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;
*/
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 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)
{
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.",
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
{
// 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
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 ?
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;
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);
}
{
// 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)
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));
}
}
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("annotation");
+ String ns = (String) content.get(ANNOTATION);
if (ns == null)
{
}
// 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
// }
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);
}
else
{
throw new IOException(MessageManager.formatMessage(
- "exception.unknown_annotation_detected", new String[] {
- annType, annContent }));
+ "exception.unknown_annotation_detected", new String[]
+ { annType, annContent }));
}
}
}
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";
{
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);
}
Vector<AlignmentAnnotation> annotation, String label,
String annots)
{
- String convert1, convert2 = null;
-
- // convert1 = OPEN_PAREN.replaceAll(annots);
- // convert2 = CLOSE_PAREN.replaceAll(convert1);
+ 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, 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"))
{
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
{
// 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);
ann.secondaryStructure = ResidueProperties.getDssp3state(pos)
.charAt(0);
- if (ann.secondaryStructure == pos.charAt(0))
- {
- ann.displayCharacter = ""; // null; // " ";
- }
- else
- {
- ann.displayCharacter = " " + ann.displayCharacter;
- }
+ if (ann.secondaryStructure == pos.charAt(0))
+ {
+ ann.displayCharacter = ""; // null; // " ";
+ }
+ else
+ {
+ ann.displayCharacter = " " + ann.displayCharacter;
+ }
}
}
return annot;
}
- @Override
- public String print(SequenceI[] s, boolean jvSuffix)
+ private String dbref_to_ac_record(DBRefEntry ref)
{
- String string = print(s, jvSuffix, false);
- return string;
+ return ref.getSource().toString() + " ; "
+ + ref.getAccessionId().toString();
}
-
- public String print(SequenceI[] s, boolean jvSuffix,
- boolean removeAnnotations)
+ @Override
+ public String print(SequenceI[] s, boolean jvSuffix)
{
out = new StringBuffer();
out.append("# STOCKHOLM 1.0");
int maxid = 0;
int in = 0;
Hashtable dataRef = null;
+ boolean isAA = s[in].isProtein();
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)
{
}
if (s[in].getDBRefs() != null)
{
- for (int idb = 0; idb < s[in].getDBRefs().length; idb++)
+ if (dataRef == null)
+ {
+ dataRef = new Hashtable();
+ }
+ List<DBRefEntry> primrefs = s[in].getPrimaryDBRefs();
+ if (primrefs.size() >= 1)
{
- if (dataRef == null)
+ dataRef.put(tmp, dbref_to_ac_record(primrefs.get(0)));
+ }
+ else
+ {
+ for (int idb = 0; idb < s[in].getDBRefs().length; idb++)
{
- dataRef = new Hashtable();
+ DBRefEntry dbref = s[in].getDBRefs()[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++;
{
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(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));
}
}
-
- // output annotations
- while (i < s.length && s[i] != null)
+ // output annotations
+ while (i < s.length && s[i] != null)
+ {
+ AlignmentAnnotation[] alAnot = s[i].getAnnotation();
+ if (alAnot != null)
{
- AlignmentAnnotation[] alAnot = s[i].getAnnotation();
- if (alAnot != null && !removeAnnotations)
+ Annotation[] ann;
+
+ for (int j = 0; j < alAnot.length; j++)
{
- Annotation[] ann;
- for (int j = 0; j < alAnot.length; j++)
+ if (alAnot[j].annotations != null)
{
-
String key = type2id(alAnot[j].label);
boolean isrna = alAnot[j].isValidStruc();
}
}
- out.append(new Format("%-" + maxid + "s")
- .form(printId(s[i], jvSuffix) + " "));
- out.append(s[i].getSequenceAsString());
- out.append(newline);
- i++;
}
- if (!removeAnnotations)
+ out.append(new Format("%-" + maxid + "s")
+ .form(printId(s[i], jvSuffix) + " "));
+ out.append(s[i].getSequenceAsString());
+ out.append(newline);
+ i++;
+ }
+
+ // alignment annotation
+ AlignmentAnnotation aa;
+ if (al.getAlignmentAnnotation() != null)
{
- // alignment annotation
- AlignmentAnnotation aa;
- if (al.getAlignmentAnnotation() != null)
+ for (int ia = 0; ia < al.getAlignmentAnnotation().length; ia++)
{
- for (int ia = 0; ia < al.getAlignmentAnnotation().length; ia++)
+ aa = al.getAlignmentAnnotation()[ia];
+ if (aa.autoCalculated || !aa.visible || aa.sequenceRef != null)
{
- aa = al.getAlignmentAnnotation()[ia];
- if (aa.autoCalculated || !aa.visible || aa.sequenceRef != null)
- {
- continue;
- }
- String seq = "";
- String label;
- String key = "";
- if (aa.label.equals("seq"))
+ continue;
+ }
+ String seq = "";
+ String label;
+ String key = "";
+ if (aa.label.equals("seq"))
+ {
+ label = "seq_cons";
+ }
+ else
+ {
+ key = type2id(aa.label.toLowerCase());
+ if (key == null)
{
- label = "seq_cons";
+ label = aa.label;
}
else
{
- key = type2id(aa.label.toLowerCase());
- if (key == null)
- {
- label = aa.label;
- }
- else
- {
- label = key + "_cons";
- }
- }
- if (label == null)
- {
- label = aa.label;
+ label = key + "_cons";
}
- label = label.replace(" ", "_");
+ }
+ if (label == null)
+ {
+ label = aa.label;
+ }
+ label = label.replace(" ", "_");
- out.append(new Format("%-" + maxid + "s")
- .form("#=GC " + label + " "));
- boolean isrna = aa.isValidStruc();
- for (int j = 0; j < aa.annotations.length; j++)
- {
- seq += outputCharacter(key, j, isrna, aa.annotations, null);
- }
- out.append(seq);
- out.append(newline);
+ out.append(
+ new Format("%-" + maxid + "s").form("#=GC " + label + " "));
+ boolean isrna = aa.isValidStruc();
+ for (int j = 0; j < aa.annotations.length; j++)
+ {
+ seq += outputCharacter(key, j, isrna, aa.annotations, null);
}
+ out.append(seq);
+ out.append(newline);
}
}
return out.toString();
}
+
/**
* add an annotation character to the output row
*
{
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)
{
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();
out.append(newline);
print(getSeqsAsArray(), false);
+ out.append("//");
out.append(newline);
return out.toString();
}
}
}
-
+
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;
}
{
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;
- }
}
git://source.jalview.org / jalview.git / blobdiff