Bio::Tools RandomDistFunctions
SummaryIncluded librariesPackage variablesSynopsisDescriptionGeneral documentationMethods
Summary
Bio::Tools::RandomDistFunctions - A set of routines useful for generating random data in different distributions
Package variables
No package variables defined.
Included modules
Bio::Root::Root
POSIX
Inherit
Bio::Root::Root
Synopsis
  use Bio::Tools::RandomDistFunctions;
  my $dist = Bio::Tools::RandomDistFunctions->new();
  for my $v ( 1..1000 ) { 
    my $birth_dist = $dist->rand_birth_distribution($lambda);
    # ... do something with the variable
  }
Description
Most of the code is based on the C implementation of these routines in
Mike Sanderson's r8s's package. See http://ginger.ucdavis.edu/ for
information on his software.
This code tries to be fast and use available faster BigInt and GMP
library methods when those modules are available.
Methods
rand_birth_distribution
No description
Code
rand_geometric_distributionDescriptionCode
rand_exponentional_distributionDescriptionCode
rand_normal_distributionDescriptionCode
Methods description
rand_geometric_distributioncode    nextTop
 Title   : rand_geometric_distribution
 Usage   : my $randvar = $dist->rand_geometric_distribution($param);
 Function: Returns a random geometric variate distributed with 
           paramater $param, according to
           c.d.f. 1 - ( 1- param) ^ n 
 Returns : integer
 Args    : $param ( 0 > $param < 1 )
rand_exponentional_distributioncodeprevnextTop
 Title   : rand_exponentional_distribution
 Usage   : my $var = $dist->rand_exponentional_distribution($param);
 Function: Returns a random exponential variate distributed with parameter
           $param, according to c.d.f 1 - e^(-param * x)
 Returns : floating point number 
 Args    : $param ( > 0 )
rand_normal_distributioncodeprevnextTop
 Title   : rand_normal_distribution
 Usage   : my $var = $dist->rand_normal_distribution()
 Function: Returns a random normal (gaussian) variate distributed 
 Returns : floating point number
 Args    : none
Methods code
rand_birth_distributiondescriptionprevnextTop
sub rand_birth_distribution {
   my ($self,$lambda) = @_;
   if( ! ref($self) && 
       $self !~ /RandomDistFunctions/ ) { 
       $lambda = $self;
   }
   unless( $lambda ) { 
       die("Cannot call birth_distribution without a valid lambda value (>0)");
   }
   return 1 - (log(rand(1) * (exp($lambda) - 1)+1)/ $lambda);
}
rand_geometric_distributiondescriptionprevnextTop
sub rand_geometric_distribution {
   my ($self,$param) = @_;
   if( ! ref($self) && 
       $self !~ /RandomDistFunctions/ ) { 
       $param = $self;
   }
   unless( $param ) { 
       die("Cannot call rand_geometric_distribution without a valid param value (>0)");
   }

   my $den;
   if( $param < 1e-8) { 
       $den = (-1 * $param) - ( $param * $param ) / 2;
   } else { 
       $den = log(1 - $param);
   }
   my $z = log(1 - rand(1)) / $den;
   return POSIX::floor($z) + 1;
   # MSanderson comments from r8s code
   # Is this the right truncation of the real-valued expression above?
   # YES
   # Checked by reference to the expected mean of the distribution in
   # 100,000 replicates
   # EX = 1/param Var = (1-param)/param^2 See Olkin, Gleser, and
   # Derman, p. 193ff. Probability Models and Applications, 1980.
}
rand_exponentional_distributiondescriptionprevnextTop
sub rand_exponentional_distribution {
   my ($self,$param) = @_;
      if( ! ref($self) && 
       $self !~ /RandomDistFunctions/ ) { 
       $param = $self;
   }
   unless( $param ) { 
       die("Cannot call rand_exponentional_distribution without a valid param value (>0)");
   }
   return log( 1- rand(1)) / $param;
}
rand_normal_distributiondescriptionprevnextTop
sub rand_normal_distribution {
    my $gset;
    my ($rsq,$v1,$v2) = ( 0,0,0);
    do { 
	$v1 = 2 * rand(1) - 1;
	$v2 = 2 * rand(1) - 1;
	$rsq= $v1**2 + $v2 ** 2;
    } while( $rsq >= 1 || $rsq == 0);
    my $fac = sqrt(-2.0 * log($rsq) / $rsq );
    $gset = $v1 * $fac;
    return $v2 * $fac;
}
General documentation
FEEDBACKTop
Mailing ListsTop
User feedback is an integral part of the evolution of this and other
Bioperl modules. Send your comments and suggestions preferably to
the Bioperl mailing list. Your participation is much appreciated.
  bioperl-l@bioperl.org              - General discussion
  http://bioperl.org/MailList.shtml  - About the mailing lists
Reporting BugsTop
Report bugs to the Bioperl bug tracking system to help us keep track
of the bugs and their resolution. Bug reports can be submitted via
email or the web:
  bioperl-bugs@bioperl.org
  http://bioperl.org/bioperl-bugs/
AUTHOR - Jason StajichTop
Email jason-at-bioperl.org
CONTRIBUTORSTop
Thanks to Mike Sanderson for assistance in the getting this
implementation together.
APPENDIXTop
The rest of the documentation details each of the object methods.
Internal methods are usually preceded with a _
birth_distributionTop
 Title   : rand_birth_distribution
 Usage   : my $randvar = $dist->
            rand_birth_distribution($lambda);
 Function: Returns a random number from a birth process waiting 
           time with a fixed interval
           1.0.  Times are measured from 0=present,1=root;
 Returns : floating point number
 Args    : $lambda ( > 0 )
 References : This is based on code by Mike Sanders in r8s.
              Ross, Stochastic Processes, p. 145 for the density