X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Fio%2FStockholmFile.java;h=50112c82d95cedaac2415de8426b2828ccf8eff5;hb=7d602d0e4b439e56af3e4551ed71f181a8025534;hp=62fec79db681ebcd0daa92a04657fcd61c4d04b7;hpb=91f6f98109a5c9ffda2ed8da3875773e3469f2f2;p=jalview.git
diff --git a/src/jalview/io/StockholmFile.java b/src/jalview/io/StockholmFile.java
index 62fec79..50112c8 100644
--- a/src/jalview/io/StockholmFile.java
+++ b/src/jalview/io/StockholmFile.java
@@ -1,31 +1,61 @@
/*
- * 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 ($$Version-Rel$$)
+ * Copyright (C) $$Year-Rel$$ The Jalview Authors
*
* This file is part of Jalview.
*
* Jalview is free software: you can redistribute it and/or
* modify it under the terms of the GNU General Public License
- * as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
+ * as published by the Free Software Foundation, either version 3
+ * of the License, or (at your option) any later version.
*
* Jalview is distributed in the hope that it will be useful, but
* WITHOUT ANY WARRANTY; without even the implied warranty
* of MERCHANTABILITY or FITNESS FOR A PARTICULAR
* 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 .
+ * 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
*/
package jalview.io;
-import java.io.*;
-import java.util.*;
-
-import com.stevesoft.pat.*;
-import jalview.datamodel.*;
+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;
+import jalview.util.Platform;
+
+import java.io.BufferedReader;
+import java.io.FileReader;
+import java.io.IOException;
+import java.util.ArrayList;
+import java.util.Enumeration;
+import java.util.Hashtable;
+import java.util.LinkedHashMap;
+import java.util.List;
+import java.util.Map;
+import java.util.Vector;
+
+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.*;
@@ -38,28 +68,136 @@ import jalview.util.Format;
* into Jalview's local representation.
*
* @author bsb at sanger.ac.uk
+ * @author Natasha Shersnev (Dundee, UK) (Stockholm file writer)
+ * @author Lauren Lui (UCSC, USA) (RNA secondary structure annotation import as
+ * stockholm)
+ * @author Anne Menard (Paris, FR) (VARNA parsing of Stockholm file data)
* @version 0.3 + jalview mods
*
*/
public class StockholmFile extends AlignFile
{
- // static Logger logger = Logger.getLogger("jalview.io.StockholmFile");
- StringBuffer out;
- AlignmentI al;
+ private static final String ANNOTATION = "annotation";
+
+ private static final char UNDERSCORE = '_';
+ // WUSS extended symbols. Avoid ambiguity with protein SS annotations by using NOT_RNASS first.
+
+ public static final String RNASS_BRACKETS = "<>[](){}AaBbCcDdEeFfGgHhIiJjKkLlMmNnOoPpQqRrSsTtUuVvWwXxYyZz";
+
+ public static final int REGEX_STOCKHOLM = 0;
+
+ public static final int REGEX_BRACKETS = 1;
+
+ // 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)
+ public static final int REGEX_NOT_RNASS = 2;
+
+ private static final int REGEX_ANNOTATION = 3;
+
+ private static final int REGEX_PFAM = 4;
+
+ private static final int REGEX_RFAM = 5;
+
+ private static final int REGEX_ALIGN_END = 6;
+
+ private static final int REGEX_SPLIT_ID = 7;
+
+ private static final int REGEX_SUBTYPE = 8;
+
+ private static final int REGEX_ANNOTATION_LINE = 9;
+
+ private static final int REGEX_REMOVE_ID = 10;
+
+ private static final int REGEX_OPEN_PAREN = 11;
+
+ private static final int REGEX_CLOSE_PAREN = 12;
+
+ public static final int REGEX_MAX = 13;
+
+ private static Regex REGEX[] = new Regex[REGEX_MAX];
+
+ /**
+ * Centralize all actual Regex instantialization in Platform.
+ * // JBPNote: Why is this 'centralisation' better ?
+ * @param id
+ * @return
+ */
+ private static Regex getRegex(int id)
+ {
+ if (REGEX[id] == null)
+ {
+ String pat = null, pat2 = null;
+ switch (id)
+ {
+ case REGEX_STOCKHOLM:
+ pat = "# STOCKHOLM ([\\d\\.]+)";
+ break;
+ case REGEX_BRACKETS:
+ // for reference; not used
+ pat = "(<|>|\\[|\\]|\\(|\\)|\\{|\\})";
+ break;
+ case REGEX_NOT_RNASS:
+ pat = "^[^<>[\\](){}A-DF-Za-df-z]*$";
+ break;
+ case REGEX_ANNOTATION:
+ pat = "(\\w+)\\s*(.*)";
+ break;
+ case REGEX_PFAM:
+ pat = "PF[0-9]{5}(.*)";
+ break;
+ case REGEX_RFAM:
+ pat = "RF[0-9]{5}(.*)";
+ break;
+ case REGEX_ALIGN_END:
+ pat = "^\\s*\\/\\/";
+ break;
+ case REGEX_SPLIT_ID:
+ pat = "(\\S+)\\/(\\d+)\\-(\\d+)";
+ break;
+ case REGEX_SUBTYPE:
+ pat = "(\\S+)\\s+(\\S*)\\s+(.*)";
+ break;
+ case REGEX_ANNOTATION_LINE:
+ pat = "#=(G[FSRC]?)\\s+(.*)";
+ break;
+ case REGEX_REMOVE_ID:
+ pat = "(\\S+)\\s+(\\S+)";
+ break;
+ case REGEX_OPEN_PAREN:
+ pat = "(<|\\[)";
+ pat2 = "(";
+ break;
+ case REGEX_CLOSE_PAREN:
+ pat = "(>|\\])";
+ pat2 = ")";
+ break;
+ default:
+ return null;
+ }
+ REGEX[id] = Platform.newRegex(pat, pat2);
+ }
+ return REGEX[id];
+ }
+
+ StringBuffer out; // output buffer
+
+ private AlignmentI al;
+
public StockholmFile()
{
}
/**
- * Creates a new StockholmFile object for output.
- */
+ * Creates a new StockholmFile object for output
+ */
public StockholmFile(AlignmentI al)
{
this.al = al;
}
-
- public StockholmFile(String inFile, String type) throws IOException
+
+ public StockholmFile(String inFile, DataSourceType type)
+ throws IOException
{
super(inFile, type);
}
@@ -69,18 +207,84 @@ public class StockholmFile extends AlignFile
super(source);
}
+ @Override
public void initData()
{
super.initData();
}
/**
+ * Parse a file in Stockholm format into Jalview's data model using VARNA
+ *
+ * @throws IOException
+ * If there is an error with the input file
+ */
+ public void parse_with_VARNA(java.io.File inFile) throws IOException
+ {
+ FileReader fr = null;
+ fr = new FileReader(inFile);
+
+ BufferedReader r = new BufferedReader(fr);
+ List result = null;
+ try
+ {
+ result = RNAFactory.loadSecStrStockholm(r);
+ } catch (ExceptionUnmatchedClosingParentheses umcp)
+ {
+ errormessage = "Unmatched parentheses in annotation. Aborting ("
+ + umcp.getMessage() + ")";
+ throw new IOException(umcp);
+ }
+ // DEBUG System.out.println("this is the secondary scructure:"
+ // +result.size());
+ SequenceI[] seqs = new SequenceI[result.size()];
+ String id = null;
+ for (int i = 0; i < result.size(); i++)
+ {
+ // DEBUG System.err.println("Processing i'th sequence in Stockholm file")
+ RNA current = result.get(i);
+
+ String seq = current.getSeq();
+ String rna = current.getStructDBN(true);
+ // DEBUG System.out.println(seq);
+ // DEBUG System.err.println(rna);
+ int begin = 0;
+ int end = seq.length() - 1;
+ id = safeName(getDataName());
+ seqs[i] = new Sequence(id, seq, begin, end);
+ String[] annot = new String[rna.length()];
+ Annotation[] ann = new Annotation[rna.length()];
+ for (int j = 0; j < rna.length(); j++)
+ {
+ annot[j] = rna.substring(j, j + 1);
+
+ }
+
+ for (int k = 0; k < rna.length(); k++)
+ {
+ ann[k] = new Annotation(annot[k], "",
+ Rna.getRNASecStrucState(annot[k]).charAt(0), 0f);
+
+ }
+ AlignmentAnnotation align = new AlignmentAnnotation("Sec. str.",
+ current.getID(), ann);
+
+ seqs[i].addAlignmentAnnotation(align);
+ seqs[i].setRNA(result.get(i));
+ this.annotations.addElement(align);
+ }
+ this.setSeqs(seqs);
+
+ }
+
+ /**
* Parse a file in Stockholm format into Jalview's data model. The file has to
* be passed at construction time
*
* @throws IOException
* If there is an error with the input file
*/
+ @Override
public void parse() throws IOException
{
StringBuffer treeString = new StringBuffer();
@@ -90,41 +294,45 @@ public class StockholmFile extends AlignFile
String version;
// String id;
Hashtable seqAnn = new Hashtable(); // Sequence related annotations
- Hashtable seqs = new Hashtable();
+ LinkedHashMap seqs = new LinkedHashMap<>();
Regex p, r, rend, s, x;
-
// Temporary line for processing RNA annotation
// String RNAannot = "";
// ------------------ Parsing File ----------------------
// First, we have to check that this file has STOCKHOLM format, i.e. the
// first line must match
- r = new Regex("# STOCKHOLM ([\\d\\.]+)");
+
+ r = getRegex(REGEX_STOCKHOLM);
if (!r.search(nextLine()))
{
- throw new IOException(
- "This file is not in valid STOCKHOLM format: First line does not contain '# STOCKHOLM'");
+ throw new IOException(MessageManager
+ .getString("exception.stockholm_invalid_format"));
}
else
{
version = r.stringMatched(1);
+
// logger.debug("Stockholm version: " + version);
}
// We define some Regexes here that will be used regularily later
- rend = new Regex("^\\s*\\/\\/"); // Find the end of an alignment
- p = new Regex("(\\S+)\\/(\\d+)\\-(\\d+)"); // split sequence id in
+ rend = getRegex(REGEX_ALIGN_END);//"^\\s*\\/\\/"); // Find the end of an alignment
+ p = getRegex(REGEX_SPLIT_ID);//"(\\S+)\\/(\\d+)\\-(\\d+)"); // split sequence id in
// id/from/to
- s = new Regex("(\\S+)\\s+(\\S*)\\s+(.*)"); // Parses annotation subtype
- r = new Regex("#=(G[FSRC]?)\\s+(.*)"); // Finds any annotation line
- x = new Regex("(\\S+)\\s+(\\S+)"); // split id from sequence
+ s = getRegex(REGEX_SUBTYPE);// "(\\S+)\\s+(\\S*)\\s+(.*)"); // Parses
+ // annotation subtype
+ r = getRegex(REGEX_ANNOTATION_LINE);// "#=(G[FSRC]?)\\s+(.*)"); // Finds any
+ // annotation line
+ x = getRegex(REGEX_REMOVE_ID);// "(\\S+)\\s+(\\S+)"); // split id from
+ // sequence
// Convert all bracket types to parentheses (necessary for passing to VARNA)
- Regex openparen = new Regex("(<|\\[)", "(");
- Regex closeparen = new Regex("(>|\\])", ")");
+ Regex openparen = getRegex(REGEX_OPEN_PAREN);//"(<|\\[)", "(");
+ Regex closeparen = getRegex(REGEX_CLOSE_PAREN);//"(>|\\])", ")");
- // 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 = getRegex("(<|>|\\[|\\]|\\(|\\))");
rend.optimize();
p.optimize();
@@ -143,15 +351,30 @@ public class StockholmFile extends AlignFile
if (rend.search(line))
{
// End of the alignment, pass stuff back
-
this.noSeqs = seqs.size();
+
+ String dbsource = null;
+ Regex pf = getRegex(REGEX_PFAM); // Finds AC for Pfam
+ Regex rf = getRegex(REGEX_RFAM); // 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())
+ for (Map.Entry skey : seqs.entrySet())
{
- String acc = (String) accs.nextElement();
// logger.debug("Processing sequence " + acc);
- String seq = (String) seqs.remove(acc);
+ String acc = skey.getKey();
+ String seq = skey.getValue();
if (maxLength < seq.length())
{
maxLength = seq.length();
@@ -196,9 +419,23 @@ 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("AC"))
+ {
+ 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 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 ?
+ }
Hashtable features = null;
// We need to adjust the positions of all features to account for gaps
@@ -224,26 +461,40 @@ 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)
+
+ /*
+ * have we added annotation rows for this type ?
+ */
+ boolean annotsAdded = false;
+ if (key != null)
{
- if (accAnnotations != null && accAnnotations.containsKey(key))
+ if (accAnnotations != null
+ && accAnnotations.containsKey(key))
{
- Vector vv = (Vector) accAnnotations.get(key);
+ Vector vv = (Vector) accAnnotations.get(key);
for (int ii = 0; ii < vv.size(); ii++)
{
- AlignmentAnnotation an = (AlignmentAnnotation) vv.elementAt(ii);
- seqO.addAlignmentAnnotation(an);
- }
+ annotsAdded = true;
+ AlignmentAnnotation an = (AlignmentAnnotation) vv
+ .elementAt(ii);
+ seqO.addAlignmentAnnotation(an);
+ annotations.add(an);
+ }
}
}
-
+
Enumeration j = content.keys();
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++)
@@ -259,7 +510,7 @@ 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);
}
@@ -285,9 +536,11 @@ public class StockholmFile extends AlignFile
if (!x.search(line))
{
// logger.error("Could not parse sequence line: " + line);
- throw new IOException("Could not parse sequence line: " + line);
+ throw new IOException(MessageManager.formatMessage(
+ "exception.couldnt_parse_sequence_line", new String[]
+ { line }));
}
- String ns = (String) seqs.get(x.stringMatched(1));
+ String ns = seqs.get(x.stringMatched(1));
if (ns == null)
{
ns = "";
@@ -343,7 +596,7 @@ public class StockholmFile extends AlignFile
*/
// Let's save the annotations, maybe we'll be able to do something
// with them later...
- Regex an = new Regex("(\\w+)\\s*(.*)");
+ Regex an = getRegex(REGEX_ANNOTATION);
if (an.search(annContent))
{
if (an.stringMatched(1).equals("NH"))
@@ -363,6 +616,9 @@ 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));
}
}
@@ -397,7 +653,8 @@ public class StockholmFile extends AlignFile
}
else
{
- throw new IOException("Error parsing " + line);
+ // throw new IOException("Error parsing " + line);
+ System.err.println(">> missing annotation: " + line);
}
}
else if (annType.equals("GC"))
@@ -427,22 +684,11 @@ 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
@@ -457,8 +703,12 @@ 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"))
@@ -477,22 +727,28 @@ public class StockholmFile extends AlignFile
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(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
+
Hashtable strucAnn;
if (seqAnn.containsKey(acc))
{
@@ -503,23 +759,31 @@ 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;
+ }
+ // 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);
}
}
else
{
- throw new IOException("Unknown annotation detected: " + annType
- + " " + annContent);
+ throw new IOException(MessageManager.formatMessage(
+ "exception.unknown_annotation_detected", new String[]
+ { annType, annContent }));
}
}
}
@@ -533,70 +797,223 @@ public class StockholmFile extends AlignFile
}
}
- protected static AlignmentAnnotation parseAnnotationRow(
- Vector annotation, String label, String annots)
+ /**
+ * 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)
{
- String convert1, convert2 = null;
-
- // Convert all bracket types to parentheses
- Regex openparen = new Regex("(<|\\[)", "(");
- Regex closeparen = new Regex("(>|\\])", ")");
+ 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 < range.length())
+ {
+ num = range.substring(p).trim();
+ try
+ {
+ en = Integer.parseInt(num);
+ } catch (NumberFormatException x)
+ {
+ // could warn here that index is invalid
+ en = -1;
+ }
+ }
+ }
+ else
+ {
+ p = range.length();
+ }
+ num = range.substring(0, p).trim();
+ try
+ {
+ st = Integer.parseInt(num);
+ } catch (NumberFormatException x)
+ {
+ // could warn here that index is invalid
+ 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";
+ if (sdbac.indexOf(".") > -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);
+ }
+ }
- // Detect if file is RNA by looking for bracket types
- Regex detectbrackets = new Regex("(<|>|\\[|\\]|\\(|\\))");
+ 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);
+ }
+ }
+ }
- convert1 = openparen.replaceAll(annots);
- convert2 = closeparen.replaceAll(convert1);
- annots = convert2;
+ protected static AlignmentAnnotation parseAnnotationRow(
+ Vector annotation, String label,
+ String annots)
+ {
+ String convert1, convert2 = null;
+ // String convert1 = OPEN_PAREN.replaceAll(annots);
+ // String convert2 = CLOSE_PAREN.replaceAll(convert1);
+ // annots = convert2;
- String type = (label.indexOf("_cons") == label.length() - 5) ? label
- .substring(0, label.length() - 5) : label;
- boolean ss = false;
+ String type = label;
+ if (label.contains("_cons"))
+ {
+ type = (label.indexOf("_cons") == label.length() - 5)
+ ? label.substring(0, label.length() - 5)
+ : label;
+ }
+ boolean ss = false, posterior = false;
type = id2type(type);
- if (type.equals("secondary structure"))
+
+ boolean isrnass = false;
+
+ if (type.equalsIgnoreCase("secondary structure"))
{
ss = true;
+ isrnass = !getRegex(REGEX_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()];
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 (ss)
{
- if (detectbrackets.search(pos))
- {
- ann.secondaryStructure = jalview.schemes.ResidueProperties
- .getRNASecStrucState(pos).charAt(0);
- }
- else
+ // if (" .-_".indexOf(pos) == -1)
{
- ann.secondaryStructure = jalview.schemes.ResidueProperties
- .getDssp3state(pos).charAt(0);
+ 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; // " ";
+ }
+ else
+ {
+ ann.displayCharacter = " " + ann.displayCharacter;
+ }
+ }
}
- if (ann.secondaryStructure == pos.charAt(0) || pos.charAt(0) == 'C')
+ }
+ if (posterior && !ann.isWhitespace()
+ && !Comparison.isGap(pos.charAt(0)))
+ {
+ float val = 0;
+ // symbol encodes values - 0..*==0..10
+ if (pos.charAt(0) == '*')
{
- ann.displayCharacter = ""; // null; // " ";
+ val = 10;
}
else
{
- ann.displayCharacter = " " + ann.displayCharacter;
+ val = pos.charAt(0) - '0';
+ if (val > 9)
+ {
+ val = 10;
+ }
}
+ ann.value = val;
}
els[i] = ann;
}
AlignmentAnnotation annot = null;
- Enumeration e = annotation.elements();
+ Enumeration e = annotation.elements();
while (e.hasMoreElements())
{
- annot = (AlignmentAnnotation) e.nextElement();
+ annot = e.nextElement();
if (annot.label.equals(type))
+ {
break;
+ }
annot = null;
}
if (annot == null)
@@ -617,153 +1034,322 @@ public class StockholmFile extends AlignFile
return annot;
}
- public String print(SequenceI[] s)
+ private String dbref_to_ac_record(DBRefEntry ref)
{
- // find max length of id
- int max = 0;
- int maxid = 0;
- int in = 0;
- 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();
- }
-
- in++;
- }
- maxid += 9;
- int i = 0;
-
- 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);
-
- // output annotations
- 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 = type2id(aa.label);
-
- if (label == null)
- label = aa.label;
-
- label += "_cons";
-
- 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();
+ return ref.getSource().toString() + " ; "
+ + ref.getAccessionId().toString();
}
+ @Override
+ public String print(SequenceI[] s, boolean jvSuffix)
+ {
+ out = new StringBuffer();
+ out.append("# STOCKHOLM 1.0");
+ out.append(newline);
+
+ // find max length of id
+ int max = 0;
+ int maxid = 0;
+ int in = 0;
+ int slen = s.length;
+ SequenceI seq;
+ Hashtable dataRef = null;
+ boolean isAA = s[in].isProtein();
+ while ((in < slen) && ((seq = s[in]) != null))
+ {
+ String tmp = printId(seq, jvSuffix);
+ max = Math.max(max, seq.getLength());
+
+ if (tmp.length() > maxid)
+ {
+ maxid = tmp.length();
+ }
+ List seqrefs = seq.getDBRefs();
+ int ndb;
+ if (seqrefs != null && (ndb = seqrefs.size()) > 0)
+ {
+ 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
+ {
+ for (int idb = 0; idb < ndb; idb++)
+ {
+ DBRefEntry dbref = seqrefs.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 : DBRefSource.EMBL)
+ .equals(DBRefUtils.getCanonicalName(dbref.getSource())))
+ {
+ break;
+ }
+ }
+ }
+ }
+ 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 = 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));
+ }
+ else
+ {
+ out.append(" DR " + type + " ");
+ }
+ out.append(newline);
+ }
+ }
+
+ // output annotations
+ while (i < slen && (seq = s[i]) != null)
+ {
+ AlignmentAnnotation[] alAnot = seq.getAnnotation();
+ if (alAnot != null)
+ {
+ 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();
+
+ if (isrna)
+ {
+ // hardwire to secondary structure if there is RNA secondary
+ // structure on the annotation
+ key = "SS";
+ }
+ if (key == null)
+ {
+ continue;
+ }
+
+ // out.append("#=GR ");
+ out.append(new Format("%-" + maxid + "s").form(
+ "#=GR " + printId(s[i], jvSuffix) + " " + key + " "));
+ ann = alAnot[j].annotations;
+ String sseq = "";
+ for (int k = 0; k < ann.length; k++)
+ {
+ sseq += outputCharacter(key, k, isrna, ann, s[i]);
+ }
+ out.append(sseq);
+ out.append(newline);
+ }
+ }
+
+ }
+
+ out.append(new Format("%-" + maxid + "s")
+ .form(printId(seq, jvSuffix) + " "));
+ out.append(seq.getSequenceAsString());
+ out.append(newline);
+ i++;
+ }
+
+ // alignment annotation
+ AlignmentAnnotation aa;
+ AlignmentAnnotation[] an = al.getAlignmentAnnotation();
+ if (an != null)
+ {
+ for (int ia = 0, na = an.length; ia < na; ia++)
+ {
+ aa = an[ia];
+ if (aa.autoCalculated || !aa.visible || aa.sequenceRef != null)
+ {
+ continue;
+ }
+ String sseq = "";
+ String label;
+ String key = "";
+ if (aa.label.equals("seq"))
+ {
+ label = "seq_cons";
+ }
+ else
+ {
+ key = type2id(aa.label.toLowerCase());
+ if (key == null)
+ {
+ label = aa.label;
+ }
+ else
+ {
+ label = key + "_cons";
+ }
+ }
+ 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, nj = aa.annotations.length; j < nj; j++)
+ {
+ sseq += outputCharacter(key, j, isrna, aa.annotations, null);
+ }
+ out.append(sseq);
+ out.append(newline);
+ }
+ }
+
+ out.append("//");
+ out.append(newline);
+
+ return out.toString();
+ }
+
+
+ /**
+ * add an annotation character to the output row
+ *
+ * @param seq
+ * @param key
+ * @param k
+ * @param isrna
+ * @param ann
+ * @param sequenceI
+ */
+ private char outputCharacter(String key, int k, boolean isrna,
+ Annotation[] ann, SequenceI sequenceI)
+ {
+ char seq = ' ';
+ Annotation annot = ann[k];
+ 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
+ ssannotchar = ' ';
+ charset = true;
+ }
+ else
+ {
+ // valid secondary structure AND no alternative label (e.g. ' B')
+ if (annot.secondaryStructure > ' ' && ch.length() < 2)
+ {
+ ssannotchar = annot.secondaryStructure;
+ charset = true;
+ }
+ }
+ if (charset)
+ {
+ return (ssannotchar == ' ' && isrna) ? '.' : ssannotchar;
+ }
+ }
+
+ if (ch.length() == 0)
+ {
+ seq = '.';
+ }
+ else if (ch.length() == 1)
+ {
+ seq = ch.charAt(0);
+ }
+ else if (ch.length() > 1)
+ {
+ seq = ch.charAt(1);
+ }
+
+ 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("# STOCKHOLM 1.0");
- out.append(newline);
- print(getSeqsAsArray());
-
+ out = new StringBuffer();
+ out.append("# STOCKHOLM 1.0");
+ out.append(newline);
+ print(getSeqsAsArray(), false);
+
out.append("//");
- out.append(newline);
+ out.append(newline);
return out.toString();
}
private static Hashtable typeIds = null;
+
static
{
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");
@@ -774,75 +1360,41 @@ 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");
}
}
-
+
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;
}
-
+
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)
+ String key = null;
+ Enumeration e = typeIds.keys();
+ while (e.hasMoreElements())
+ {
+ Object ll = e.nextElement();
+ if (typeIds.get(ll).toString().equalsIgnoreCase(type))
{
- return (String) key;
+ key = (String) ll;
+ break;
}
- System.err.println("Warning : Unknown Stockholm annotation type: "
- + type);
+ }
+ if (key != null)
+ {
+ return 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) {
- *
- * // decide on secondary structure or not. Annotation[] els = new
- * Annotation[ssline.length()]; for (int i = 0; i < ssline.length(); i++) {
- * String pos = ssline.substring(i, i + 1); Annotation ann; ann = new
- * Annotation(pos, "", ' ', 0f); // 0f is 'valid' null - will not
- *
- * ann.secondaryStructure =
- * jalview.schemes.ResidueProperties.getRNAssState(pos).charAt(0);
- *
- * ann.displayCharacter = "x" + ann.displayCharacter;
- *
- * System.out.println(ann.displayCharacter);
- *
- * els[i] = ann; } AlignmentAnnotation helicesAnnot = null; Enumeration e =
- * annotation.elements(); while (e.hasMoreElements()) { helicesAnnot =
- * (AlignmentAnnotation) e.nextElement(); if (helicesAnnot.label.equals(type))
- * break; helicesAnnot = null; } if (helicesAnnot == null) { helicesAnnot =
- * new AlignmentAnnotation(type, type, els);
- * annotation.addElement(helicesAnnot); } else { Annotation[] anns = new
- * Annotation[helicesAnnot.annotations.length + els.length];
- * System.arraycopy(helicesAnnot.annotations, 0, anns, 0,
- * helicesAnnot.annotations.length); System.arraycopy(els, 0, anns,
- * helicesAnnot.annotations.length, els.length); helicesAnnot.annotations =
- * anns; }
- *
- * helicesAnnot.features = Rna.GetBasePairs(ssline);
- * Rna.HelixMap(helicesAnnot.features);
- *
- *
- * return helicesAnnot; }
- */
}