newmodel-class: Class newmodel

In fedeago/NewWave: Negative binomial model for scRNA-seq

Class newmodel

Description

Objects of this class store all the values needed to work with a negative binomial model, as described in the vignette. They contain all information to fit a model by penalized maximum likelihood or simulate data from a model.

Usage

## S4 method for signature 'newmodel'
show(object)

## S4 method for signature 'newmodel'
numberSamples(x)

## S4 method for signature 'newmodel'
numberFeatures(x)

## S4 method for signature 'newmodel'
numberFactors(x)

## S4 method for signature 'newmodel'
newX(object)

## S4 method for signature 'newmodel'
newV(object)

## S4 method for signature 'newmodel'
newLogMu(object)

## S4 method for signature 'newmodel'
newMu(object)

## S4 method for signature 'newmodel'
newZeta(object)

## S4 method for signature 'newmodel'
newPhi(object)

## S4 method for signature 'newmodel'
newTheta(object)

## S4 method for signature 'newmodel'
newEpsilon_beta(object)

## S4 method for signature 'newmodel'
newEpsilon_gamma(object)

## S4 method for signature 'newmodel'
newEpsilon_W(object)

## S4 method for signature 'newmodel'
newEpsilon_alpha(object)

## S4 method for signature 'newmodel'
newEpsilon_zeta(object)

## S4 method for signature 'newmodel'
newW(object)

## S4 method for signature 'newmodel'
newBeta(object)

## S4 method for signature 'newmodel'
newGamma(object)

## S4 method for signature 'newmodel'
newAlpha(object)

Arguments

object	an object of class newmodel.
x	an object of class newmodel.

Details

For the full description of the model see the model vignette. Internally, the slots are checked so that the matrices are of the appropriate dimensions: in particular, X, O and W need to have n rows, V needs to have J rows, zeta must be of length J.

Value

numberSamples returns the number of samples; numberFeaturesreturns the number of features; numberFactors returns the number of latent factors.

Methods (by generic)

• show: show useful info on the object.

• numberSamples: returns the number of samples.

• numberFeatures: returns the number of features.

• numberFactors: returns the number of latent factors.

• newX: returns the sample-level design matrix for mu.

• newV: returns the gene-level design matrix for mu.

• newLogMu: returns the logarithm of the mean of the non-zero component.

• newMu: returns the mean of the non-zero component.

• newZeta: returns the log of the inverse of the dispersion parameter.

• newPhi: returns the dispersion parameter.

• newTheta: returns the inverse of the dispersion parameter.

• newEpsilon_beta: returns the regularization parameters for beta.

• newEpsilon_gamma: returns the regularization parameters for gamma.

• newEpsilon_W: returns the regularization parameters for W.

• newEpsilon_alpha: returns the regularization parameters for alpha.

• newEpsilon_zeta: returns the regularization parameters for zeta.

• newW: returns the matrix W of inferred sample-level covariates.

• newBeta: returns the matrix beta of inferred parameters.

• newGamma: returns the matrix gamma of inferred parameters.

• newAlpha: returns the matrix alpha of inferred parameters.

Slots

X

matrix. The design matrix containing sample-level covariates, one sample per row.

V

matrix. The design matrix containing gene-level covariates, one gene per row.

X_intercept

logical. TRUE if X contains an intercept.

V_intercept

logical. TRUE if V contains an intercept.

W

matrix. The factors of sample-level latent factors.

beta

matrix or NULL. The coefficients of X in the regression.

gamma

matrix or NULL. The coefficients of V in the regression.

alpha

matrix. The weight of sample-level latent factors.

zeta

numeric. A vector of log of inverse dispersion parameters.

epsilon_beta

nonnegative scalar. Regularization parameter for beta

epsilon_gamma

nonnegative scalar. Regularization parameter for gamma

epsilon_W

nonnegative scalar. Regularization parameter for W

epsilon_alpha

nonnegative scalar. Regularization parameter for alpha

epsilon_zeta

nonnegative scalar. Regularization parameter for zeta

fedeago/NewWave documentation built on March 28, 2022, 5:46 a.m.