source: trunk/yat/statistics/SAMScore.h @ 966

#ifndef _theplu_yat_statistics_sam_score_
#define _theplu_yat_statistics_sam_score_

// $Id: SAMScore.h 966 2007-10-11 17:01:01Z peter $

/*
  Copyright (C) 2006 Jari Häkkinen, Peter Johansson
  Copyright (C) 2007 Peter Johansson

  This file is part of the yat library, http://trac.thep.lu.se/trac/yat

  The yat library 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 2 of the
  License, or (at your option) any later version.

  The yat library 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 this program; if not, write to the Free Software
  Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA
  02111-1307, USA.
*/

#include "Score.h"

#include <cmath>

namespace theplu {
namespace yat {
  namespace utility {
    class vector;
  }
  namespace classifier {
    class DataLookWeighted1D;
  }
namespace statistics {  

  /**
     @brief Class for score used in Significance Analysis of
     Microarrays (SAM).
  
     The score is similar to the Student t-test but with an added
     fudge factor in denominator to avoid groups with small variance
     getting a large score. \f$ \frac{m_x-m_y}{s+s_0} \f$ 

     see http://www.pnas.org/cgi/content/abstract/98/9/5116 for
     details
  */   
  class SAMScore : public Score
  {

  public:
    ///
    /// @param s0 \f$ s_0 \f$ is a fudge factor
    /// @param absolute if true max(score, -score) is used
    ///
    SAMScore(const double s0, bool absolute=true);

    /**
       \f$ \frac{m_x-m_y}{s+s_0} \f$ where \f$ m = \frac{1}{n_x}\sum
       x_i \f$, \f$ s^2 = \left(\frac{1}{n_x}+\frac{1}{n_y} \right)
       \frac{\sum (x_i-m_x)^2 + \sum(y_i-m_y)^2}{n_x+n_y-2} \f$, and
       \f$ s_0 \f$ is the fudge factor.

       @return SAM score. If absolute=true absolute value of t-score
       is returned
    */
    double score(const classifier::Target& target, 
                 const utility::vector& value) const; 

    /**
       \f$ \frac{m_x-m_y}{s+s_0} \f$ where \f$ m = \frac{\sum
       w_ix_i}{w_i} \f$, \f$ s_0 \f$ is the fudge factor, and \f$ s^2
       = \left(\frac{1}{n_x}+\frac{1}{n_y} \right) \frac{\sum
       w_i(x_i-m_x)^2 + \sum w_i(y_i-m_y)^2}{n_x+n_y-2} \f$ where \f$
       n \f$ is weighted version of number of data points \f$
       \frac{\left(\sum w_i\right)^2}{\sum w_i^2} \f$.

       @return weighted version of SAM score. If absolute=true
       absolute value is returned
    */
    double score(const classifier::Target& target, 
                 const classifier::DataLookupWeighted1D& value) const;          

    /**
       \f$ \frac{m_x-m_y}{s+s_0} \f$ where \f$ m = \frac{\sum
       w_ix_i}{w_i} \f$, \f$ s_0 \f$ is the fudge factor, and \f$ s^2
       = \left(\frac{1}{n_x}+\frac{1}{n_y} \right) \frac{\sum
       w_i(x_i-m_x)^2 + \sum w_i(y_i-m_y)^2}{n_x+n_y-2} \f$ where \f$
       n \f$ is weighted version of number of data points \f$
       \frac{\left(\sum w_i\right)^2}{\sum w_i^2} \f$.

       @return weighted version of SAM score. If absolute=true
       absolute value is returned
    */
    double score(const classifier::Target& target, 
                 const utility::vector& value, 
                 const utility::vector& weight) const;          
  private:
    double s0_;

    template<class T> 
    double score(const T& positive, const T& negative) const
    {
      if(positive.n()+negative.n()<=2) 
        return 0;
      double diff = positive.mean() - negative.mean();
      double s2 = ( (1.0/positive.n()+1.0/negative.n()) * 
                    (positive.sum_xx_centered()+negative.sum_xx_centered()) /
                    (positive.n()+negative.n()-2) );
      if (diff<0 && absolute_)
        return -diff/(sqrt(s2)+s0_);
      return diff/(sqrt(s2)+s0_);
    }
  };

}}} // of namespace statistics, yat, and theplu

#endif