*/\r
package jalview.io;\r
\r
+import jalview.datamodel.AlignmentI;\r
import jalview.datamodel.AlignmentAnnotation;\r
import jalview.datamodel.Annotation;\r
+import jalview.datamodel.DBRefEntry;\r
+import jalview.datamodel.Mapping;\r
import jalview.datamodel.Sequence;\r
+import jalview.datamodel.SequenceFeature;\r
import jalview.datamodel.SequenceI;\r
+import jalview.util.Format;\r
\r
import java.io.BufferedReader;\r
import java.io.FileReader;\r
import java.util.ArrayList;\r
import java.util.Enumeration;\r
import java.util.Hashtable;\r
+import java.util.List;\r
+import java.util.StringTokenizer;\r
import java.util.Vector;\r
\r
import com.stevesoft.pat.Regex;\r
{\r
// static Logger logger = Logger.getLogger("jalview.io.StockholmFile");\r
protected ArrayList<RNA> result;\r
+ \r
+ StringBuffer out; // output buffer\r
\r
- public String id;\r
+ AlignmentI al;\r
\r
public StockholmFile()\r
{\r
}\r
\r
+ /**
+ * Creates a new StockholmFile object for output.
+ */
+ public StockholmFile(AlignmentI al)
+ {
+ this.al = al;
+ }
+
public StockholmFile(String inFile, String type) throws IOException\r
{\r
super(inFile, type);\r
*/\r
public void parse() throws IOException\r
{\r
+ StringBuffer treeString = new StringBuffer();\r
+ String treeName = null;\r
+ // --------------- Variable Definitions -------------------\r
+ String line;\r
+ String version;\r
+ // String id;
+ Hashtable seqAnn = new Hashtable(); // Sequence related annotations\r
+ Hashtable seqs = new Hashtable();\r
+ Regex p, r, rend, s, x;\r
+ // Temporary line for processing RNA annotation\r
+ // String RNAannot = "";\r
+\r
+ // ------------------ Parsing File ----------------------\r
+ // First, we have to check that this file has STOCKHOLM format, i.e. the\r
+ // first line must match\r
+ \r
+ \r
+ r = new Regex("# STOCKHOLM ([\\d\\.]+)");\r
+ if (!r.search(nextLine()))\r
+ {\r
+ throw new IOException(\r
+ "This file is not in valid STOCKHOLM format: First line does not contain '# STOCKHOLM'");\r
+ }\r
+ else\r
+ {\r
+ version = r.stringMatched(1);\r
+ \r
+ // logger.debug("Stockholm version: " + version);\r
+ }\r
+\r
+ // We define some Regexes here that will be used regularily later\r
+ rend = new Regex("^\\s*\\/\\/"); // Find the end of an alignment\r
+ p = new Regex("(\\S+)\\/(\\d+)\\-(\\d+)"); // split sequence id in\r
+ // id/from/to\r
+ s = new Regex("(\\S+)\\s+(\\S*)\\s+(.*)"); // Parses annotation subtype\r
+ r = new Regex("#=(G[FSRC]?)\\s+(.*)"); // Finds any annotation line\r
+ x = new Regex("(\\S+)\\s+(\\S+)"); // split id from sequence\r
+\r
+ // Convert all bracket types to parentheses (necessary for passing to VARNA)\r
+ Regex openparen = new Regex("(<|\\[)", "(");\r
+ Regex closeparen = new Regex("(>|\\])", ")");\r
+\r
+ // Detect if file is RNA by looking for bracket types\r
+ Regex detectbrackets = new Regex("(<|>|\\[|\\]|\\(|\\))");\r
+\r
+ rend.optimize();\r
+ p.optimize();\r
+ s.optimize();\r
+ r.optimize();\r
+ x.optimize();\r
+ openparen.optimize();\r
+ closeparen.optimize();\r
+ \r
+ while ((line = nextLine()) != null)\r
+ {\r
+ if (line.length() == 0)\r
+ {\r
+ continue;\r
+ }\r
+ if (rend.search(line))\r
+ {\r
+ // End of the alignment, pass stuff back\r
+ this.noSeqs = seqs.size();
+ \r
+ 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);\r
+ Enumeration accs = seqs.keys();\r
+ while (accs.hasMoreElements())\r
+ {\r
+ String acc = (String) accs.nextElement();\r
+ // logger.debug("Processing sequence " + acc);\r
+ String seq = (String) seqs.remove(acc);\r
+ if (maxLength < seq.length())\r
+ {\r
+ maxLength = seq.length();\r
+ }\r
+ int start = 1;\r
+ int end = -1;\r
+ String sid = acc;\r
+ /*\r
+ * Retrieve hash of annotations for this accession Associate
+ * Annotation with accession
+ */\r
+ Hashtable accAnnotations = null;\r
+ \r
+ if (seqAnn != null && seqAnn.containsKey(acc))\r
+ {\r
+ accAnnotations = (Hashtable) seqAnn.remove(acc);\r
+ //TODO: add structures to sequence\r
+ }\r
+ \r
+ // Split accession in id and from/to\r
+ if (p.search(acc))\r
+ {\r
+ sid = p.stringMatched(1);\r
+ start = Integer.parseInt(p.stringMatched(2));\r
+ end = Integer.parseInt(p.stringMatched(3));\r
+ }\r
+ // logger.debug(sid + ", " + start + ", " + end);\r
+ \r
+ Sequence seqO = new Sequence(sid, seq, start, end);\r
+ // Add Description (if any)\r
+ if (accAnnotations != null && accAnnotations.containsKey("DE"))\r
+ {\r
+ String desc = (String) accAnnotations.get("DE");\r
+ seqO.setDescription((desc == null) ? "" : desc);\r
+ }\r
+ // Add DB References (if any)\r
+ if (accAnnotations != null && accAnnotations.containsKey("DR"))\r
+ {\r
+ String dbr = (String) accAnnotations.get("DR");\r
+ if (dbr != null && dbr.indexOf(";") > -1)\r
+ {\r
+ String src = dbr.substring(0, dbr.indexOf(";"));\r
+ String acn = dbr.substring(dbr.indexOf(";") + 1);\r
+ jalview.util.DBRefUtils.parseToDbRef(seqO, src, "0", acn);\r
+ }\r
+ } \r
+
+ if (accAnnotations != null && accAnnotations.containsKey("AC"))
+ {
+ if (dbsource != null)
+ {
+ String dbr = (String) accAnnotations.get("AC");
+ if (dbr != null)
+ {\r
+ // 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);
+ \r
+ }\r
+ }\r
+ // else - do what ? add the data anyway and prompt the user to specify what references these are ?
+ }
+ \r
+ Hashtable features = null;\r
+ // We need to adjust the positions of all features to account for gaps\r
+ try\r
+ {\r
+ features = (Hashtable) accAnnotations.remove("features");\r
+ } catch (java.lang.NullPointerException e)\r
+ {\r
+ // loggerwarn("Getting Features for " + acc + ": " +\r
+ // e.getMessage());\r
+ // continue;\r
+ }\r
+ // if we have features\r
+ if (features != null)\r
+ {\r
+ int posmap[] = seqO.findPositionMap();\r
+ Enumeration i = features.keys();\r
+ while (i.hasMoreElements())\r
+ {\r
+ // TODO: parse out secondary structure annotation as annotation\r
+ // row\r
+ // TODO: parse out scores as annotation row\r
+ // TODO: map coding region to core jalview feature types\r
+ String type = i.nextElement().toString();\r
+ Hashtable content = (Hashtable) features.remove(type);\r
+
+ // 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();\r
+ while (j.hasMoreElements())\r
+ {\r
+ String desc = j.nextElement().toString();\r
+ String ns = content.get(desc).toString();\r
+ char[] byChar = ns.toCharArray();\r
+ for (int k = 0; k < byChar.length; k++)\r
+ {\r
+ char c = byChar[k];\r
+ if (!(c == ' ' || c == '_' || c == '-' || c == '.')) // PFAM\r
+ // uses\r
+ // '.'\r
+ // for\r
+ // feature\r
+ // background\r
+ {\r
+ int new_pos = posmap[k]; // look up nearest seqeunce\r
+ // position to this column\r
+ SequenceFeature feat = new SequenceFeature(type, desc,\r
+ new_pos, new_pos, 0f, null);\r
+ \r
+ seqO.addSequenceFeature(feat);\r
+ }\r
+ }\r
+ }\r
+ \r
+ }\r
+ \r
+ }\r
+ // garbage collect\r
+ \r
+ // logger.debug("Adding seq " + acc + " from " + start + " to " + end\r
+ // + ": " + seq);\r
+ this.seqs.addElement(seqO);\r
+ }\r
+ return; // finished parsing this segment of source\r
+ }\r
+ else if (!r.search(line))\r
+ {\r
+ // System.err.println("Found sequence line: " + line);\r
+ \r
+ // Split sequence in sequence and accession parts\r
+ if (!x.search(line))\r
+ {\r
+ // logger.error("Could not parse sequence line: " + line);\r
+ throw new IOException("Could not parse sequence line: " + line);\r
+ }\r
+ String ns = (String) seqs.get(x.stringMatched(1));\r
+ if (ns == null)\r
+ {\r
+ ns = "";\r
+ }\r
+ ns += x.stringMatched(2);\r
+ \r
+ seqs.put(x.stringMatched(1), ns);\r
+ }\r
+ else\r
+ {\r
+ String annType = r.stringMatched(1);\r
+ String annContent = r.stringMatched(2);\r
+ \r
+ // System.err.println("type:" + annType + " content: " + annContent);\r
+ \r
+ if (annType.equals("GF"))\r
+ {\r
+ /*\r
+ * Generic per-File annotation, free text Magic features: #=GF NH\r
+ * <tree in New Hampshire eXtended format> #=GF TN <Unique identifier\r
+ * for the next tree> Pfam descriptions: 7. DESCRIPTION OF FIELDS\r
+ * \r
+ * Compulsory fields: ------------------\r
+ * \r
+ * AC Accession number: Accession number in form PFxxxxx.version or\r
+ * PBxxxxxx. ID Identification: One word name for family. DE\r
+ * Definition: Short description of family. AU Author: Authors of the\r
+ * entry. SE Source of seed: The source suggesting the seed members\r
+ * belong to one family. GA Gathering method: Search threshold to\r
+ * build the full alignment. TC Trusted Cutoff: Lowest sequence score\r
+ * and domain score of match in the full alignment. NC Noise Cutoff:\r
+ * Highest sequence score and domain score of match not in full\r
+ * alignment. TP Type: Type of family -- presently Family, Domain,\r
+ * Motif or Repeat. SQ Sequence: Number of sequences in alignment. AM\r
+ * Alignment Method The order ls and fs hits are aligned to the model\r
+ * to build the full align. // End of alignment.\r
+ * \r
+ * Optional fields: ----------------\r
+ * \r
+ * DC Database Comment: Comment about database reference. DR Database\r
+ * Reference: Reference to external database. RC Reference Comment:\r
+ * Comment about literature reference. RN Reference Number: Reference\r
+ * Number. RM Reference Medline: Eight digit medline UI number. RT\r
+ * Reference Title: Reference Title. RA Reference Author: Reference\r
+ * Author RL Reference Location: Journal location. PI Previous\r
+ * identifier: Record of all previous ID lines. KW Keywords: Keywords.\r
+ * CC Comment: Comments. NE Pfam accession: Indicates a nested domain.\r
+ * NL Location: Location of nested domains - sequence ID, start and\r
+ * end of insert.\r
+ * \r
+ * Obsolete fields: ----------- AL Alignment method of seed: The\r
+ * method used to align the seed members.\r
+ */\r
+ // Let's save the annotations, maybe we'll be able to do something\r
+ // with them later...\r
+ Regex an = new Regex("(\\w+)\\s*(.*)");\r
+ if (an.search(annContent))\r
+ {\r
+ if (an.stringMatched(1).equals("NH"))\r
+ {\r
+ treeString.append(an.stringMatched(2));\r
+ }\r
+ else if (an.stringMatched(1).equals("TN"))\r
+ {\r
+ if (treeString.length() > 0)\r
+ {\r
+ if (treeName == null)\r
+ {\r
+ treeName = "Tree " + (getTreeCount() + 1);\r
+ }\r
+ addNewickTree(treeName, treeString.toString());\r
+ }\r
+ treeName = an.stringMatched(2);\r
+ treeString = new StringBuffer();\r
+ }\r
+ setAlignmentProperty(an.stringMatched(1), an.stringMatched(2));\r
+ }\r
+ }\r
+ else if (annType.equals("GS"))\r
+ {\r
+ // Generic per-Sequence annotation, free text\r
+ /*\r
+ * Pfam uses these features: Feature Description ---------------------\r
+ * ----------- AC <accession> ACcession number DE <freetext>\r
+ * DEscription DR <db>; <accession>; Database Reference OS <organism>\r
+ * OrganiSm (species) OC <clade> Organism Classification (clade, etc.)\r
+ * LO <look> Look (Color, etc.)\r
+ */\r
+ if (s.search(annContent))\r
+ {\r
+ String acc = s.stringMatched(1);\r
+ String type = s.stringMatched(2);\r
+ String content = s.stringMatched(3);\r
+ // TODO: store DR in a vector.\r
+ // TODO: store AC according to generic file db annotation.\r
+ Hashtable ann;\r
+ if (seqAnn.containsKey(acc))\r
+ {\r
+ ann = (Hashtable) seqAnn.get(acc);\r
+ }\r
+ else\r
+ {\r
+ ann = new Hashtable();\r
+ }\r
+ ann.put(type, content);\r
+ seqAnn.put(acc, ann);\r
+ }\r
+ else\r
+ {\r
+ throw new IOException("Error parsing " + line);\r
+ }\r
+ }\r
+ else if (annType.equals("GC"))\r
+ {\r
+ // Generic per-Column annotation, exactly 1 char per column\r
+ // always need a label.\r
+ if (x.search(annContent))\r
+ {\r
+ // parse out and create alignment annotation directly.\r
+ parseAnnotationRow(annotations, x.stringMatched(1),\r
+ x.stringMatched(2));\r
+ }\r
+ }\r
+ else if (annType.equals("GR"))\r
+ {\r
+ // Generic per-Sequence AND per-Column markup, exactly 1 char per\r
+ // column\r
+ /*\r
+ * Feature Description Markup letters ------- -----------\r
+ * -------------- SS Secondary Structure [HGIEBTSCX] SA Surface\r
+ * Accessibility [0-9X] (0=0%-10%; ...; 9=90%-100%) TM TransMembrane\r
+ * [Mio] PP Posterior Probability [0-9*] (0=0.00-0.05; 1=0.05-0.15;\r
+ * *=0.95-1.00) LI LIgand binding [*] AS Active Site [*] IN INtron (in\r
+ * or after) [0-2]\r
+ */\r
+ if (s.search(annContent))\r
+ {\r
+ String acc = s.stringMatched(1);\r
+ String type = s.stringMatched(2);\r
+ String seq = new String(s.stringMatched(3));\r
+ String description = null;\r
+ // Check for additional information about the current annotation\r
+ // We use a simple string tokenizer here for speed\r
+ StringTokenizer sep = new StringTokenizer(seq, " \t");\r
+ description = sep.nextToken();\r
+ if (sep.hasMoreTokens())\r
+ {\r
+ seq = sep.nextToken();\r
+ }\r
+ else\r
+ {\r
+ seq = description;\r
+ description = new String();\r
+ }\r
+ // sequence id with from-to fields\r
+ \r
+ Hashtable ann;\r
+ // Get an object with all the annotations for this sequence\r
+ if (seqAnn.containsKey(acc))\r
+ {\r
+ // logger.debug("Found annotations for " + acc);\r
+ ann = (Hashtable) seqAnn.get(acc);\r
+ }\r
+ else\r
+ {\r
+ // logger.debug("Creating new annotations holder for " + acc);\r
+ ann = new Hashtable();\r
+ seqAnn.put(acc, ann);\r
+ }\r
+ // TODO test structure, call parseAnnotationRow with vector from
+ // hashtable for specific sequence
+ Hashtable features;\r
+ // Get an object with all the content for an annotation\r
+ if (ann.containsKey("features"))\r
+ {\r
+ // logger.debug("Found features for " + acc);\r
+ features = (Hashtable) ann.get("features");\r
+ }\r
+ else\r
+ {\r
+ // logger.debug("Creating new features holder for " + acc);\r
+ features = new Hashtable();\r
+ ann.put("features", features);\r
+ }\r
+ \r
+ Hashtable content;\r
+ if (features.containsKey(this.id2type(type)))\r
+ {\r
+ // logger.debug("Found content for " + this.id2type(type));\r
+ content = (Hashtable) features.get(this.id2type(type));\r
+ }\r
+ else\r
+ {\r
+ // logger.debug("Creating new content holder for " +\r
+ // this.id2type(type));\r
+ content = new Hashtable();\r
+ features.put(this.id2type(type), content);\r
+ }\r
+ String ns = (String) content.get(description);\r
+ if (ns == null)\r
+ {\r
+ ns = "";\r
+ }\r
+ ns += seq;\r
+ content.put(description, ns);\r
+
+// if(type.equals("SS")){\r
+ Hashtable strucAnn;\r
+ if (seqAnn.containsKey(acc))\r
+ {\r
+ strucAnn = (Hashtable) seqAnn.get(acc);\r
+ }\r
+ else\r
+ {\r
+ strucAnn = new Hashtable();\r
+ }\r
+ \r
+ Vector newStruc=new Vector();\r
+ parseAnnotationRow(newStruc, type,ns);\r
+ \r
+ strucAnn.put(type, newStruc);\r
+ seqAnn.put(acc, strucAnn);\r
+ }\r
+// }\r
+ else\r
+ {\r
+ System.err\r
+ .println("Warning - couldn't parse sequence annotation row line:\n"\r
+ + line);\r
+ // throw new IOException("Error parsing " + line);\r
+ }\r
+ }\r
+ else\r
+ {\r
+ throw new IOException("Unknown annotation detected: " + annType\r
+ + " " + annContent);\r
+ }\r
+ }\r
+ }\r
+ if (treeString.length() > 0)\r
+ {\r
+ if (treeName == null)\r
+ {\r
+ treeName = "Tree " + (1 + getTreeCount());\r
+ }\r
+ addNewickTree(treeName, treeString.toString());\r
+ }\r
+ }\r
+\r
+/**\r
+ * Demangle an accession string and guess the originating sequence database for a given sequence\r
+ * @param seqO sequence to be annotated\r
+ * @param dbr Accession string for sequence\r
+ * @param dbsource source database for alignment (PFAM or RFAM)\r
+ */\r
+ private void guessDatabaseFor(Sequence seqO, String dbr, String dbsource)\r
+ {\r
+ DBRefEntry dbrf=null;\r
+ List<DBRefEntry> dbrs=new ArrayList<DBRefEntry>();\r
+ String seqdb="Unknown",sdbac=""+dbr;\r
+ int st=-1,en=-1,p;\r
+ if ((st=sdbac.indexOf("/"))>-1)\r
+ {\r
+ String num,range=sdbac.substring(st+1);\r
+ sdbac = sdbac.substring(0,st);\r
+ if ((p=range.indexOf("-"))>-1)\r
+ {\r
+ p++;\r
+ if (p<range.length())\r
+ {\r
+ num = range.substring(p).trim();\r
+ try {\r
+ en = Integer.parseInt(num);\r
+ } catch (NumberFormatException x)\r
+ {\r
+ // could warn here that index is invalid\r
+ en = -1;\r
+ }\r
+ }\r
+ } else {\r
+ p=range.length();\r
+ }\r
+ num=range.substring(0,p).trim();\r
+ try {\r
+ st = Integer.parseInt(num);\r
+ } catch (NumberFormatException x)\r
+ {\r
+ // could warn here that index is invalid\r
+ st = -1;\r
+ }\r
+ }\r
+ if (dbsource.equals("PFAM")) {\r
+ seqdb = "UNIPROT";\r
+ if (sdbac.indexOf(".")>-1)\r
+ {\r
+ // strip of last subdomain\r
+ sdbac = sdbac.substring(0,sdbac.indexOf("."));\r
+ dbrf = jalview.util.DBRefUtils.parseToDbRef(seqO, seqdb, dbsource, sdbac);\r
+ if (dbrf!=null)\r
+ {\r
+ dbrs.add(dbrf);\r
+ }\r
+ }\r
+ dbrf = jalview.util.DBRefUtils.parseToDbRef(seqO, dbsource, dbsource, dbr);\r
+ if (dbr!=null)\r
+ {\r
+ dbrs.add(dbrf);\r
+ }\r
+ } else {\r
+ seqdb = "EMBL"; // total guess - could be ENA, or something else these days\r
+ if (sdbac.indexOf(".")>-1)\r
+ {\r
+ // strip off last subdomain\r
+ sdbac = sdbac.substring(0,sdbac.indexOf("."));\r
+ dbrf = jalview.util.DBRefUtils.parseToDbRef(seqO, seqdb, dbsource, sdbac);\r
+ if (dbrf!=null)\r
+ {\r
+ dbrs.add(dbrf);\r
+ }\r
+ }\r
+ \r
+ dbrf = jalview.util.DBRefUtils.parseToDbRef(seqO, dbsource, dbsource, dbr);\r
+ if (dbrf!=null)\r
+ {\r
+ dbrs.add(dbrf);\r
+ }\r
+ }\r
+ if (st!=-1 && en!=-1)\r
+ {\r
+ for (DBRefEntry d:dbrs)\r
+ {\r
+ jalview.util.MapList mp = new jalview.util.MapList(new int[] { seqO.getStart(),seqO.getEnd()},new int[] { st,en},1,1);\r
+ jalview.datamodel.Mapping mping = new Mapping(mp);\r
+ d.setMap(mping);\r
+ }\r
+ }\r
+ }\r
+\r
+\r
+ /**\r
+ * Parse a file in Stockholm format into Jalview's data model using VARNA\r
+ * \r
+ * @throws IOException\r
+ * If there is an error with the input file\r
+ */\r
+ public void parse_with_VARNA() throws IOException\r
+ {\r
FileReader fr = null;\r
fr = new FileReader(inFile);\r
\r
// DEBUG System.out.println("this is the secondary scructure:"\r
// +result.size());\r
SequenceI[] seqs = new SequenceI[result.size()];\r
+ String id=null;\r
for (int i = 0; i < result.size(); i++)\r
{\r
// DEBUG System.err.println("Processing i'th sequence in Stockholm file")\r
convert2 = closeparen.replaceAll(convert1);\r
annots = convert2;\r
\r
- String type = (label.indexOf("_cons") == label.length() - 5) ? label\r
- .substring(0, label.length() - 5) : label;\r
+ String type = label;\r
+ if (label.contains("_cons"))\r
+ {\r
+ type = (label.indexOf("_cons") == label.length() - 5) ? label\r
+ .substring(0, label.length() - 5) : label;\r
+ }\r
boolean ss = false;\r
type = id2type(type);\r
if (type.equals("secondary structure"))\r
return annot;\r
}\r
\r
- public static String print(SequenceI[] s)\r
+ public String print(SequenceI[] s)
+ {
+ // 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)
{\r
- return "not yet implemented";\r
+ seq += ch.charAt(1);
+ }
+ }
+ out.append(seq);
+ out.append(newline);
+ }
+ }
+ return out.toString();
}\r
\r
public String print()\r
{\r
- return print(getSeqsAsArray());\r
+ out = new StringBuffer();
+ out.append("# STOCKHOLM 1.0");
+ out.append(newline);
+ print(getSeqsAsArray());
+
+ out.append("//");
+ out.append(newline);
+ return out.toString();
}\r
\r
private static Hashtable typeIds = null;\r
+ id);\r
return id;\r
}\r
+
+ 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;
+ }
/**\r
* make a friendly ID string.\r
* \r
dataName = dataName.substring(1, e).trim();\r
return dataName;\r
}\r
-}
\ No newline at end of file
+}\r