Optimization methods for phylogenetic inference. More...

#include <Bpp/Phyl/OptimizationTools.h>

Collaboration diagram for bpp::OptimizationTools:

Classes
class	ScaleFunction

Public Member Functions
	OptimizationTools ()

virtual	~OptimizationTools ()

Static Public Member Functions
static unsigned int	optimizeNumericalParameters (DiscreteRatesAcrossSitesTreeLikelihood tl, const ParameterList &parameters, OptimizationListener listener=0, unsigned int nstep=1, double tolerance=0.000001, unsigned int tlEvalMax=1000000, OutputStream messageHandler=ApplicationTools::message, OutputStream profiler=ApplicationTools::message, bool reparametrization=false, unsigned int verbose=1, const std::string &optMethodDeriv=OPTIMIZATION_NEWTON, const std::string &optMethodModel=OPTIMIZATION_BRENT) throw (Exception)
	Optimize numerical parameters (branch length, substitution model & rate distribution) of a TreeLikelihood function. More...

static unsigned int	optimizeNumericalParameters2 (DiscreteRatesAcrossSitesTreeLikelihood tl, const ParameterList &parameters, OptimizationListener listener=0, double tolerance=0.000001, unsigned int tlEvalMax=1000000, OutputStream messageHandler=ApplicationTools::message, OutputStream profiler=ApplicationTools::message, bool reparametrization=false, bool useClock=false, unsigned int verbose=1, const std::string &optMethodDeriv=OPTIMIZATION_NEWTON) throw (Exception)
	Optimize numerical parameters (branch length, substitution model & rate distribution) of a TreeLikelihood function. More...

static unsigned int	optimizeBranchLengthsParameters (DiscreteRatesAcrossSitesTreeLikelihood tl, const ParameterList &parameters, OptimizationListener listener=0, double tolerance=0.000001, unsigned int tlEvalMax=1000000, OutputStream messageHandler=ApplicationTools::message, OutputStream profiler=ApplicationTools::message, unsigned int verbose=1, const std::string &optMethodDeriv=OPTIMIZATION_NEWTON) throw (Exception)
	Optimize branch lengths parameters of a TreeLikelihood function. More...

static unsigned int	optimizeNumericalParametersWithGlobalClock (DiscreteRatesAcrossSitesClockTreeLikelihood cl, const ParameterList &parameters, OptimizationListener listener=0, unsigned int nstep=1, double tolerance=0.000001, unsigned int tlEvalMax=1000000, OutputStream messageHandler=ApplicationTools::message, OutputStream profiler=ApplicationTools::message, unsigned int verbose=1, const std::string &optMethodDeriv=OPTIMIZATION_GRADIENT) throw (Exception)
	Optimize numerical parameters assuming a global clock (branch heights, substitution model & rate distribution) of a ClockTreeLikelihood function. More...

static unsigned int	optimizeNumericalParametersWithGlobalClock2 (DiscreteRatesAcrossSitesClockTreeLikelihood cl, const ParameterList &parameters, OptimizationListener listener=0, double tolerance=0.000001, unsigned int tlEvalMax=1000000, OutputStream messageHandler=ApplicationTools::message, OutputStream profiler=ApplicationTools::message, unsigned int verbose=1, const std::string &optMethodDeriv=OPTIMIZATION_GRADIENT) throw (Exception)
	Optimize numerical parameters assuming a global clock (branch heights, substitution model & rate distribution) of a ClockTreeLikelihood function. More...

static unsigned int	optimizeTreeScale (TreeLikelihood tl, double tolerance=0.000001, unsigned int tlEvalMax=1000000, OutputStream messageHandler=ApplicationTools::message, OutputStream *profiler=ApplicationTools::message, unsigned int verbose=1) throw (Exception)
	Optimize the scale of a TreeLikelihood. More...

static NNIHomogeneousTreeLikelihood *	optimizeTreeNNI (NNIHomogeneousTreeLikelihood tl, const ParameterList &parameters, bool optimizeNumFirst=true, double tolBefore=100, double tolDuring=100, unsigned int tlEvalMax=1000000, unsigned int numStep=1, OutputStream messageHandler=ApplicationTools::message, OutputStream *profiler=ApplicationTools::message, bool reparametrization=false, unsigned int verbose=1, const std::string &optMethod=OptimizationTools::OPTIMIZATION_NEWTON, unsigned int nStep=1, const std::string &nniMethod=NNITopologySearch::PHYML) throw (Exception)
	Optimize all parameters from a TreeLikelihood object, including tree topology using Nearest Neighbor Interchanges. More...

static NNIHomogeneousTreeLikelihood *	optimizeTreeNNI2 (NNIHomogeneousTreeLikelihood tl, const ParameterList &parameters, bool optimizeNumFirst=true, double tolBefore=100, double tolDuring=100, unsigned int tlEvalMax=1000000, unsigned int numStep=1, OutputStream messageHandler=ApplicationTools::message, OutputStream *profiler=ApplicationTools::message, bool reparametrization=false, unsigned int verbose=1, const std::string &optMethod=OptimizationTools::OPTIMIZATION_NEWTON, const std::string &nniMethod=NNITopologySearch::PHYML) throw (Exception)
	Optimize all parameters from a TreeLikelihood object, including tree topology using Nearest Neighbor Interchanges. More...

static DRTreeParsimonyScore *	optimizeTreeNNI (DRTreeParsimonyScore *tp, unsigned int verbose=1)
	Optimize tree topology from a DRTreeParsimonyScore using Nearest Neighbor Interchanges. More...

static DistanceMatrix *	estimateDistanceMatrix (DistanceEstimation &estimationMethod, const ParameterList &parametersToIgnore, const std::string &param=DISTANCEMETHOD_INIT, unsigned int verbose=0) throw (Exception)
	Estimate a distance matrix using maximum likelihood. More...

static TreeTemplate< Node > *	buildDistanceTree (DistanceEstimation &estimationMethod, AgglomerativeDistanceMethod &reconstructionMethod, const ParameterList &parametersToIgnore, bool optimizeBrLen=false, const std::string &param=DISTANCEMETHOD_INIT, double tolerance=0.000001, unsigned int tlEvalMax=1000000, OutputStream profiler=0, OutputStream messenger=0, unsigned int verbose=0) throw (Exception)
	Build a tree using a distance method. More...

Static Public Attributes
static std::string	OPTIMIZATION_GRADIENT = "gradient"

static std::string	OPTIMIZATION_NEWTON = "newton"

static std::string	OPTIMIZATION_BRENT = "Brent"

static std::string	OPTIMIZATION_BFGS = "BFGS"

static std::string	DISTANCEMETHOD_INIT = "init"

static std::string	DISTANCEMETHOD_PAIRWISE = "pairwise"

static std::string	DISTANCEMETHOD_ITERATIONS = "iterations"

Detailed Description

Optimization methods for phylogenetic inference.

This class provides optimization methods for phylogenetics. Parameters of the optimization methods are set to work with TreeLikelihood object. Some non trivial parameters are left to the user choice (tolerance, maximum number of function evaluation, output streams).

Definition at line 300 of file OptimizationTools.h.

Constructor & Destructor Documentation

◆ OptimizationTools()

OptimizationTools::OptimizationTools ( )

Definition at line 64 of file OptimizationTools.cpp.

◆ ~OptimizationTools()

OptimizationTools::~OptimizationTools ( )

virtual

Definition at line 66 of file OptimizationTools.cpp.

Member Function Documentation

◆ buildDistanceTree()

TreeTemplate< Node > * OptimizationTools::buildDistanceTree	(	DistanceEstimation &	estimationMethod,
		AgglomerativeDistanceMethod &	reconstructionMethod,
		const ParameterList &	parametersToIgnore,
		bool	optimizeBrLen = `false`,
		const std::string &	param = `DISTANCEMETHOD_INIT`,
		double	tolerance = `0.000001`,
		unsigned int	tlEvalMax = `1000000`,
		OutputStream *	profiler = `0`,
		OutputStream *	messenger = `0`,
		unsigned int	verbose = `0`
	)
throw	(	Exception
	)

static

Build a tree using a distance method.

This method estimate a distance matrix using a DistanceEstimation object, and then builds the phylogenetic tree using a AgglomerativeDistanceMethod object. The main issue here is to estimate non-branch lengths parameters, as substitution model and rate distribution parameters. Three options are provideed here:

DISTANCEMETHOD_INIT (default) keep parameters to there initial value,
DISTANCEMETHOD_PAIRWISE estimated parameters in a pairwise manner, which is standard but not that satisfying...
DISTANCEMETHOD_ITERATIONS uses Ninio et al's iterative algorithm, which uses Maximum Likelihood to estimate these parameters, and then update the distance matrix. Ninio M, Privman E, Pupko T, Friedman N. Phylogeny reconstruction: increasing the accuracy of pairwise distance estimation using Bayesian inference of evolutionary rates. Bioinformatics. 2007 Jan 15;23(2):e136-41.

Parameters

estimationMethod	The distance estimation object to use.
reconstructionMethod	The tree reconstruction object to use.
parametersToIgnore	A list of parameters to ignore while optimizing parameters.
optimizeBrLen	Tell if branch lengths should be optimized together with other parameters. This may lead to more accurate parameter estimation, but is slower.
param	String describing the type of optimization to use.
tolerance	Threshold on likelihood for stopping the iterative procedure. Used only with param=DISTANCEMETHOD_ITERATIONS.
tlEvalMax	Maximum number of likelihood computations to perform when optimizing parameters. Used only with param=DISTANCEMETHOD_ITERATIONS.
profiler	Output stream used by optimizer. Used only with param=DISTANCEMETHOD_ITERATIONS.
messenger	Output stream used by optimizer. Used only with param=DISTANCEMETHOD_ITERATIONS.
verbose	Verbose level.

Definition at line 677 of file OptimizationTools.cpp.

References DISTANCEMETHOD_ITERATIONS, DISTANCEMETHOD_PAIRWISE, bpp::AbstractHomogeneousTreeLikelihood::initialize(), optimizeNumericalParameters(), and bpp::TreeTools::robinsonFouldsDistance().

◆ estimateDistanceMatrix()

DistanceMatrix * OptimizationTools::estimateDistanceMatrix	(	DistanceEstimation &	estimationMethod,
		const ParameterList &	parametersToIgnore,
		const std::string &	param = `DISTANCEMETHOD_INIT`,
		unsigned int	verbose = `0`
	)
throw	(	Exception
	)

static

Estimate a distance matrix using maximum likelihood.

This method estimate a distance matrix using a DistanceEstimation object. The main issue here is to estimate non-branch lengths parameters, as substitution model and rate distribution parameters. Twoe options are provideed here:

DISTANCEMETHOD_INIT (default) keep parameters to there initial value,
DISTANCEMETHOD_PAIRWISE estimated parameters in a pairwise manner, which is standard but not that satisfying...

Parameters

estimationMethod	The distance estimation object to use.
parametersToIgnore	A list of parameters to ignore while optimizing parameters.
param	String describing the type of optimization to use.
verbose	Verbose level.

See also: buildDistanceTree for a procedure to jointly estimate the distance matrix and underlying tree.

Definition at line 647 of file OptimizationTools.cpp.

References DISTANCEMETHOD_INIT, and DISTANCEMETHOD_PAIRWISE.

◆ optimizeBranchLengthsParameters()

unsigned int OptimizationTools::optimizeBranchLengthsParameters	(	DiscreteRatesAcrossSitesTreeLikelihood *	tl,
		const ParameterList &	parameters,
		OptimizationListener *	listener = `0`,
		double	tolerance = `0.000001`,
		unsigned int	tlEvalMax = `1000000`,
		OutputStream *	messageHandler = `ApplicationTools::message`,
		OutputStream *	profiler = `ApplicationTools::message`,
		unsigned int	verbose = `1`,
		const std::string &	optMethodDeriv = `OPTIMIZATION_NEWTON`
	)
throw	(	Exception
	)

static

Optimize branch lengths parameters of a TreeLikelihood function.

Uses Newton's method.

A condition over function values is used as a stop condition for the algorithm.

See also: NewtonBrentMetaOptimizer

Parameters

tl	A pointer toward the TreeLikelihood object to optimize.
parameters	The list of parameters to optimize. The intersection of branch length parameters and the input set will be used. Use tl->getBranchLengthsParameters() in order to estimate all branch length parameters.
listener	A pointer toward an optimization listener, if needed.
tolerance	The tolerance to use in the algorithm.
tlEvalMax	The maximum number of function evaluations.
messageHandler	The massage handler.
profiler	The profiler.
verbose	The verbose level.
optMethodDeriv	Optimization type for derivable parameters (first or second order derivatives).

See also: OPTIMIZATION_NEWTON, OPTIMIZATION_GRADIENT

Exceptions

Exception any exception thrown by the Optimizer.

Definition at line 348 of file OptimizationTools.cpp.

References OPTIMIZATION_BFGS, OPTIMIZATION_GRADIENT, and OPTIMIZATION_NEWTON.

◆ optimizeNumericalParameters()

unsigned int OptimizationTools::optimizeNumericalParameters	(	DiscreteRatesAcrossSitesTreeLikelihood *	tl,
		const ParameterList &	parameters,
		OptimizationListener *	listener = `0`,
		unsigned int	nstep = `1`,
		double	tolerance = `0.000001`,
		unsigned int	tlEvalMax = `1000000`,
		OutputStream *	messageHandler = `ApplicationTools::message`,
		OutputStream *	profiler = `ApplicationTools::message`,
		bool	reparametrization = `false`,
		unsigned int	verbose = `1`,
		const std::string &	optMethodDeriv = `OPTIMIZATION_NEWTON`,
		const std::string &	optMethodModel = `OPTIMIZATION_BRENT`
	)
throw	(	Exception
	)

static

Optimize numerical parameters (branch length, substitution model & rate distribution) of a TreeLikelihood function.

Uses Newton's method for branch length and Brent or BFGS one dimensional method for other parameters.

A condition over function values is used as a stop condition for the algorithm.

See also: BrentOneDimension, BFGSMultiDimensions

Parameters

tl	A pointer toward the TreeLikelihood object to optimize.
parameters	The list of parameters to optimize. Use tl->getIndependentParameters() in order to estimate all parameters.
listener	A pointer toward an optimization listener, if needed.
nstep	The number of progressive steps to perform (see NewtonBrentMetaOptimizer). 1 means full precision from start.
tolerance	The tolerance to use in the algorithm.
tlEvalMax	The maximum number of function evaluations.
messageHandler	The massage handler.
profiler	The profiler.
reparametrization	Tell if parameters should be transformed in order to remove constraints. This can improve optimization, but is a bit slower.
verbose	The verbose level.
optMethodDeriv	Optimization type for derivable parameters (first or second order derivatives).

See also: OPTIMIZATION_NEWTON, OPTIMIZATION_GRADIENT

Parameters

optMethodModel Optimization type for model parameters (Brent or BFGS).

See also: OPTIMIZATION_BRENT, OPTIMIZATION_BFGS

Exceptions

Exception any exception thrown by the Optimizer.

Definition at line 150 of file OptimizationTools.cpp.

References OPTIMIZATION_BFGS, OPTIMIZATION_BRENT, OPTIMIZATION_GRADIENT, and OPTIMIZATION_NEWTON.

Referenced by buildDistanceTree(), bpp::PhylogeneticsApplicationTools::optimizeParameters(), optimizeTreeNNI(), and bpp::NNITopologyListener::topologyChangeSuccessful().

◆ optimizeNumericalParameters2()

unsigned int OptimizationTools::optimizeNumericalParameters2	(	DiscreteRatesAcrossSitesTreeLikelihood *	tl,
		const ParameterList &	parameters,
		OptimizationListener *	listener = `0`,
		double	tolerance = `0.000001`,
		unsigned int	tlEvalMax = `1000000`,
		OutputStream *	messageHandler = `ApplicationTools::message`,
		OutputStream *	profiler = `ApplicationTools::message`,
		bool	reparametrization = `false`,
		bool	useClock = `false`,
		unsigned int	verbose = `1`,
		const std::string &	optMethodDeriv = `OPTIMIZATION_NEWTON`
	)
throw	(	Exception
	)

static

Optimize numerical parameters (branch length, substitution model & rate distribution) of a TreeLikelihood function.

Uses Newton's method for all parameters, branch length derivatives are computed analytically, derivatives for other parameters numerically.

See also: PseudoNewtonOptimizer

Parameters

tl	A pointer toward the TreeLikelihood object to optimize.
parameters	The list of parameters to optimize. Use tl->getIndependentParameters() in order to estimate all parameters.
listener	A pointer toward an optimization listener, if needed.
tolerance	The tolerance to use in the algorithm.
tlEvalMax	The maximum number of function evaluations.
messageHandler	The massage handler.
profiler	The profiler.
reparametrization	Tell if parameters should be transformed in order to remove constraints. This can improve optimization, but is a bit slower.
useClock	Tell if branch lengths have to be optimized under a global molecular clock constraint.
verbose	The verbose level.
optMethodDeriv	Optimization type for derivable parameters (first or second order derivatives).

See also: OPTIMIZATION_NEWTON, OPTIMIZATION_GRADIENT

Exceptions

Exception any exception thrown by the Optimizer.

Definition at line 255 of file OptimizationTools.cpp.

References bpp::GlobalClockTreeLikelihoodFunctionWrapper::getHeightParameters(), OPTIMIZATION_BFGS, OPTIMIZATION_GRADIENT, and OPTIMIZATION_NEWTON.

Referenced by bpp::PhylogeneticsApplicationTools::optimizeParameters(), optimizeTreeNNI2(), and bpp::NNITopologyListener2::topologyChangeSuccessful().

◆ optimizeNumericalParametersWithGlobalClock()

unsigned int OptimizationTools::optimizeNumericalParametersWithGlobalClock	(	DiscreteRatesAcrossSitesClockTreeLikelihood *	cl,
		const ParameterList &	parameters,
		OptimizationListener *	listener = `0`,
		unsigned int	nstep = `1`,
		double	tolerance = `0.000001`,
		unsigned int	tlEvalMax = `1000000`,
		OutputStream *	messageHandler = `ApplicationTools::message`,
		OutputStream *	profiler = `ApplicationTools::message`,
		unsigned int	verbose = `1`,
		const std::string &	optMethodDeriv = `OPTIMIZATION_GRADIENT`
	)
throw	(	Exception
	)

static

Optimize numerical parameters assuming a global clock (branch heights, substitution model & rate distribution) of a ClockTreeLikelihood function.

Uses Newton or conjugate gradient method for branch length and Brent's one dimensional method for other parameters (NewtonBrentMetaOptimizer). Derivatives are computed analytically.

A condition over function values is used as a stop condition for the algorithm.

See also: NewtonBrentMetaOptimizer

Parameters

cl	A pointer toward the ClockTreeLikelihood object to optimize.
parameters	The list of parameters to optimize. Use cl->getIndependentParameters() in order to estimate all parameters.
listener	A pointer toward an optimization listener, if needed.
nstep	The number of progressive steps to perform (see NewtonBrentMetaOptimizer). 1 means full precision from start.
tolerance	The tolerance to use in the algorithm.
tlEvalMax	The maximum number of function evaluations.
messageHandler	The massage handler.
profiler	The profiler.
verbose	The verbose level.
optMethodDeriv	Optimization type for derivable parameters (first or second order derivatives).

See also: OPTIMIZATION_NEWTON, OPTIMIZATION_GRADIENT

Exceptions

Exception any exception thrown by the Optimizer.

Deprecated:: See optimizeNumericalParameters2 as a more general replacement.

Definition at line 406 of file OptimizationTools.cpp.

References OPTIMIZATION_GRADIENT, and OPTIMIZATION_NEWTON.

Referenced by bpp::PhylogeneticsApplicationTools::optimizeParameters().

◆ optimizeNumericalParametersWithGlobalClock2()

unsigned int OptimizationTools::optimizeNumericalParametersWithGlobalClock2	(	DiscreteRatesAcrossSitesClockTreeLikelihood *	cl,
		const ParameterList &	parameters,
		OptimizationListener *	listener = `0`,
		double	tolerance = `0.000001`,
		unsigned int	tlEvalMax = `1000000`,
		OutputStream *	messageHandler = `ApplicationTools::message`,
		OutputStream *	profiler = `ApplicationTools::message`,
		unsigned int	verbose = `1`,
		const std::string &	optMethodDeriv = `OPTIMIZATION_GRADIENT`
	)
throw	(	Exception
	)

static

Optimize numerical parameters assuming a global clock (branch heights, substitution model & rate distribution) of a ClockTreeLikelihood function.

Uses Newton or conjugate gradient method for all parameters, branch length derivatives are computed analytically, derivatives for other parameters numerically.

See also: PseudoNewtonOptimizer

Parameters

cl	A pointer toward the ClockTreeLikelihood object to optimize.
parameters	The list of parameters to optimize. Use cl->getIndependentParameters() in order to estimate all parameters.
listener	A pointer toward an optimization listener, if needed.
tolerance	The tolerance to use in the algorithm.
tlEvalMax	The maximum number of function evaluations.
messageHandler	The massage handler.
profiler	The profiler.
verbose	The verbose level.
optMethodDeriv	Optimization type for derivable parameters (first or second order derivatives).

See also: OPTIMIZATION_NEWTON, OPTIMIZATION_GRADIENT

Exceptions

Exception any exception thrown by the Optimizer.

Deprecated:: See optimizeNumericalParameters2 as a more general replacement.

Definition at line 476 of file OptimizationTools.cpp.

References OPTIMIZATION_GRADIENT, and OPTIMIZATION_NEWTON.

Referenced by bpp::PhylogeneticsApplicationTools::optimizeParameters().

◆ optimizeTreeNNI() [1/2]

NNIHomogeneousTreeLikelihood * OptimizationTools::optimizeTreeNNI	(	NNIHomogeneousTreeLikelihood *	tl,
		const ParameterList &	parameters,
		bool	optimizeNumFirst = `true`,
		double	tolBefore = `100`,
		double	tolDuring = `100`,
		unsigned int	tlEvalMax = `1000000`,
		unsigned int	numStep = `1`,
		OutputStream *	messageHandler = `ApplicationTools::message`,
		OutputStream *	profiler = `ApplicationTools::message`,
		bool	reparametrization = `false`,
		unsigned int	verbose = `1`,
		const std::string &	optMethod = `OptimizationTools::OPTIMIZATION_NEWTON`,
		unsigned int	nStep = `1`,
		const std::string &	nniMethod = `NNITopologySearch::PHYML`
	)
throw	(	Exception
	)

static

Optimize all parameters from a TreeLikelihood object, including tree topology using Nearest Neighbor Interchanges.

This function takes as input a TreeLikelihood object implementing the NNISearchable interface.

Details: A NNITopologySearch object is instanciated and is associated an additional TopologyListener. This listener is used to re-estimate numerical parameters after one or several topology change. By default, the PHYML option is used for the NNITopologySearch object, and numerical parameters are re-estimated every 4 NNI runs (as in the phyml software).

The optimizeNumericalParameters method is used for estimating numerical parameters. The tolerance passed to this function is specified as input parameters. They are generally very high to avoid local optima.

Parameters

tl	A pointer toward the TreeLikelihood object to optimize.
parameters	The list of parameters to optimize. Use tl->getIndependentParameters() in order to estimate all parameters.
optimizeNumFirst	Tell if we must optimize numerical parameters before searching topology.
tolBefore	The tolerance to use when estimating numerical parameters before topology search (if optimizeNumFirst is set to 'true').
tolDuring	The tolerance to use when estimating numerical parameters during the topology search.
tlEvalMax	The maximum number of function evaluations.
numStep	Number of NNI rounds before re-estimating numerical parameters.
messageHandler	The massage handler.
profiler	The profiler.
reparametrization	Tell if parameters should be transformed in order to remove constraints. This can improve optimization, but is a bit slower.
verbose	The verbose level.
optMethod	Option passed to optimizeNumericalParameters.
nStep	Option passed to optimizeNumericalParameters.
nniMethod	NNI algorithm to use.

Returns: A pointer toward the final likelihood object. This pointer may be the same as passed in argument (tl), but in some cases the algorithm clone this object. We may change this bahavior in the future... You hence should write something like
tl = OptimizationTools::optimizeTreeNNI(tl, ...);

Exceptions

Exception any exception thrown by the optimizer.

Definition at line 564 of file OptimizationTools.cpp.

References bpp::NNITopologySearch::addTopologyListener(), bpp::NNITopologySearch::getSearchableObject(), optimizeNumericalParameters(), bpp::NNITopologySearch::search(), and bpp::NNITopologyListener::setNumericalOptimizationCounter().

Referenced by bpp::TreeTools::MRP(), bpp::TreeTools::MRPMultilabel(), and bpp::PhylogeneticsApplicationTools::optimizeParameters().

◆ optimizeTreeNNI() [2/2]

DRTreeParsimonyScore * OptimizationTools::optimizeTreeNNI	(	DRTreeParsimonyScore *	tp,
		unsigned int	verbose = `1`
	)

static

Optimize tree topology from a DRTreeParsimonyScore using Nearest Neighbor Interchanges.

Parameters

tp	A pointer toward the DRTreeParsimonyScore object to optimize.
verbose	The verbose level.

Returns: A pointer toward the final parsimony score object. This pointer may be the same as passed in argument (tl), but in some cases the algorithm clone this object. We may change this bahavior in the future... You hence should write something like
tp = OptimizationTools::optimizeTreeNNI(tp, ...);

Definition at line 629 of file OptimizationTools.cpp.

References bpp::NNITopologySearch::getSearchableObject(), bpp::NNITopologySearch::PHYML, and bpp::NNITopologySearch::search().

◆ optimizeTreeNNI2()

NNIHomogeneousTreeLikelihood * OptimizationTools::optimizeTreeNNI2	(	NNIHomogeneousTreeLikelihood *	tl,
		const ParameterList &	parameters,
		bool	optimizeNumFirst = `true`,
		double	tolBefore = `100`,
		double	tolDuring = `100`,
		unsigned int	tlEvalMax = `1000000`,
		unsigned int	numStep = `1`,
		OutputStream *	messageHandler = `ApplicationTools::message`,
		OutputStream *	profiler = `ApplicationTools::message`,
		bool	reparametrization = `false`,
		unsigned int	verbose = `1`,
		const std::string &	optMethod = `OptimizationTools::OPTIMIZATION_NEWTON`,
		const std::string &	nniMethod = `NNITopologySearch::PHYML`
	)
throw	(	Exception
	)

static

Optimize all parameters from a TreeLikelihood object, including tree topology using Nearest Neighbor Interchanges.

This function takes as input a TreeLikelihood object implementing the NNISearchable interface.

Details: A NNITopologySearch object is instanciated and is associated an additional TopologyListener. This listener is used to re-estimate numerical parameters after one or several topology change. By default, the PHYML option is used for the NNITopologySearch object, and numerical parameters are re-estimated every 4 NNI runs (as in the phyml software).

The optimizeNumericalParameters2 method is used for estimating numerical parameters. The tolerance passed to this function is specified as input parameters. They are generally very high to avoid local optima.

Parameters

tl	A pointer toward the TreeLikelihood object to optimize.
parameters	The list of parameters to optimize. Use tl->getIndependentParameters() in order to estimate all parameters.
optimizeNumFirst	Tell if we must optimize numerical parameters before searching topology.
tolBefore	The tolerance to use when estimating numerical parameters before topology search (if optimizeNumFirst is set to 'true').
tolDuring	The tolerance to use when estimating numerical parameters during the topology search.
tlEvalMax	The maximum number of function evaluations.
numStep	Number of NNI rounds before re-estimating numerical parameters.
messageHandler	The massage handler.
profiler	The profiler.
reparametrization	Tell if parameters should be transformed in order to remove constraints. This can improve optimization, but is a bit slower.
verbose	The verbose level.
optMethod	Option passed to optimizeNumericalParameters2.
nniMethod	NNI algorithm to use.

Returns: A pointer toward the final likelihood object. This pointer may be the same as passed in argument (tl), but in some cases the algorithm clone this object. We may change this bahavior in the future... You hence should write something like
tl = OptimizationTools::optimizeTreeNNI2(tl, ...);

Exceptions

Exception any exception thrown by the optimizer.

Definition at line 597 of file OptimizationTools.cpp.

References bpp::NNITopologySearch::addTopologyListener(), bpp::NNITopologySearch::getSearchableObject(), optimizeNumericalParameters2(), bpp::NNITopologySearch::search(), and bpp::NNITopologyListener2::setNumericalOptimizationCounter().

Referenced by bpp::PhylogeneticsApplicationTools::optimizeParameters().

◆ optimizeTreeScale()

unsigned int OptimizationTools::optimizeTreeScale	(	TreeLikelihood *	tl,
		double	tolerance = `0.000001`,
		unsigned int	tlEvalMax = `1000000`,
		OutputStream *	messageHandler = `ApplicationTools::message`,
		OutputStream *	profiler = `ApplicationTools::message`,
		unsigned int	verbose = `1`
	)
throw	(	Exception
	)

static

Optimize the scale of a TreeLikelihood.

This method only works on branch lengths parameters. It multiply all branch length by a factor 'x' which is optimized using Brent's algorithm in one dimension. This method may be usefull for scaling a tree whose branch lengths come from the Neighbor-Joining algorithm for instance.

Practically, and contrarily to what one may expect, this does not speed up the optimization!

A condition over parameters is used as a stop condition for the algorithm.

Parameters

tl	A pointer toward the TreeLikelihood object to optimize.
tolerance	The tolerance to use in the algorithm.
tlEvalMax	The maximum number of function evaluations.
messageHandler	The massage handler.
profiler	The profiler.
verbose	The verbose level.

Exceptions

Exception any exception thrown by the optimizer.

Definition at line 121 of file OptimizationTools.cpp.

References bpp::OptimizationTools::ScaleFunction::getParameters().

Referenced by bpp::PhylogeneticsApplicationTools::optimizeParameters().

Member Data Documentation

◆ DISTANCEMETHOD_INIT

std::string OptimizationTools::DISTANCEMETHOD_INIT = "init"

static

Definition at line 763 of file OptimizationTools.h.

Referenced by estimateDistanceMatrix().

◆ DISTANCEMETHOD_ITERATIONS

std::string OptimizationTools::DISTANCEMETHOD_ITERATIONS = "iterations"

static

Definition at line 765 of file OptimizationTools.h.

Referenced by buildDistanceTree().

◆ DISTANCEMETHOD_PAIRWISE

std::string OptimizationTools::DISTANCEMETHOD_PAIRWISE = "pairwise"

static

Definition at line 764 of file OptimizationTools.h.

Referenced by buildDistanceTree(), and estimateDistanceMatrix().

◆ OPTIMIZATION_BFGS

std::string OptimizationTools::OPTIMIZATION_BFGS = "BFGS"

static

Definition at line 310 of file OptimizationTools.h.

Referenced by optimizeBranchLengthsParameters(), optimizeNumericalParameters(), optimizeNumericalParameters2(), and bpp::PhylogeneticsApplicationTools::optimizeParameters().

◆ OPTIMIZATION_BRENT

std::string OptimizationTools::OPTIMIZATION_BRENT = "Brent"

static

Definition at line 309 of file OptimizationTools.h.

Referenced by optimizeNumericalParameters(), and bpp::PhylogeneticsApplicationTools::optimizeParameters().

◆ OPTIMIZATION_GRADIENT

std::string OptimizationTools::OPTIMIZATION_GRADIENT = "gradient"

static

Definition at line 307 of file OptimizationTools.h.

Referenced by optimizeBranchLengthsParameters(), optimizeNumericalParameters(), optimizeNumericalParameters2(), optimizeNumericalParametersWithGlobalClock(), optimizeNumericalParametersWithGlobalClock2(), and bpp::PhylogeneticsApplicationTools::optimizeParameters().

◆ OPTIMIZATION_NEWTON

std::string OptimizationTools::OPTIMIZATION_NEWTON = "newton"

static

Definition at line 308 of file OptimizationTools.h.

Referenced by optimizeBranchLengthsParameters(), optimizeNumericalParameters(), optimizeNumericalParameters2(), optimizeNumericalParametersWithGlobalClock(), optimizeNumericalParametersWithGlobalClock2(), and bpp::PhylogeneticsApplicationTools::optimizeParameters().

The documentation for this class was generated from the following files:

Bpp/Phyl/OptimizationTools.h
Bpp/Phyl/OptimizationTools.cpp

Classes

Public Member Functions

Static Public Member Functions

Static Public Attributes

Detailed Description

Constructor & Destructor Documentation

◆ OptimizationTools()

◆ ~OptimizationTools()

Member Function Documentation

◆ buildDistanceTree()

◆ estimateDistanceMatrix()

◆ optimizeBranchLengthsParameters()

◆ optimizeNumericalParameters()

◆ optimizeNumericalParameters2()

◆ optimizeNumericalParametersWithGlobalClock()

◆ optimizeNumericalParametersWithGlobalClock2()

◆ optimizeTreeNNI() [1/2]

◆ optimizeTreeNNI() [2/2]

◆ optimizeTreeNNI2()

◆ optimizeTreeScale()

Member Data Documentation

◆ DISTANCEMETHOD_INIT

◆ DISTANCEMETHOD_ITERATIONS

◆ DISTANCEMETHOD_PAIRWISE

◆ OPTIMIZATION_BFGS

◆ OPTIMIZATION_BRENT

◆ OPTIMIZATION_GRADIENT

◆ OPTIMIZATION_NEWTON