public class DifferentiableStatisticalModelFactory extends Object
Constructor and Description |
---|
DifferentiableStatisticalModelFactory() |
Modifier and Type | Method and Description |
---|---|
static HomogeneousMMDiffSM |
createHomogeneousMarkovModel(AlphabetContainer con,
double ess,
int order,
int priorLength)
This method returns a homogeneous Markov model with user-specified order.
|
static BayesianNetworkDiffSM |
createInhomogeneousMarkovModel(AlphabetContainer con,
int length,
double ess,
int order)
This method returns a inhomogeneous Markov model (IMM) with user-specified order.
|
static MarkovRandomFieldDiffSM |
createMarkovRandomField(AlphabetContainer con,
int length,
String constraintType)
This method allows to create a
MarkovRandomFieldDiffSM of the specified length and with the given constraint type. |
static MixtureDiffSM |
createMixtureModel(DifferentiableStatisticalModel[] models)
This method allows to create a
MixtureDiffSM that models a mixture of individual component DifferentiableStatisticalModel s. |
static BayesianNetworkDiffSM |
createPWM(AlphabetContainer con,
int length,
double ess)
This method returns a position weight matrix (PWM).
|
static StrandDiffSM |
createStrandModel(DifferentiableStatisticalModel model)
This method allows to create a
StrandDiffSM that allows to score binding sites on both strand of DNA. |
static ExtendedZOOPSDiffSM |
createZOOPS(int length,
DifferentiableStatisticalModel motif,
HomogeneousDiffSM bg)
This method allows to create a "zero or one occurrence per sequence" (ZOOPS) model that allows to discover binding sites in a
DataSet . |
public DifferentiableStatisticalModelFactory()
public static BayesianNetworkDiffSM createPWM(AlphabetContainer con, int length, double ess) throws Exception
con
- the AlphabetContainer
of the PWMlength
- the length of the PWM, i.e., the length of the sequences that can be modeledess
- the equivalent sample size (ess) of the PWM for the BDeu prior on its parametersException
- if the model can not be created correctlypublic static BayesianNetworkDiffSM createInhomogeneousMarkovModel(AlphabetContainer con, int length, double ess, int order) throws Exception
con
- the AlphabetContainer
of the IMMlength
- the length of the IMM, i.e., the length of the sequences that can be modeledess
- the equivalent sample size (ess) of the IMM for the BDeu prior on its parametersorder
- the order of the IMM, i.e., the number of directly preceding random variables (=positions) that might have an influence on the probability of outcome of a random variable (=position)Exception
- if the model can not be created correctlypublic static HomogeneousMMDiffSM createHomogeneousMarkovModel(AlphabetContainer con, double ess, int order, int priorLength)
con
- the AlphabetContainer
of the modeless
- the equivalent sample size (ess) of the class of this model, used for the BDeu prior on its parameters in conjunction with priorLength
order
- the order of the model, i.e., the number of directly preceding random variables (=positions) that might have an influence on the probability of outcome of a random variable (=position)priorLength
- the a-priorily expected length of input sequences, is multiplied by ess
before computing hyper-parameterspublic static StrandDiffSM createStrandModel(DifferentiableStatisticalModel model) throws CloneNotSupportedException, WrongAlphabetException
StrandDiffSM
that allows to score binding sites on both strand of DNA.
The strand preferences is learned together with the supplied DifferentiableStatisticalModel
.model
- the internally used modelStrandDiffSM
CloneNotSupportedException
- if the supplied DifferentiableStatisticalModel
could not be clonedWrongAlphabetException
- if the AlphabetContainer
of the supplied DifferentiableStatisticalModel
is not AlphabetContainer.isReverseComplementable()
public static MixtureDiffSM createMixtureModel(DifferentiableStatisticalModel[] models) throws CloneNotSupportedException
MixtureDiffSM
that models a mixture of individual component DifferentiableStatisticalModel
s.models
- the internally used modelsMixtureDiffSM
CloneNotSupportedException
- if the supplied DifferentiableStatisticalModel
s could not be clonedpublic static ExtendedZOOPSDiffSM createZOOPS(int length, DifferentiableStatisticalModel motif, HomogeneousDiffSM bg) throws Exception
DataSet
.public static MarkovRandomFieldDiffSM createMarkovRandomField(AlphabetContainer con, int length, String constraintType)
MarkovRandomFieldDiffSM
of the specified length and with the given constraint type.con
- the AlphabetContainer
of the MarkovRandomFieldDiffSM
length
- the length of the Sequence
s the MarkovRandomFieldDiffSM
can handleconstraintType
- the constraint type, see ConstraintManager.extract(int, String)
MarkovRandomFieldDiffSM
ConstraintManager.extract(int, String)