*/
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.Comparison;
-import jalview.util.Format;
-import jalview.util.MessageManager;
-
import java.io.BufferedReader;
import java.io.FileReader;
import java.io.IOException;
import java.util.Hashtable;
import java.util.LinkedHashMap;
import java.util.List;
+import java.util.Locale;
import java.util.Map;
import java.util.Vector;
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
{
private static final String ANNOTATION = "annotation";
- 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(
+ "^[^<>[\\](){}ADFJ-RUVWYZadfj-ruvwyz]*$");
+
StringBuffer out; // output buffer
AlignmentI al;
String version;
// String id;
Hashtable seqAnn = new Hashtable(); // Sequence related annotations
- LinkedHashMap<String, String> seqs = new LinkedHashMap<String, String>();
+ LinkedHashMap<String, String> seqs = new LinkedHashMap<>();
Regex p, r, rend, s, x;
// Temporary line for processing RNA annotation
// String RNAannot = "";
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();
// 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)
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 ?
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));
}
}
strucAnn = new Hashtable();
}
- Vector<AlignmentAnnotation> newStruc = new Vector<AlignmentAnnotation>();
+ Vector<AlignmentAnnotation> newStruc = new Vector<>();
parseAnnotationRow(newStruc, type, ns);
for (AlignmentAnnotation alan : newStruc)
{
private void guessDatabaseFor(Sequence seqO, String dbr, String dbsource)
{
DBRefEntry dbrf = null;
- List<DBRefEntry> dbrs = new ArrayList<DBRefEntry>();
+ List<DBRefEntry> dbrs = new ArrayList<>();
String seqdb = "Unknown", sdbac = "" + dbr;
int st = -1, en = -1, p;
if ((st = sdbac.indexOf("/")) > -1)
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";
}
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"))
{
{
// 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);
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)
{
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<String, String> dataRef = null;
+ boolean isAA = s[in].isProtein();
+ while ((in < slen) && ((seq = s[in]) != null))
{
- String tmp = printId(s[in], jvSuffix);
- max = Math.max(max, s[in].getLength());
+ String tmp = printId(seq, jvSuffix);
+ max = Math.max(max, seq.getLength());
if (tmp.length() > maxid)
{
maxid = tmp.length();
}
- if (s[in].getDBRefs() != null)
+ List<DBRefEntry> 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<DBRefEntry> primrefs = seq.getPrimaryDBRefs();
+ if (primrefs.size() >= 1)
{
- if (dataRef == null)
+ dataRef.put(tmp, dbref_to_ac_record(primrefs.get(0)));
+ }
+ else
+ {
+ 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++;
// output database accessions
if (dataRef != null)
{
- Enumeration en = dataRef.keys();
+ Enumeration<String> en = dataRef.keys();
while (en.hasMoreElements())
{
Object idd = en.nextElement();
- String type = (String) dataRef.remove(idd);
+ String type = dataRef.remove(idd);
out.append(new Format("%-" + (maxid - 2) + "s")
.form("#=GS " + idd.toString() + " "));
- if (type.contains("PFAM") || type.contains("RFAM"))
+ 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)
+ 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);
// 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"))
}
else
{
- key = type2id(aa.label.toLowerCase());
+ key = type2id(aa.label.toLowerCase(Locale.ROOT));
if (key == null)
{
label = aa.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);
}
}
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)
{
seq = ch.charAt(1);
}
- return seq;
+
+ return (seq == ' ' && key != null && key.equals("SS") && isrna) ? '.'
+ : seq;
}
public String print()