X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Fio%2FStockholmFile.java;h=9b5a5cdb56420af978d70feebb27cd53faea721f;hb=b2f9a8d7bce642ff4011bc6d49e02bb0569fbb11;hp=5ff3551315971255d28c24a13c1b062be01924bb;hpb=59d682209891099d46b960509907c79e3fb276fe;p=jalview.git
diff --git a/src/jalview/io/StockholmFile.java b/src/jalview/io/StockholmFile.java
index 5ff3551..9b5a5cd 100644
--- a/src/jalview/io/StockholmFile.java
+++ b/src/jalview/io/StockholmFile.java
@@ -1,6 +1,6 @@
/*
- * Jalview - A Sequence Alignment Editor and Viewer (Version 2.8)
- * Copyright (C) 2012 J Procter, AM Waterhouse, LM Lui, J Engelhardt, G Barton, M Clamp, S Searle
+ * Jalview - A Sequence Alignment Editor and Viewer (Version 2.8.1)
+ * Copyright (C) 2014 The Jalview Authors
*
* This file is part of Jalview.
*
@@ -14,6 +14,7 @@
* PURPOSE. See the GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License along with Jalview. If not, see .
+ * The Jalview Authors are detailed in the 'AUTHORS' file.
*/
/*
* This extension was written by Benjamin Schuster-Boeckler at sanger.ac.uk
@@ -25,7 +26,7 @@ import java.util.*;
import com.stevesoft.pat.*;
import jalview.datamodel.*;
-import jalview.analysis.Rna;
+import jalview.util.Format;
// import org.apache.log4j.*;
@@ -44,11 +45,22 @@ import jalview.analysis.Rna;
public class StockholmFile extends AlignFile
{
// static Logger logger = Logger.getLogger("jalview.io.StockholmFile");
+ StringBuffer out; // output buffer
+
+ AlignmentI al;
public StockholmFile()
{
}
+ /**
+ * Creates a new StockholmFile object for output.
+ */
+ public StockholmFile(AlignmentI al)
+ {
+ this.al = al;
+ }
+
public StockholmFile(String inFile, String type) throws IOException
{
super(inFile, type);
@@ -133,8 +145,24 @@ public class StockholmFile extends AlignFile
if (rend.search(line))
{
// End of the alignment, pass stuff back
-
this.noSeqs = seqs.size();
+
+ String seqdb,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)
+ {
+ String dbType = getAlignmentProperty("AC").toString();
+ if (pf.search(dbType))
+ {
+ // PFAM Alignment - so references are typically from Uniprot
+ dbsource = "PFAM";
+ }
+ else if (rf.search(dbType))
+ {
+ dbsource = "RFAM";
+ }
+ }
// logger.debug("Number of sequences: " + this.noSeqs);
Enumeration accs = seqs.keys();
while (accs.hasMoreElements())
@@ -177,6 +205,7 @@ public class StockholmFile extends AlignFile
String desc = (String) accAnnotations.get("DE");
seqO.setDescription((desc == null) ? "" : desc);
}
+
// Add DB References (if any)
if (accAnnotations != null && accAnnotations.containsKey("DR"))
{
@@ -186,19 +215,22 @@ public class StockholmFile extends AlignFile
String src = dbr.substring(0, dbr.indexOf(";"));
String acn = dbr.substring(dbr.indexOf(";") + 1);
jalview.util.DBRefUtils.parseToDbRef(seqO, src, "0", acn);
- // seqO.addDBRef(dbref);
}
}
- if (accAnnotations != null && accAnnotations.containsKey("SS"))
- {
- Vector v = (Vector) accAnnotations.get("SS");
- for (int i = 0; i < v.size(); i++)
+ if (accAnnotations != null && accAnnotations.containsKey("AC"))
+ {
+ if (dbsource != null)
{
- AlignmentAnnotation an = (AlignmentAnnotation) v.elementAt(i);
- seqO.addAlignmentAnnotation(an);
- // annotations.add(an);
- }
+ 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);
+
+ }
+ }
+ // else - do what ? add the data anyway and prompt the user to specify what references these are ?
}
Hashtable features = null;
@@ -225,6 +257,24 @@ public class StockholmFile extends AlignFile
// TODO: map coding region to core jalview feature types
String type = i.nextElement().toString();
Hashtable content = (Hashtable) features.remove(type);
+
+ // add alignment annotation for this feature
+ String key = type2id(type);
+ if (key != null)
+ {
+ if (accAnnotations != null
+ && accAnnotations.containsKey(key))
+ {
+ Vector vv = (Vector) accAnnotations.get(key);
+ for (int ii = 0; ii < vv.size(); ii++)
+ {
+ AlignmentAnnotation an = (AlignmentAnnotation) vv
+ .elementAt(ii);
+ seqO.addAlignmentAnnotation(an);
+ }
+ }
+ }
+
Enumeration j = content.keys();
while (j.hasMoreElements())
{
@@ -478,25 +528,20 @@ public class StockholmFile extends AlignFile
}
ns += seq;
content.put(description, ns);
-
- if (type.equals("SS"))
+ Hashtable strucAnn;
+ if (seqAnn.containsKey(acc))
{
- Hashtable strucAnn;
- if (seqAnn.containsKey(acc))
- {
- strucAnn = (Hashtable) seqAnn.get(acc);
- }
- else
- {
- strucAnn = new Hashtable();
- }
-
- Vector newStruc = new Vector();
- parseAnnotationRow(newStruc, type, ns);
-
- strucAnn.put(type, newStruc);
- seqAnn.put(acc, strucAnn);
+ strucAnn = (Hashtable) seqAnn.get(acc);
}
+ else
+ {
+ strucAnn = new Hashtable();
+ }
+
+ Vector newStruc = new Vector();
+ parseAnnotationRow(newStruc, type, ns);
+ strucAnn.put(type, newStruc);
+ seqAnn.put(acc, strucAnn);
}
else
{
@@ -523,6 +568,95 @@ public class StockholmFile extends AlignFile
}
}
+ /**
+ * Demangle an accession string and guess the originating sequence database for a given sequence
+ * @param seqO sequence to be annotated
+ * @param dbr Accession string for sequence
+ * @param dbsource source database for alignment (PFAM or RFAM)
+ */
+ private void guessDatabaseFor(Sequence seqO, String dbr, String dbsource)
+ {
+ DBRefEntry dbrf=null;
+ List dbrs=new ArrayList();
+ String seqdb="Unknown",sdbac=""+dbr;
+ int st=-1,en=-1,p;
+ if ((st=sdbac.indexOf("/"))>-1)
+ {
+ String num,range=sdbac.substring(st+1);
+ sdbac = sdbac.substring(0,st);
+ if ((p=range.indexOf("-"))>-1)
+ {
+ p++;
+ if (p-1)
+ {
+ // strip of last subdomain
+ sdbac = sdbac.substring(0,sdbac.indexOf("."));
+ dbrf = jalview.util.DBRefUtils.parseToDbRef(seqO, seqdb, dbsource, sdbac);
+ if (dbrf!=null)
+ {
+ dbrs.add(dbrf);
+ }
+ }
+ dbrf = jalview.util.DBRefUtils.parseToDbRef(seqO, dbsource, dbsource, dbr);
+ if (dbr!=null)
+ {
+ dbrs.add(dbrf);
+ }
+ } else {
+ seqdb = "EMBL"; // total guess - could be ENA, or something else these days
+ if (sdbac.indexOf(".")>-1)
+ {
+ // strip off last subdomain
+ sdbac = sdbac.substring(0,sdbac.indexOf("."));
+ dbrf = jalview.util.DBRefUtils.parseToDbRef(seqO, seqdb, dbsource, sdbac);
+ if (dbrf!=null)
+ {
+ dbrs.add(dbrf);
+ }
+ }
+
+ dbrf = jalview.util.DBRefUtils.parseToDbRef(seqO, dbsource, dbsource, dbr);
+ if (dbrf!=null)
+ {
+ dbrs.add(dbrf);
+ }
+ }
+ if (st!=-1 && en!=-1)
+ {
+ 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.datamodel.Mapping mping = new Mapping(mp);
+ d.setMap(mping);
+ }
+ }
+ }
+
protected static AlignmentAnnotation parseAnnotationRow(
Vector annotation, String label, String annots)
{
@@ -539,8 +673,12 @@ public class StockholmFile extends AlignFile
convert2 = closeparen.replaceAll(convert1);
annots = convert2;
- String type = (label.indexOf("_cons") == label.length() - 5) ? label
- .substring(0, label.length() - 5) : label;
+ String type = label;
+ if (label.contains("_cons"))
+ {
+ type = (label.indexOf("_cons") == label.length() - 5) ? label
+ .substring(0, label.length() - 5) : label;
+ }
boolean ss = false;
type = id2type(type);
if (type.equals("secondary structure"))
@@ -607,14 +745,211 @@ public class StockholmFile extends AlignFile
return annot;
}
- public static String print(SequenceI[] s)
+ public String print(SequenceI[] s)
{
- return "not yet implemented";
+ // find max length of id
+ int max = 0;
+ int maxid = 0;
+ int in = 0;
+ Hashtable dataRef = null;
+ while ((in < s.length) && (s[in] != null))
+ {
+ String tmp = printId(s[in]);
+ if (s[in].getSequence().length > max)
+ {
+ max = s[in].getSequence().length;
+ }
+
+ if (tmp.length() > maxid)
+ {
+ maxid = tmp.length();
+ }
+ if (s[in].getDBRef() != null)
+ {
+ for (int idb = 0; idb < s[in].getDBRef().length; idb++)
+ {
+ if (dataRef == null)
+ dataRef = new Hashtable();
+
+ String datAs1 = s[in].getDBRef()[idb].getSource().toString()
+ + " ; "
+ + s[in].getDBRef()[idb].getAccessionId().toString();
+ dataRef.put(tmp, datAs1);
+ }
+ }
+ in++;
+ }
+ maxid += 9;
+ int i = 0;
+
+ // output database type
+ if (al.getProperties() != null)
+ {
+ if (!al.getProperties().isEmpty())
+ {
+ Enumeration key = al.getProperties().keys();
+ Enumeration val = al.getProperties().elements();
+ while (key.hasMoreElements())
+ {
+ out.append("#=GF " + key.nextElement() + " " + val.nextElement());
+ out.append(newline);
+ }
+ }
+ }
+
+ // output database accessions
+ if (dataRef != null)
+ {
+ 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"))
+ {
+
+ out.append(" AC " + type.substring(type.indexOf(";") + 1));
+ }
+ else
+ {
+ out.append(" DR " + type + " ");
+ }
+ out.append(newline);
+ }
+ }
+
+ // output annotations
+ while (i < s.length && s[i] != null)
+ {
+ if (s[i].getDatasetSequence() != null)
+ {
+ SequenceI ds = s[i].getDatasetSequence();
+ AlignmentAnnotation[] alAnot;
+ Annotation[] ann;
+ Annotation annot;
+ alAnot = s[i].getAnnotation();
+ String feature = "";
+ if (alAnot != null)
+ {
+ for (int j = 0; j < alAnot.length; j++)
+ {
+ if (ds.getSequenceFeatures() != null)
+ {
+ feature = ds.getSequenceFeatures()[0].type;
+ }
+ String key = type2id(feature);
+
+ if (key == null)
+ continue;
+
+ // out.append("#=GR ");
+ out.append(new Format("%-" + maxid + "s").form("#=GR "
+ + printId(s[i]) + " " + key + " "));
+ ann = alAnot[j].annotations;
+ String seq = "";
+ for (int k = 0; k < ann.length; k++)
+ {
+ annot = ann[k];
+ String ch = (annot == null) ? Character.toString(s[i]
+ .getCharAt(k)) : annot.displayCharacter;
+ if (ch.length() == 0)
+ {
+ if (key.equals("SS"))
+ {
+ char ll = annot.secondaryStructure;
+ seq = (Character.toString(ll).equals(" ")) ? seq + "C"
+ : seq + ll;
+ }
+ else
+ {
+ seq += ".";
+ }
+ }
+ else if (ch.length() == 1)
+ {
+ seq += ch;
+ }
+ else if (ch.length() > 1)
+ {
+ seq += ch.charAt(1);
+ }
+ }
+ out.append(seq);
+ out.append(newline);
+ }
+ }
+ }
+
+ out.append(new Format("%-" + maxid + "s").form(printId(s[i]) + " "));
+ out.append(s[i].getSequenceAsString());
+ out.append(newline);
+ i++;
+ }
+
+ // alignment annotation
+ AlignmentAnnotation aa;
+ if (al.getAlignmentAnnotation() != null)
+ {
+ for (int ia = 0; ia < al.getAlignmentAnnotation().length; ia++)
+ {
+ aa = al.getAlignmentAnnotation()[ia];
+ if (aa.autoCalculated || !aa.visible)
+ {
+ continue;
+ }
+ String seq = "";
+ String label;
+
+ if (aa.label.equals("seq"))
+ label = "seq_cons";
+ else
+ label = type2id(aa.label.toLowerCase()) + "_cons";
+
+ if (label == null)
+ label = aa.label;
+
+ out.append(new Format("%-" + maxid + "s").form("#=GC " + label
+ + " "));
+ for (int j = 0; j < aa.annotations.length; j++)
+ {
+ String ch = (aa.annotations[j] == null) ? "-"
+ : aa.annotations[j].displayCharacter;
+ if (ch.length() == 0)
+ {
+ char ll = aa.annotations[j].secondaryStructure;
+ if (Character.toString(ll).equals(" "))
+ seq += "C";
+ else
+ seq += ll;
+ }
+ else if (ch.length() == 1)
+ {
+ seq += ch;
+ }
+ else if (ch.length() > 1)
+ {
+ seq += ch.charAt(1);
+ }
+ }
+ out.append(seq);
+ out.append(newline);
+ }
+ }
+ return out.toString();
}
public String print()
{
- return print(getSeqsAsArray());
+ out = new StringBuffer();
+ out.append("# STOCKHOLM 1.0");
+ out.append(newline);
+ print(getSeqsAsArray());
+
+ out.append("//");
+ out.append(newline);
+ return out.toString();
}
private static Hashtable typeIds = null;
@@ -652,6 +987,28 @@ public class StockholmFile extends AlignFile
+ id);
return id;
}
+
+ protected static String type2id(String type)
+ {
+ String key = null;
+ Enumeration e = typeIds.keys();
+ while (e.hasMoreElements())
+ {
+ Object ll = e.nextElement();
+ if (typeIds.get(ll).toString().equals(type))
+ {
+ key = (String) ll;
+ break;
+ }
+ }
+ if (key != null)
+ {
+ return (String) key;
+ }
+ System.err.println("Warning : Unknown Stockholm annotation type: "
+ + type);
+ return key;
+ }
/**
* //ssline is complete secondary structure line private AlignmentAnnotation
* addHelices(Vector annotation, String label, String ssline) {