src/uk/ac/vamsas/objects/utils/Seq.java

   1 /*
   2  * Created on 17-May-2005
   3  *
   4  * TODO To change the template for this generated file go to
   5  * Window - Preferences - Java - Code Style - Code Templates
   6  */
   7 package uk.ac.vamsas.objects.utils;
   8
   9 import java.io.BufferedWriter;
  10 import java.io.IOException;
  11 import java.io.OutputStream;
  12 import java.io.OutputStreamWriter;
  13 import java.util.regex.Pattern;
  14
  15 import uk.ac.vamsas.objects.core.AlignmentSequence;
  16 import uk.ac.vamsas.objects.core.Sequence;
  17 import uk.ac.vamsas.objects.core.SequenceType;
  18
  19 /**
  20  * @author jimp
  21  *
  22  * TODO To change the template for this generated type comment go to
  23  * Window - Preferences - Java - Code Style - Code Templates
  24  */
  25 public class Seq {
  26
  27         public static void write_PirSeq(OutputStream os, SequenceType seq, int wid) throws IOException {
  28                 BufferedWriter pir_out = new BufferedWriter(new OutputStreamWriter(os));
  29                 pir_out.write(">P1;"+seq.getName()+"\n");
  30                 int width = (wid<1) ? 80 : wid;
  31                 for (int j=0,k=seq.getSequence().length(); j<k; j+=width)
  32                         if (j+width<k)
  33                                 pir_out.write(seq.getSequence().substring(j,j+width)+"\n");
  34                         else
  35                                 pir_out.write(seq.getSequence().substring(j)+"\n");
  36                 pir_out.flush();
  37           }
  38
  39         public static void write_FastaSeq(OutputStream os, SequenceType seq) throws IOException {
  40                 BufferedWriter fasta_out = new BufferedWriter(new OutputStreamWriter(os));
  41                 fasta_out.write(">"+seq.getName()+"\n");
  42                 fasta_out.write(seq.getSequence()+"\n");
  43                 fasta_out.flush();
  44           }
  45
  46         public static void write_FastaSeq(OutputStream os, SequenceType seq, int wid) throws IOException {
  47                 BufferedWriter fasta_out = new BufferedWriter(new OutputStreamWriter(os));
  48                 fasta_out.write(">"+seq.getName()+"\n");
  49                 int width = (wid<1) ? 80 : wid;
  50                 for (int j=0,k=seq.getSequence().length(); j<k; j+=width)
  51                         if (j+width<k)
  52                                 fasta_out.write(seq.getSequence().substring(j,j+width)+"\n");
  53                         else
  54                                 fasta_out.write(seq.getSequence().substring(j)+"\n");
  55                 fasta_out.flush();
  56           }
  57         /**
  58    *validate a SequenceType Vobject as an info:iubmb.org/aminoacid SequenceType
  59    *This version resolves references to Sequence objects from AlignmentSequence
  60    *TODO: Define info: urn for dictionary string (could also be regex of valid characters!)
  61    * @param s
  62          * @param dict TODO
  63    * @return true if a valid amino acid sequence Vobject
  64          */
  65         private static boolean valid_aadictionary_string(String s, String dict) {
  66     if (s==null)
  67       return false;
  68     // validate against dictionary
  69     // TODO generalise to resolve dictionary against info: urn for dictionary type
  70       Pattern aa_repl = Pattern.compile("[ARNDCQEGHILKMFPSTWYVUX]+", Pattern.CASE_INSENSITIVE);
  71       String remnants = aa_repl.matcher(s).replaceAll("");
  72       return !remnants.matches("//S+");
  73   }
  74
  75   public static Sequence newSequence(String Name, String Sequence, String Dictionary, int start, int end) {
  76     //TODO: make hierarchy reflecting the SeqType Vobject.
  77     Sequence seq= new Sequence();
  78      seq.setDictionary(Dictionary);
  79      seq.setName(Name);
  80      seq.setSequence(Sequence);
  81      seq.setStart(start);
  82      if (start<=end) {
  83        if ((end-start)!=Sequence.length())
  84          seq.setEnd(start+Sequence.length());
  85      } else {
  86        // reverse topology mapping. TODO: VAMSAS: decide if allowed to do start>end on Sequence Vobject
  87        if ((start-end)!=Sequence.length())
  88          seq.setEnd(end+Sequence.length());
  89      }
  90      return seq;
  91   }
  92   public static AlignmentSequence newAlignmentSequence(String name, String alSequence, Sequence refseq, long start, long end) {
  93     if (refseq!=null) {
  94       AlignmentSequence asq = new AlignmentSequence();
  95       asq.setName(name);
  96       asq.setSequence(alSequence);
  97       asq.setRefid(refseq);
  98       if (end>refseq.getEnd() || end<start || end==-1)
  99         end = refseq.getEnd();
 100       asq.setEnd(end);
 101       if (start<refseq.getStart())
 102         start = refseq.getStart();
 103       asq.setStart(start);
 104       return asq;
 105     }
 106     return null;
 107   }
 108   public static boolean is_valid_aa_seq(SequenceType s) {
 109     Sequence q;
 110     boolean validref=false;
 111     if (s instanceof Sequence) {
 112       q=(Sequence) s;
 113       if (q.getDictionary()!=null
 114
 115         && q.getDictionary().length()>0
 116         || !q.getDictionary().equals(SymbolDictionary.STANDARD_AA))
 117         return false;
 118       return valid_aadictionary_string(q.getSequence(), SymbolDictionary.STANDARD_AA);
 119     }
 120
 121     // follow references
 122     if (s instanceof AlignmentSequence) {
 123       Object w = (((AlignmentSequence) s).getRefid());
 124       if (w!=null && w!=s && w instanceof SequenceType)
 125         return is_valid_aa_seq((SequenceType) w)
 126         && valid_aadictionary_string(((AlignmentSequence) s).getSequence(), SymbolDictionary.STANDARD_AA);
 127     }
 128
 129     return false;
 130   }
 131 }