, new_pos, null); seqO.addSequenceFeature(feat); } @@ -428,8 +509,8 @@ public class StockholmFile extends AlignFile { // logger.error("Could not parse sequence line: " + line); throw new IOException(MessageManager.formatMessage( - "exception.couldnt_parse_sequence_line", - new String[] { line })); + "exception.couldnt_parse_sequence_line", new String[] + { line })); } String ns = seqs.get(x.stringMatched(1)); if (ns == null) @@ -572,22 +653,11 @@ public class StockholmFile extends AlignFile { String acc = s.stringMatched(1); String type = s.stringMatched(2); - String seq = new String(s.stringMatched(3)); - String description = null; - // Check for additional information about the current annotation - // We use a simple string tokenizer here for speed - StringTokenizer sep = new StringTokenizer(seq, " \t"); - description = sep.nextToken(); - if (sep.hasMoreTokens()) - { - seq = sep.nextToken(); - } - else - { - seq = description; - description = new String(); - } - // sequence id with from-to fields + String oseq = s.stringMatched(3); + /* + * copy of annotation field that may be processed into whitespace chunks + */ + String seq = new String(oseq); Hashtable ann; // Get an object with all the annotations for this sequence @@ -602,8 +672,12 @@ public class StockholmFile extends AlignFile ann = new Hashtable(); seqAnn.put(acc, ann); } + + // // start of block for appending annotation lines for wrapped + // stokchholm file // TODO test structure, call parseAnnotationRow with vector from // hashtable for specific sequence + Hashtable features; // Get an object with all the content for an annotation if (ann.containsKey("features")) @@ -619,27 +693,31 @@ 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(description); + String ns = (String) content.get(ANNOTATION); + if (ns == null) { ns = ""; } + // finally, append the annotation line ns += seq; - content.put(description, ns); + content.put(ANNOTATION, ns); + // // end of wrapped annotation block. + // // Now a new row is created with the current set of data - // if(type.equals("SS")){ Hashtable strucAnn; if (seqAnn.containsKey(acc)) { @@ -650,21 +728,22 @@ public class StockholmFile extends AlignFile strucAnn = new Hashtable(); } - Vector newStruc = new Vector(); + Vector newStruc = new Vector<>(); parseAnnotationRow(newStruc, type, ns); for (AlignmentAnnotation alan : newStruc) { alan.visible = false; } - // annotations.addAll(newStruc); + // new annotation overwrites any existing annotation... + strucAnn.put(type, newStruc); seqAnn.put(acc, strucAnn); } // } else { - System.err - .println("Warning - couldn't parse sequence annotation row line:\n" + System.err.println( + "Warning - couldn't parse sequence annotation row line:\n" + line); // throw new IOException("Error parsing " + line); } @@ -672,8 +751,8 @@ public class StockholmFile extends AlignFile else { throw new IOException(MessageManager.formatMessage( - "exception.unknown_annotation_detected", new String[] { - annType, annContent })); + "exception.unknown_annotation_detected", new String[] + { annType, annContent })); } } } @@ -701,7 +780,7 @@ public class StockholmFile extends AlignFile private void guessDatabaseFor(Sequence seqO, String dbr, String dbsource) { DBRefEntry dbrf = null; - List dbrs = new ArrayList(); + List dbrs = new ArrayList<>(); String seqdb = "Unknown", sdbac = "" + dbr; int st = -1, en = -1, p; if ((st = sdbac.indexOf("/")) > -1) @@ -786,8 +865,10 @@ public class StockholmFile extends AlignFile { for (DBRefEntry d : dbrs) { - jalview.util.MapList mp = new jalview.util.MapList(new int[] { - seqO.getStart(), seqO.getEnd() }, new int[] { st, en }, 1, 1); + jalview.util.MapList mp = new jalview.util.MapList( + new int[] + { seqO.getStart(), seqO.getEnd() }, new int[] { st, en }, 1, + 1); jalview.datamodel.Mapping mping = new Mapping(mp); d.setMap(mping); } @@ -798,29 +879,36 @@ 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); - annots = convert2; + // String convert1 = OPEN_PAREN.replaceAll(annots); + // String convert2 = CLOSE_PAREN.replaceAll(convert1); + // annots = convert2; String type = label; if (label.contains("_cons")) { - type = (label.indexOf("_cons") == label.length() - 5) ? label - .substring(0, label.length() - 5) : label; + type = (label.indexOf("_cons") == label.length() - 5) + ? label.substring(0, label.length() - 5) + : label; } - boolean ss = false; - type = id2type(type); - if (type.equals("secondary structure")) + boolean ss = false, posterior = false; + type = typeToDescription(type); + if (type.equalsIgnoreCase("secondary structure")) { ss = true; } + if (type.equalsIgnoreCase("posterior probability")) + { + posterior = true; + } // decide on secondary structure or not. Annotation[] els = new Annotation[annots.length()]; 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 @@ -831,23 +919,43 @@ public class StockholmFile extends AlignFile if (DETECT_BRACKETS.search(pos)) { ann.secondaryStructure = Rna.getRNASecStrucState(pos).charAt(0); + ann.displayCharacter = "" + pos.charAt(0); } else { ann.secondaryStructure = ResidueProperties.getDssp3state(pos) .charAt(0); - } - if (ann.secondaryStructure == pos.charAt(0)) - { - ann.displayCharacter = ""; // null; // " "; + if (ann.secondaryStructure == pos.charAt(0)) + { + ann.displayCharacter = ""; // null; // " "; + } + else + { + ann.displayCharacter = " " + ann.displayCharacter; + } } - else + } + + } + if (posterior && !ann.isWhitespace() + && !Comparison.isGap(pos.charAt(0))) + { + float val = 0; + // symbol encodes values - 0..*==0..10 + if (pos.charAt(0) == '*') + { + val = 10; + } + else + { + val = pos.charAt(0) - '0'; + if (val > 9) { - ann.displayCharacter = " " + ann.displayCharacter; + val = 10; } } - + ann.value = val; } els[i] = ann; @@ -882,153 +990,123 @@ 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); - if (s[in].getSequence().length > max) + if (seq != null) { - max = s[in].getSequence().length; - } - - if (tmp.length() > maxid) - { - maxid = tmp.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); - } + String formattedId = printId(seq, jvSuffix); + maxIdWidth = Math.max(maxIdWidth, formattedId.length()); } - 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) { - if (s[i].getDatasetSequence() != null) + if (seq != null) { - SequenceI ds = s[i].getDatasetSequence(); - AlignmentAnnotation[] alAnot; - Annotation[] ann; - Annotation annot; - alAnot = s[i].getAnnotation(); - String feature = ""; + AlignmentAnnotation[] alAnot = seq.getAnnotation(); if (alAnot != null) { for (int j = 0; j < alAnot.length; j++) { - - String key = type2id(alAnot[j].label); - boolean isrna = alAnot[j].isValidStruc(); - + AlignmentAnnotation ann = alAnot[j]; + String key = descriptionToType(ann.label); + boolean isrna = ann.isValidStruc(); if (isrna) { - // hardwire to secondary structure if there is RNA secondary - // structure on the annotation + /* + * 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++) + 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++) { - seq += outputCharacter(key, k, isrna, ann, s[i]); + seqString + .append(getAnnotationCharacter(key, k, anns[k], seq)); } - out.append(seq); + out.append(seqString.toString()); 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")) { @@ -1036,30 +1114,28 @@ 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(); + out.append( + 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); } } @@ -1071,28 +1147,50 @@ 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))) : annot.displayCharacter; + String ch = (annot == null) + ? ((sequenceI == null) ? "-" + : Character.toString(sequenceI.getCharAt(k))) + : annot.displayCharacter; if (key != null && key.equals("SS")) { if (annot == null) { - // sensible gap character if one is available or make one up - return sequenceI == null ? '-' : sequenceI.getCharAt(k); + // Stockholm format requires underscore, not space + return UNDERSCORE; } else { @@ -1119,77 +1217,6 @@ public class StockholmFile extends AlignFile return seq; } - public String print() - { - out = new StringBuffer(); - out.append("# STOCKHOLM 1.0"); - out.append(newline); - print(getSeqsAsArray(), false); - - 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().equals(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. *