NucleotideDiversity.hpp

/*
    Copyright 2008-2009 Stéphane De Mita, Mathieu Siol

    This file is part of the EggLib library.

    EggLib 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.

    EggLib 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 EggLib.  If not, see <http://www.gnu.org/licenses/>.
*/


#ifndef EGGLIB_NUCLEOTIDEDIVERSITY_HPP
#define EGGLIB_NUCLEOTIDEDIVERSITY_HPP


#include "BaseDiversity.hpp"
#include <string>
#include <vector>



namespace egglib {


    class NucleotideDiversity : public BaseDiversity {

        public:

            NucleotideDiversity();


            virtual ~NucleotideDiversity();


            virtual void load(
                CharMatrix& data,
                bool allowMultipleMutations=false,
                double minimumExploitableData=1.,
                unsigned int ignoreFrequency=0,
                std::string characterMapping=dnaMapping,
                bool useZeroAsAncestral=false
            );


        // accessors for the "site analysis" section

            unsigned int S() const;
            
            unsigned int So() const;
            
            unsigned int eta() const;
            
            double nseff() const;
            
            unsigned int lseff() const;
            
            double nseffo() const;
            
            unsigned int lseffo() const;
            
            unsigned int  npop() const;
            
            unsigned int popLabel(unsigned int popIndex) const; // no check!


        // accessors for the "diversity" section

            double Pi();
            
            double thetaW();
            
            double average_Pi();
            
            double pop_Pi(unsigned int popIndex);  // no check!
            
            double D();

        // accessors for the "outgroup diversity" section
        
            double thetaH();
            
            double thetaL();
            
            double H();
            
            double Z();
            
            double E();

         // accessors for the "differentiation" section
         
            unsigned int FixedDifferences();
            
            unsigned int CommonAlleles();
            
            unsigned int SharedAlleles();
            
            unsigned int SpecificAlleles();
            
            unsigned int SpecificDerivedAlleles();
            
            unsigned int Polymorphisms(unsigned int pop);
            
            unsigned int SpecificAlleles(unsigned int pop);
            
            unsigned int SpecificDerivedAlleles(unsigned int pop);
            
            unsigned int FixedDifferences(unsigned int pop1, unsigned int pop2);

            unsigned int CommonAlleles(unsigned int pop1, unsigned int pop2);

            unsigned int SharedAlleles(unsigned int pop1, unsigned int pop2);


        // accessor for the "triConfigurations" section

            unsigned int triConfiguration(unsigned int index);


        std::vector<unsigned int> polymorphic_positions() const;


        std::vector<unsigned int> singleton_positions() const;


        protected:

            NucleotideDiversity(const NucleotideDiversity& source) { }


            NucleotideDiversity& operator=(const NucleotideDiversity& source) { return *this; }


            void init();  // initializes values
            void clear();  // free memory but doesn't initializes
            
            // diversity (without outgroup)
            void diversity();
            
            // diversity with outgroup
            void outgroupDiversity();
            
            // site patterns
            void differentiation();
            
            // triconfigurations
            void triConfigurations();
            

            // holders for statistics, with booleans flagging groups of stats
            
            bool b_analysisSites;
            
            bool b_diversity;
            
            double  v_Pi;             // nucleotide diversity
            double  v_thetaW;         // theta (Watterson estimator)
            double  v_average_Pi;     // average diversity across populations
            double *v_pop_Pi;         // diversity per population
            double  v_D;              // Tajima's D
            
            bool b_outgroupDiversity;
            
            double v_thetaH;        // theta (Fay and Wu estimator)
            double v_thetaL;        // theta (Zeng estimator)
            double v_H;             // Fay and Wu's H
            double v_Z;             // normalized Fay and Wu's H
            double v_E;             // Zeng et al.'s E
            
            bool b_differentiation;
            
            unsigned int  *v_pairwiseFixedDifferences;
            unsigned int  *v_pairwiseCommonAlleles;
            unsigned int  *v_pairwiseSharedAlleles;
            unsigned int  *v_popPolymorphic;
            unsigned int  *v_popSpecific;
            unsigned int  *v_popSpecificDerived;
            unsigned int   v_countFixedDifferences;
            unsigned int   v_countCommonAlleles;
            unsigned int   v_countSharedAlleles;
            unsigned int   v_countSpecificAlleles;
            unsigned int   v_countSpecificDerivedAlleles;
        
            
            bool b_triConfigurations;
            
            unsigned int  *v_triConfigurations;

    };
}

#endif