X-Git-Url: http://source.jalview.org/gitweb/?p=jalview.git;a=blobdiff_plain;f=src%2Fjalview%2Fio%2FStockholmFile.java;h=5d645cafe30874f85fe0b58a06619da2571c91c8;hp=8527f8b388a520250b7e034ebb25c790b244acb3;hb=0ae965fc73680bdc69b06986c060ff17342c4787;hpb=5edb9a0c935f7c153bbddedfe2ce36bb389dbb53 diff --git a/src/jalview/io/StockholmFile.java b/src/jalview/io/StockholmFile.java index 8527f8b..5d645ca 100644 --- a/src/jalview/io/StockholmFile.java +++ b/src/jalview/io/StockholmFile.java @@ -23,19 +23,6 @@ */ package jalview.io; -import jalview.analysis.Rna; -import jalview.datamodel.AlignmentAnnotation; -import jalview.datamodel.AlignmentI; -import jalview.datamodel.Annotation; -import jalview.datamodel.DBRefEntry; -import jalview.datamodel.Mapping; -import jalview.datamodel.Sequence; -import jalview.datamodel.SequenceFeature; -import jalview.datamodel.SequenceI; -import jalview.schemes.ResidueProperties; -import jalview.util.Format; -import jalview.util.MessageManager; - import java.io.BufferedReader; import java.io.FileReader; import java.io.IOException; @@ -44,8 +31,8 @@ import java.util.Enumeration; import java.util.Hashtable; import java.util.LinkedHashMap; import java.util.List; +import java.util.Locale; import java.util.Map; -import java.util.StringTokenizer; import java.util.Vector; import com.stevesoft.pat.Regex; @@ -53,8 +40,21 @@ import com.stevesoft.pat.Regex; import fr.orsay.lri.varna.exceptions.ExceptionUnmatchedClosingParentheses; import fr.orsay.lri.varna.factories.RNAFactory; import fr.orsay.lri.varna.models.rna.RNA; - -// import org.apache.log4j.*; +import jalview.analysis.Rna; +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.DBRefUtils; +import jalview.util.Format; +import jalview.util.MessageManager; /** * This class is supposed to parse a Stockholm format file into Jalview There @@ -74,13 +74,24 @@ 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( + "^[^<>[\\](){}ADFJ-RUVWYZadfj-ruvwyz]*$"); + StringBuffer out; // output buffer AlignmentI al; @@ -163,8 +174,8 @@ 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 +206,7 @@ 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 +218,8 @@ 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 { @@ -230,8 +240,8 @@ 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(); @@ -252,7 +262,7 @@ 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) @@ -323,17 +333,15 @@ public class StockholmFile extends AlignFile if (accAnnotations != null && accAnnotations.containsKey("AC")) { - if (dbsource != null) + String dbr = (String) accAnnotations.get("AC"); + if (dbr != null) { - String dbr = (String) accAnnotations.get("AC"); - if (dbr != null) - { - // we could get very clever here - but for now - just try to - // guess accession type from source of alignment plus structure - // of accession - guessDatabaseFor(seqO, dbr, dbsource); - - } + // we could get very clever here - but for now - just try to + // guess accession type from type of sequence, source of alignment + // plus + // structure + // of accession + guessDatabaseFor(seqO, dbr, dbsource); } // else - do what ? add the data anyway and prompt the user to // specify what references these are ? @@ -366,6 +374,11 @@ 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 +387,7 @@ 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 +400,11 @@ 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 +420,7 @@ 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 +447,8 @@ 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) @@ -507,6 +526,11 @@ public class StockholmFile extends AlignFile treeName = an.stringMatched(2); treeString = new StringBuffer(); } + // TODO: JAL-3532 - this is where GF comments and database + // references are lost + // suggest overriding this method for Stockholm files to catch and + // properly + // process CC, DR etc into multivalued properties setAlignmentProperty(an.stringMatched(1), an.stringMatched(2)); } } @@ -572,22 +596,11 @@ 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 +615,12 @@ 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")) @@ -631,15 +648,18 @@ public class StockholmFile extends AlignFile content = new Hashtable(); features.put(this.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 +670,22 @@ 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 +693,8 @@ 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 +722,7 @@ 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) @@ -738,6 +759,13 @@ public class StockholmFile extends AlignFile st = -1; } } + if (dbsource == null) + { + // make up an origin based on whether the sequence looks like it is + // nucleotide + // or protein + dbsource = (seqO.isProtein()) ? "PFAM" : "RFAM"; + } if (dbsource.equals("PFAM")) { seqdb = "UNIPROT"; @@ -786,8 +814,10 @@ 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); } @@ -807,14 +837,24 @@ public class StockholmFile extends AlignFile 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()]; @@ -828,26 +868,46 @@ 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.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; @@ -881,6 +941,12 @@ public class StockholmFile extends AlignFile return annot; } + private String dbref_to_ac_record(DBRefEntry ref) + { + return ref.getSource().toString() + " ; " + + ref.getAccessionId().toString(); + } + @Override public String print(SequenceI[] s, boolean jvSuffix) { @@ -892,32 +958,50 @@ 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; + boolean isAA = s[in].isProtein(); + 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<>(); + } + List primrefs = seq.getPrimaryDBRefs(); + if (primrefs.size() >= 1) + { + dataRef.put(tmp, dbref_to_ac_record(primrefs.get(0))); + } + else { - if (dataRef == null) + for (int idb = 0; idb < seq.getDBRefs().size(); idb++) { - dataRef = new Hashtable(); + DBRefEntry dbref = seq.getDBRefs().get(idb); + dataRef.put(tmp, dbref_to_ac_record(dbref)); + // if we put in a uniprot or EMBL record then we're done: + if (isAA && DBRefSource.UNIPROT + .equals(DBRefUtils.getCanonicalName(dbref.getSource()))) + { + break; + } + if (!isAA && DBRefSource.EMBL + .equals(DBRefUtils.getCanonicalName(dbref.getSource()))) + { + break; + } } - - String datAs1 = s[in].getDBRefs()[idb].getSource().toString() - + " ; " - + s[in].getDBRefs()[idb].getAccessionId().toString(); - dataRef.put(tmp, datAs1); } } in++; @@ -943,14 +1027,15 @@ 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(); - String type = (String) dataRef.remove(idd); - out.append(new Format("%-" + (maxid - 2) + "s").form("#=GS " - + idd.toString() + " ")); - if (type.contains("PFAM") || type.contains("RFAM")) + String type = dataRef.remove(idd); + out.append(new Format("%-" + (maxid - 2) + "s") + .form("#=GS " + idd.toString() + " ")); + if (isAA && type.contains("UNIPROT") + || (!isAA && type.contains("EMBL"))) { out.append(" AC " + type.substring(type.indexOf(";") + 1)); @@ -964,13 +1049,13 @@ 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); @@ -989,38 +1074,39 @@ public class StockholmFile extends AlignFile } // 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(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")) @@ -1029,7 +1115,7 @@ public class StockholmFile extends AlignFile } else { - key = type2id(aa.label.toLowerCase()); + key = type2id(aa.label.toLowerCase(Locale.ROOT)); if (key == null) { label = aa.label; @@ -1045,14 +1131,14 @@ 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++) + 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); } } @@ -1078,23 +1164,39 @@ 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 - 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) @@ -1109,7 +1211,9 @@ public class StockholmFile extends AlignFile { seq = ch.charAt(1); } - return seq; + + return (seq == ' ' && key != null && key.equals("SS") && isrna) ? '.' + : seq; } public String print() @@ -1131,10 +1235,10 @@ 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"); @@ -1145,7 +1249,7 @@ 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"); } } @@ -1156,8 +1260,8 @@ public class StockholmFile extends AlignFile { 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; } @@ -1168,7 +1272,7 @@ 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; @@ -1178,8 +1282,8 @@ 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; }