Blmodel.h

Go to the documentation of this file.

#ifndef BLMODEL_H
#define BLMODeL_H
#include "Match.h"
#include <mysql++.h>

class Blmodel : public Match {
   public:
      Blmodel(const string &q, const string &t, int fp, int ql, int tl,
            int qb, int qe, int tb, int te, int ex, double sc, 
            double ide, double qc, float ol)
         : Match(qb,qe,tb,te), qid(q), tid(t), fpnum(fp), qlen(ql), tlen(tl),
            numex(ex), score(sc), iden(ide), qcov(qc), olp(ol)
            { }

      /* the order of the column is in the qbegin,qend,tbegin,tend
       * in the input table */
      Blmodel(const mysqlpp::Row &r)
         : Match(int(r[5]), int(r[6]), int(r[7]), int(r[8])),
               qid(string(r.at(0))), tid(string(r[1])), fpnum(int(r[2])), 
               qlen(int(r[3])), tlen(int(r[4])), numex(int(r[9])), 
               score(double(r[10])), iden(double(r[11])), 
               qcov(double(r[12])), olp(float(r[13])) 
               { }

      const string& getQid() const { return qid; }
      const string& getTid() const { return tid; }
      int getFpnum() const { return fpnum; }

      /* Obtain the garget boundary that is likely to contain the
       * complete gene. If the guide protein sequence start from 1
       * then the 5' end of the gene is only expanded by 6 nt.
       * Same is true if the guide protein end is used up in the
       * alignment.
       *
       * Parameters: 
       *     ease is the 1/2 of intergenic + 5'-UTR on the 5'-side
       *         and the 1/2 of intergenic + 3'-UTR
       *     cr is the genomic/CDS ratio this parameter depends on genomes.
       *         This ratio will be computed from the underlying model by
       *         target range/ query range.  The program will pick the larger
       *         of t/q ratio and cr to compute the covered region in genomic.
       *
       * Return: the start and end of the target.  
       *
       * Use bioseq coordinates: start from 1
       * The cr is the ratio of total DNA to coding.
       * For human this number is very high 50 to 100,
       * for lower eukaryotes this number is cloase to 3-4.
       * Default value is 4.1.
       * */
      pair<int,int> getExpectedTargetRange(int ease=500, double cr=4.1) const;

      int getExonNum() const { return numex; }
      /* the score is derived from the sum of the scores of the
       * matching fragments
       */
      double getScore() const { return score; }
      /* normalized score, removed the overlapping regions
       */
      double getNormScore() const { return score/(1+olp); }
      int queryLength() const { return qlen; }
      int targetLength() const { return tlen; }
      friend ostream& operator<<(ostream &ous, const Blmodel &bm);
      bool queryHasStart() const { return 1 == queryBegin(); }
      bool queryHasStop() const { return qlen == queryEnd(); }

   private:
      string qid;
      string tid;
      int fpnum, qlen, tlen, numex;
      double score, iden, qcov;
      float olp;
      //bool hasbegin; has been removed from the new version
      //bool hasend; has been removed from the new version
};

#endif