X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Fio%2FStockholmFile.java;h=f7a2df257a2f911580651ae4f7a48153fee5a8ce;hb=1beac3545a78d4c5c3274dbb53296708d693efe0;hp=3fd0505738e23bbdd0da3dde1e7f9bd8868c491b;hpb=8b27085fa7fc5f2877e078421284c2636b85b8c6;p=jalview.git diff --git a/src/jalview/io/StockholmFile.java b/src/jalview/io/StockholmFile.java index 3fd0505..f7a2df2 100644 --- a/src/jalview/io/StockholmFile.java +++ b/src/jalview/io/StockholmFile.java @@ -33,6 +33,7 @@ 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; @@ -45,7 +46,6 @@ import java.util.Hashtable; import java.util.LinkedHashMap; import java.util.List; import java.util.Map; -import java.util.StringTokenizer; import java.util.Vector; import com.stevesoft.pat.Regex; @@ -74,23 +74,38 @@ import fr.orsay.lri.varna.models.rna.RNA; */ 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 DETECT_BRACKETS = 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( + "(<|>|\\[|\\]|\\(|\\)|\\{|\\})"); + + // WUSS extended symbols. Avoid ambiguity with protein SS annotations by using NOT_RNASS first. + public static final String RNASS_BRACKETS = "<>[](){}AaBbCcDdEeFfGgHhIiJjKkLlMmNnOoPpQqRrSsTtUuVvWwXxYyZz"; + + // use the following regex to decide an annotations (whole) line is NOT an RNA + // SS (it contains only E,H,e,h and other non-brace/non-alpha chars) + private static final Regex NOT_RNASS = new Regex( + "^[^<>[\\](){}A-DF-Za-df-z]*$"); StringBuffer out; // output buffer - AlignmentI al; + private AlignmentI al; public StockholmFile() { } /** - * Creates a new StockholmFile object for output. + * Creates a new StockholmFile object for output */ public StockholmFile(AlignmentI al) { @@ -163,8 +178,8 @@ 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 +210,7 @@ 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 +222,8 @@ 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 +232,7 @@ 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 @@ -366,6 +380,11 @@ public class StockholmFile extends AlignFile // add alignment annotation for this feature String key = type2id(type); + + /* + * have we added annotation rows for this type ? + */ + boolean annotsAdded = false; if (key != null) { if (accAnnotations != null @@ -374,6 +393,7 @@ 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 +406,11 @@ 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 +426,7 @@ 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 +453,8 @@ 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 +597,11 @@ 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 +616,12 @@ 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")) @@ -622,24 +640,28 @@ public class StockholmFile extends AlignFile if (features.containsKey(this.id2type(type))) { // logger.debug("Found content for " + this.id2type(type)); - content = (Hashtable) features.get(this.id2type(type)); + content = (Hashtable) features + .get(this.id2type(type)); } else { // logger.debug("Creating new content holder for " + // this.id2type(type)); content = new Hashtable(); - features.put(this.id2type(type), content); + features.put(id2type(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 +672,22 @@ 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 +695,8 @@ 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 +724,7 @@ 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 +809,10 @@ 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 +823,43 @@ 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, convert2 = null; + // 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; + boolean ss = false, posterior = false; type = id2type(type); - if (type.equals("secondary structure")) + + boolean isrnass = false; + + if (type.equalsIgnoreCase("secondary structure")) { ss = true; + isrnass = !NOT_RNASS.search(annots); // sorry about the double negative + // here (it's easier for dealing with + // other non-alpha-non-brace chars) + } + 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 @@ -828,27 +867,46 @@ public class StockholmFile extends AlignFile { // if (" .-_".indexOf(pos) == -1) { - if (DETECT_BRACKETS.search(pos)) + if (isrnass && RNASS_BRACKETS.indexOf(pos) >= 0) { - ann.secondaryStructure = Rna.getRNASecStrucState( - pos).charAt(0); + 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; @@ -897,10 +955,7 @@ public class StockholmFile extends AlignFile while ((in < s.length) && (s[in] != null)) { String tmp = printId(s[in], jvSuffix); - if (s[in].getSequence().length > max) - { - max = s[in].getSequence().length; - } + max = Math.max(max, s[in].getLength()); if (tmp.length() > maxid) { @@ -949,8 +1004,8 @@ public class StockholmFile extends AlignFile { Object idd = en.nextElement(); String type = (String) dataRef.remove(idd); - out.append(new Format("%-" + (maxid - 2) + "s").form("#=GS " - + idd.toString() + " ")); + out.append(new Format("%-" + (maxid - 2) + "s") + .form("#=GS " + idd.toString() + " ")); if (type.contains("PFAM") || type.contains("RFAM")) { @@ -967,35 +1022,34 @@ public class StockholmFile extends AlignFile // output annotations while (i < s.length && s[i] != null) { - if (s[i].getDatasetSequence() != null) + AlignmentAnnotation[] alAnot = s[i].getAnnotation(); + if (alAnot != null) { - SequenceI ds = s[i].getDatasetSequence(); - AlignmentAnnotation[] alAnot; Annotation[] ann; - Annotation annot; - alAnot = s[i].getAnnotation(); - String feature = ""; - if (alAnot != null) + + for (int j = 0; j < alAnot.length; j++) { - for (int j = 0; j < alAnot.length; j++) + if (alAnot[j].annotations != null) { - if (ds.getSequenceFeatures() != null) + String key = type2id(alAnot[j].label); + boolean isrna = alAnot[j].isValidStruc(); + + if (isrna) { - feature = ds.getSequenceFeatures()[0].type; + // hardwire to secondary structure if there is RNA secondary + // structure on the annotation + key = "SS"; } - // ?bug - feature may still have previous loop value - String key = type2id(feature); - if (key == null) { + continue; } // out.append("#=GR "); - out.append(new Format("%-" + maxid + "s").form("#=GR " - + printId(s[i], jvSuffix) + " " + key + " ")); + out.append(new Format("%-" + maxid + "s").form( + "#=GR " + printId(s[i], jvSuffix) + " " + key + " ")); ann = alAnot[j].annotations; - boolean isrna = alAnot[j].isValidStruc(); String seq = ""; for (int k = 0; k < ann.length; k++) { @@ -1005,6 +1059,7 @@ public class StockholmFile extends AlignFile out.append(newline); } } + } out.append(new Format("%-" + maxid + "s") @@ -1050,8 +1105,8 @@ public class StockholmFile extends AlignFile } label = label.replace(" ", "_"); - out.append(new Format("%-" + maxid + "s").form("#=GC " + label - + " ")); + out.append( + new Format("%-" + maxid + "s").form("#=GC " + label + " ")); boolean isrna = aa.isValidStruc(); for (int j = 0; j < aa.annotations.length; j++) { @@ -1068,6 +1123,7 @@ public class StockholmFile extends AlignFile return out.toString(); } + /** * add an annotation character to the output row * @@ -1083,23 +1139,39 @@ public class StockholmFile extends AlignFile { 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 == null + ? String.valueOf(annot.secondaryStructure) + : annot.displayCharacter); + if (ch == null) + { + ch = " "; + } if (key != null && key.equals("SS")) { + char ssannotchar = ' '; + boolean charset = false; if (annot == null) { - // sensible gap character if one is available or make one up - return sequenceI == null ? '-' : sequenceI.getCharAt(k); + // sensible gap character + ssannotchar = ' '; + charset = true; } else { // valid secondary structure AND no alternative label (e.g. ' B') if (annot.secondaryStructure > ' ' && ch.length() < 2) { - return annot.secondaryStructure; + ssannotchar = annot.secondaryStructure; + charset = true; } } + if (charset) + { + return (ssannotchar == ' ' && isrna) ? '.' : ssannotchar; + } } if (ch.length() == 0) @@ -1114,9 +1186,31 @@ public class StockholmFile extends AlignFile { seq = ch.charAt(1); } - return seq; + + return (seq == ' ' && key != null && key.equals("SS") && isrna) ? '.' + : seq; } + /** + * make a friendly ID string. + * + * @param dataName + * @return truncated dataName to after last '/' + */ + private String safeName(String dataName) + { + int b = 0; + while ((b = dataName.indexOf("/")) > -1 && b < dataName.length()) + { + dataName = dataName.substring(b + 1).trim(); + + } + int e = (dataName.length() - dataName.indexOf(".")) + 1; + dataName = dataName.substring(1, e).trim(); + return dataName; + } + + public String print() { out = new StringBuffer(); @@ -1136,10 +1230,10 @@ public class StockholmFile extends AlignFile if (typeIds == null) { typeIds = new Hashtable(); - typeIds.put("SS", "secondary structure"); - typeIds.put("SA", "surface accessibility"); + typeIds.put("SS", "Secondary Structure"); + typeIds.put("SA", "Surface Accessibility"); typeIds.put("TM", "transmembrane"); - typeIds.put("PP", "posterior probability"); + typeIds.put("PP", "Posterior Probability"); typeIds.put("LI", "ligand binding"); typeIds.put("AS", "active site"); typeIds.put("IN", "intron"); @@ -1150,19 +1244,19 @@ public class StockholmFile extends AlignFile typeIds.put("DE", "description"); typeIds.put("DR", "reference"); typeIds.put("LO", "look"); - typeIds.put("RF", "reference positions"); + 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); + System.err.println( + "Warning : Unknown Stockholm annotation type code " + id); return id; } @@ -1173,7 +1267,7 @@ public class StockholmFile extends AlignFile while (e.hasMoreElements()) { Object ll = e.nextElement(); - if (typeIds.get(ll).toString().equals(type)) + if (typeIds.get(ll).toString().equalsIgnoreCase(type)) { key = (String) ll; break; @@ -1183,27 +1277,8 @@ public class StockholmFile extends AlignFile { return key; } - System.err.println("Warning : Unknown Stockholm annotation type: " - + type); + System.err.println( + "Warning : Unknown Stockholm annotation type: " + type); return key; } - - /** - * make a friendly ID string. - * - * @param dataName - * @return truncated dataName to after last '/' - */ - private String safeName(String dataName) - { - int b = 0; - while ((b = dataName.indexOf("/")) > -1 && b < dataName.length()) - { - dataName = dataName.substring(b + 1).trim(); - - } - int e = (dataName.length() - dataName.indexOf(".")) + 1; - dataName = dataName.substring(1, e).trim(); - return dataName; - } }