X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Fio%2FStockholmFile.java;h=84e629e043e85055e787ef5d33bc6d4d90d81e58;hb=47806656477f02076cac0467c7374e231a59ab7e;hp=e22ee0a40800dc58e93b64a04d54dbc29f127bb0;hpb=3d0101179759ef157b088ea135423cd909512d9f;p=jalview.git diff --git a/src/jalview/io/StockholmFile.java b/src/jalview/io/StockholmFile.java index e22ee0a..84e629e 100644 --- a/src/jalview/io/StockholmFile.java +++ b/src/jalview/io/StockholmFile.java @@ -74,13 +74,23 @@ 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 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; @@ -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 = ""; @@ -229,8 +239,8 @@ public class StockholmFile extends AlignFile Regex openparen = new Regex("(<|\\[)", "("); Regex closeparen = new Regex("(>|\\])", ")"); - // Detect if file is RNA by looking for bracket types - Regex detectbrackets = new Regex("(<|>|\\[|\\]|\\(|\\))"); +// // Detect if file is RNA by looking for bracket types +// Regex detectbrackets = new Regex("(<|>|\\[|\\]|\\(|\\))"); rend.optimize(); p.optimize(); @@ -251,7 +261,7 @@ public class StockholmFile extends AlignFile // End of the alignment, pass stuff back this.noSeqs = seqs.size(); - String seqdb, dbsource = null; + String dbsource = null; Regex pf = new Regex("PF[0-9]{5}(.*)"); // Finds AC for Pfam Regex rf = new Regex("RF[0-9]{5}(.*)"); // Finds AC for Rfam if (getAlignmentProperty("AC") != null) @@ -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); @@ -926,31 +941,33 @@ public class StockholmFile extends AlignFile int max = 0; int maxid = 0; int in = 0; - Hashtable dataRef = null; - while ((in < s.length) && (s[in] != null)) + int slen = s.length; + SequenceI seq; + Hashtable dataRef = null; + while ((in < slen) && ((seq = s[in]) != null)) { - String tmp = printId(s[in], jvSuffix); - if (s[in].getSequence().length > max) - { - max = s[in].getSequence().length; - } + String tmp = printId(seq, jvSuffix); + max = Math.max(max, seq.getLength()); if (tmp.length() > maxid) { maxid = tmp.length(); } - if (s[in].getDBRefs() != null) + List seqrefs = seq.getDBRefs(); + int ndb; + if (seqrefs != null && (ndb = seqrefs.size()) > 0) { - for (int idb = 0; idb < s[in].getDBRefs().length; idb++) + if (dataRef == null) + { + dataRef = new Hashtable<>(); + } + for (int idb = 0; idb < ndb; idb++) { - if (dataRef == null) - { - dataRef = new Hashtable(); - } - String datAs1 = s[in].getDBRefs()[idb].getSource().toString() + DBRefEntry ref = seqrefs.get(idb); + String datAs1 = ref.getSource().toString() + " ; " - + s[in].getDBRefs()[idb].getAccessionId().toString(); + + ref.getAccessionId().toString(); dataRef.put(tmp, datAs1); } } @@ -977,7 +994,7 @@ public class StockholmFile extends AlignFile // output database accessions if (dataRef != null) { - Enumeration en = dataRef.keys(); + Enumeration en = dataRef.keys(); while (en.hasMoreElements()) { Object idd = en.nextElement(); @@ -998,13 +1015,13 @@ public class StockholmFile extends AlignFile } // output annotations - while (i < s.length && s[i] != null) + while (i < slen && (seq = s[i]) != null) { - AlignmentAnnotation[] alAnot = s[i].getAnnotation(); + AlignmentAnnotation[] alAnot = seq.getAnnotation(); if (alAnot != null) { Annotation[] ann; - for (int j = 0; j < alAnot.length; j++) + for (int j = 0, nj = alAnot.length; j < nj; j++) { String key = type2id(alAnot[j].label); @@ -1024,37 +1041,38 @@ public class StockholmFile extends AlignFile // out.append("#=GR "); out.append(new Format("%-" + maxid + "s").form( - "#=GR " + printId(s[i], jvSuffix) + " " + key + " ")); + "#=GR " + printId(seq, jvSuffix) + " " + key + " ")); ann = alAnot[j].annotations; - String seq = ""; - for (int k = 0; k < ann.length; k++) + String sseq = ""; + for (int k = 0, nk = ann.length; k < nk; k++) { - seq += outputCharacter(key, k, isrna, ann, s[i]); + sseq += outputCharacter(key, k, isrna, ann, seq); } - out.append(seq); + out.append(sseq); out.append(newline); } } out.append(new Format("%-" + maxid + "s") - .form(printId(s[i], jvSuffix) + " ")); - out.append(s[i].getSequenceAsString()); + .form(printId(seq, jvSuffix) + " ")); + out.append(seq.getSequenceAsString()); out.append(newline); i++; } // alignment annotation AlignmentAnnotation aa; - if (al.getAlignmentAnnotation() != null) + AlignmentAnnotation[] an = al.getAlignmentAnnotation(); + if (an != null) { - for (int ia = 0; ia < al.getAlignmentAnnotation().length; ia++) + for (int ia = 0, na = an.length; ia < na; ia++) { - aa = al.getAlignmentAnnotation()[ia]; + aa = an[ia]; if (aa.autoCalculated || !aa.visible || aa.sequenceRef != null) { continue; } - String seq = ""; + String sseq = ""; String label; String key = ""; if (aa.label.equals("seq")) @@ -1082,11 +1100,11 @@ public class StockholmFile extends AlignFile out.append( new Format("%-" + maxid + "s").form("#=GC " + label + " ")); boolean isrna = aa.isValidStruc(); - for (int j = 0; j < aa.annotations.length; j++) + for (int j = 0, nj = aa.annotations.length; j < nj; j++) { - seq += outputCharacter(key, j, isrna, aa.annotations, null); + sseq += outputCharacter(key, j, isrna, aa.annotations, null); } - out.append(seq); + out.append(sseq); out.append(newline); } } @@ -1115,22 +1133,36 @@ 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 +1177,9 @@ public class StockholmFile extends AlignFile { seq = ch.charAt(1); } - return seq; + + return (seq == ' ' && key != null && key.equals("SS") && isrna) ? '.' + : seq; } public String print()