X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Fio%2FStockholmFile.java;h=0e73af13cc025548f130c4db7f2cabbae6e8dab0;hb=14208707a204bf8f7f2f714eebfdfd937094b943;hp=e22ee0a40800dc58e93b64a04d54dbc29f127bb0;hpb=3d0101179759ef157b088ea135423cd909512d9f;p=jalview.git diff --git a/src/jalview/io/StockholmFile.java b/src/jalview/io/StockholmFile.java index e22ee0a..0e73af1 100644 --- a/src/jalview/io/StockholmFile.java +++ b/src/jalview/io/StockholmFile.java @@ -74,6 +74,8 @@ import fr.orsay.lri.varna.models.rna.RNA; */ 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("(>|\\])", ")"); @@ -81,6 +83,14 @@ public class StockholmFile extends AlignFile 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; @@ -195,7 +205,7 @@ 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 = ""; @@ -391,7 +401,7 @@ public class StockholmFile extends AlignFile while (j.hasMoreElements()) { String desc = j.nextElement().toString(); - if ("annotations".equals(desc) && annotsAdded) + if (ANNOTATION.equals(desc) && annotsAdded) { // don't add features if we already added an annotation row continue; @@ -411,7 +421,7 @@ 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); } @@ -634,7 +644,7 @@ public class StockholmFile extends AlignFile content = new Hashtable(); features.put(this.id2type(type), content); } - String ns = (String) content.get("annotation"); + String ns = (String) content.get(ANNOTATION); if (ns == null) { @@ -642,7 +652,7 @@ public class StockholmFile extends AlignFile } // finally, append the annotation line ns += seq; - content.put("annotation", ns); + content.put(ANNOTATION, ns); // // end of wrapped annotation block. // // Now a new row is created with the current set of data @@ -656,7 +666,7 @@ public class StockholmFile extends AlignFile strucAnn = new Hashtable(); } - Vector newStruc = new Vector(); + Vector newStruc = new Vector<>(); parseAnnotationRow(newStruc, type, ns); for (AlignmentAnnotation alan : newStruc) { @@ -708,7 +718,7 @@ 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) @@ -822,9 +832,14 @@ public class StockholmFile extends AlignFile } boolean ss = false, posterior = false; type = id2type(type); + + 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")) { @@ -842,7 +857,7 @@ 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.displayCharacter = "" + pos.charAt(0); @@ -930,10 +945,7 @@ 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) { @@ -1115,22 +1127,36 @@ public class StockholmFile extends AlignFile String ch = (annot == null) ? ((sequenceI == null) ? "-" : Character.toString(sequenceI.getCharAt(k))) - : annot.displayCharacter; + : (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 - return ' '; + 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) @@ -1145,7 +1171,9 @@ public class StockholmFile extends AlignFile { seq = ch.charAt(1); } - return seq; + + return (seq == ' ' && key != null && key.equals("SS") && isrna) ? '.' + : seq; } public String print()