pal.eval
Interface UnconstrainedLikelihoodModel.External

All Superinterfaces:

java.io.Serializable

Enclosing interface:

UnconstrainedLikelihoodModel

public static interface UnconstrainedLikelihoodModel.External
extends java.io.Serializable

The External calculator does not maintain any state and is approapriate for calculation where a store is provided

Method Summary

void	calculateExtended(double distance, PatternInfo centerPattern, ConditionalProbabilityStore leftConditionalProbabilities, ConditionalProbabilityStore rightConditionalProbabilities, ConditionalProbabilityStore resultStore)
void	calculateFlat(PatternInfo centerPattern, ConditionalProbabilityStore leftConditionalProbabilities, ConditionalProbabilityStore rightConditionalProbabilities, ConditionalProbabilityStore resultStore)
double	calculateLogLikelihood(double distance, PatternInfo centerPattern, ConditionalProbabilityStore leftFlatConditionalProbabilities, ConditionalProbabilityStore rightFlatConditionalProbabilities, ConditionalProbabilityStore tempStore) Calculate the likelihood given two sub trees (left, right) and their flat (unextend) likeihood probabilities
double	calculateLogLikelihood(PatternInfo centerPattern, ConditionalProbabilityStore leftConditionalProbabilities, ConditionalProbabilityStore rightConditionalProbabilities) Calculate the likelihood given two sub trees (left, right) and their extended likeihood probabilities
double	calculateLogLikelihoodSingle(int[] patternWeights, int numberOfPatterns, ConditionalProbabilityStore conditionalProbabilityStore) Calculate the likelihood given the conditional probabilites at the root
void	calculateSingleExtendedDirect(double distance, int numberOfPatterns, ConditionalProbabilityStore conditionalProbabilities) Extend the conditionals back in time by some distance
void	calculateSingleExtendedIndirect(double distance, int numberOfPatterns, ConditionalProbabilityStore baseConditionalProbabilities, ConditionalProbabilityStore resultConditionalProbabilities) Extend the conditionals back in time by some distance
SiteDetails	calculateSiteDetailsRooted(PatternInfo centerPattern, ConditionalProbabilityStore leftConditionalProbabilitiesStore, ConditionalProbabilityStore rightConditionalProbabilitiesStore) Calculate the conditional probabilities of each pattern for each category
SiteDetails	calculateSiteDetailsUnrooted(double distance, PatternInfo centerPattern, ConditionalProbabilityStore leftConditionalProbabilitiesStore, ConditionalProbabilityStore rightConditionalProbabilitiesStore, ConditionalProbabilityStore tempStore) Calculate the conditional probabilities of each pattern for each category

Method Detail

calculateFlat

void calculateFlat(PatternInfo centerPattern,
                   ConditionalProbabilityStore leftConditionalProbabilities,
                   ConditionalProbabilityStore rightConditionalProbabilities,
                   ConditionalProbabilityStore resultStore)

Parameters:

centerPattern - the pattern information

leftConditionalProbabilities - Implementations must not overwrite or change

rightConditionalProbabilities - Implementations must not overwrite or change

resultStore - Where to stick the created categoryPatternState information

calculateExtended

void calculateExtended(double distance,
                       PatternInfo centerPattern,
                       ConditionalProbabilityStore leftConditionalProbabilities,
                       ConditionalProbabilityStore rightConditionalProbabilities,
                       ConditionalProbabilityStore resultStore)

Parameters:

distance - the evolutionary distance

centerPattern - the pattern information

leftConditionalProbabilities - Implementations must not overwrite or change

rightConditionalProbabilities - Implementations must not overwrite or change

resultStore - Where to stick the created categoryPatternState information

calculateSingleExtendedDirect

void calculateSingleExtendedDirect(double distance,
                                   int numberOfPatterns,
                                   ConditionalProbabilityStore conditionalProbabilities)

Extend the conditionals back in time by some distance

Parameters:

distance - The evolutionary distance to extend by

numberOfPatterns - the number of patterns

conditionalProbabilities - The probabilities to extend

calculateSingleExtendedIndirect

void calculateSingleExtendedIndirect(double distance,
                                     int numberOfPatterns,
                                     ConditionalProbabilityStore baseConditionalProbabilities,
                                     ConditionalProbabilityStore resultConditionalProbabilities)

Extend the conditionals back in time by some distance

Parameters:

distance - The evolutionary distance to extend by

numberOfPatterns - the number of patterns

baseConditionalProbabilities - The probabilities to extend

resultConditionalProbabilities - The probabilities to extend

calculateLogLikelihood

double calculateLogLikelihood(double distance,
                              PatternInfo centerPattern,
                              ConditionalProbabilityStore leftFlatConditionalProbabilities,
                              ConditionalProbabilityStore rightFlatConditionalProbabilities,
                              ConditionalProbabilityStore tempStore)

Calculate the likelihood given two sub trees (left, right) and their flat (unextend) likeihood probabilities

Parameters:

distance - The evolutionary distance

centerPattern - the pattern information

leftFlatConditionalProbabilities - The left conditional probabilities (unextended)

rightFlatConditionalProbabilities - The right conditional probabilities (unextended)

tempStore - may be used internally to calculate likelihood

Returns:

the log likelihood

calculateLogLikelihood

double calculateLogLikelihood(PatternInfo centerPattern,
                              ConditionalProbabilityStore leftConditionalProbabilities,
                              ConditionalProbabilityStore rightConditionalProbabilities)

Calculate the likelihood given two sub trees (left, right) and their extended likeihood probabilities

Parameters:

centerPattern - the pattern information

leftConditionalProbabilities - The left conditional probabilities

rightConditionalProbabilities - The right conditional probabilities

Returns:

the Log likelihood

calculateLogLikelihoodSingle

double calculateLogLikelihoodSingle(int[] patternWeights,
                                    int numberOfPatterns,
                                    ConditionalProbabilityStore conditionalProbabilityStore)

Calculate the likelihood given the conditional probabilites at the root

Parameters:

patternWeights - the weights of each pattern

numberOfPatterns - the number of patterns

Returns:

the Log likelihood

calculateSiteDetailsRooted

SiteDetails calculateSiteDetailsRooted(PatternInfo centerPattern,
                                       ConditionalProbabilityStore leftConditionalProbabilitiesStore,
                                       ConditionalProbabilityStore rightConditionalProbabilitiesStore)

Calculate the conditional probabilities of each pattern for each category

Parameters:

centerPattern - the pattern information

leftConditionalProbabilitiesStore - The left conditional probabilities

rightConditionalProbabilitiesStore - The right conditional probabilities

calculateSiteDetailsUnrooted

SiteDetails calculateSiteDetailsUnrooted(double distance,
                                         PatternInfo centerPattern,
                                         ConditionalProbabilityStore leftConditionalProbabilitiesStore,
                                         ConditionalProbabilityStore rightConditionalProbabilitiesStore,
                                         ConditionalProbabilityStore tempStore)

Calculate the conditional probabilities of each pattern for each category

Parameters:

distance - The distance between the two nodes

centerPattern - the pattern information

leftConditionalProbabilitiesStore - The left conditional probabilities

rightConditionalProbabilitiesStore - The right conditional probabilities

tempStore - after call will hold a matrix of values in the form [cat][pattern], where [cat][pattern] represents the site probability under a particular category/class, *not* multiplied by the category probability or pattern weights