Package index • PLNmodels

Top-level fitting functions

Set of functions to fit variants of the Poisson lognormal model

PLNmodels PLNmodels-package: PLNmodels: Poisson Lognormal Models
PLN(): Poisson lognormal model
ZIPLN(): Zero Inflated Poisson lognormal model
PLNLDA(): Poisson lognormal model towards Linear Discriminant Analysis
PLNPCA(): Poisson lognormal model towards Principal Component Analysis
PLNnetwork(): Sparse Poisson lognormal model for network inference
ZIPLNnetwork(): Zero Inflated Sparse Poisson lognormal model for network inference
PLNmixture(): Poisson lognormal mixture model

Poisson lognormal fit

Description of the PLNfit object and methods for its manipulation. Any PLN variant in the package inherits from this class (PLNPCAfit, PLNnetworkfit, PLNLDAfit).

PLNfit: An R6 Class to represent a PLNfit in a standard, general framework
PLNfit_diagonal PLNLDAfit_spherical: An R6 Class to represent a PLNfit in a standard, general framework, with diagonal residual covariance
PLNfit_fixedcov: An R6 Class to represent a PLNfit in a standard, general framework, with fixed (inverse) residual covariance
PLNfit_genpop: An R6 Class to represent a PLNfit with a residual covariance structured by a fixed correlation matrix (e.g. a genetic relationship matrix), motivated by population genetics
PLNfit_spherical: An R6 Class to represent a PLNfit in a standard, general framework, with spherical residual covariance
PLN_param(): Control of a PLN fit
coef(<PLNfit>): Extract model coefficients
vcov(<PLNfit>): Calculate Variance-Covariance Matrix for a fitted PLN() model object
sigma(<PLNfit>): Extract variance-covariance of residuals 'Sigma'
predict(<PLNfit>): Predict counts of a new sample
predict_cond(): Predict counts conditionally
fitted(<PLNfit>): Extracts model fitted values from objects returned by PLN() and its variants
standard_error(): Component-wise standard errors of B
logLik(<PLNfit>): Extract log-likelihood of a fitted PLN model
AIC(<PLNfit>): Akaike Information Criterion for a fitted PLN model
BIC(<PLNfit>): Bayesian Information Criterion for a fitted PLN model

Zero Inflated Poisson lognormal fit

Description of the ZIPLNfit object and methods for its manipulation.

ZIPLNfit: An R6 Class to represent a ZIPLNfit
ZIPLNfit_diagonal: An R6 Class to represent a ZIPLNfit in a standard, general framework, with diagonal residual covariance
ZIPLNfit_fixed: An R6 Class to represent a ZIPLNfit in a standard, general framework, with fixed (inverse) residual covariance
ZIPLNfit_sparse: An R6 Class to represent a ZIPLNfit in a standard, general framework, with sparse inverse residual covariance
ZIPLNfit_spherical: An R6 Class to represent a ZIPLNfit in a standard, general framework, with spherical residual covariance
ZIPLN_param(): Control of a ZIPLN fit
coef(<ZIPLNfit>): Extract model coefficients
sigma(<ZIPLNfit>): Extract variance-covariance of residuals 'Sigma'
predict(<ZIPLNfit>): Predict counts of a new sample
fitted(<ZIPLNfit>): Extracts model fitted values from objects returned by ZIPLN() and its variants
plot(<ZIPLNfit_sparse>): Extract and plot the network (partial correlation, support or inverse covariance) from a ZIPLNfit_sparse object
logLik(<ZIPLNfit>): Extract log-likelihood of a fitted ZIPLN model
AIC(<ZIPLNfit>): Akaike Information Criterion for a fitted ZIPLN model
BIC(<ZIPLNfit>): Bayesian Information Criterion for a fitted ZIPLN model

Linear discriminant analysis via a Poisson lognormal fit

Description of the PLNLDAfit object and methods for its manipulation.

PLNLDAfit: An R6 Class to represent a PLNfit in a LDA framework
PLNLDAfit_diagonal: An R6 Class to represent a PLNfit in a LDA framework with diagonal covariance
PLNfit_diagonal PLNLDAfit_spherical: An R6 Class to represent a PLNfit in a standard, general framework, with diagonal residual covariance
PLNLDA_param(): Control of a PLNLDA fit
plot(<PLNLDAfit>): LDA visualization (individual and/or variable factor map(s)) for a PLNPCAfit object
predict(<PLNLDAfit>): Predict group of new samples
coef(<PLNLDAfit>): Extracts model coefficients from objects returned by PLNLDA()

Poisson Lognormal PCA fit

Description of the PLNPCAfit and PLNPCAfamily objects and methods for their manipulation.

PLNPCAfit: An R6 Class to represent a PLNfit in a PCA framework
PLNPCA_param(): Control of PLNPCA fit
plot(<PLNPCAfit>): PCA visualization (individual and/or variable factor map(s)) for a PLNPCAfit object
PLNPCAfamily: An R6 Class to represent a collection of PLNPCAfit
plot(<PLNPCAfamily>): Display the criteria associated with a collection of PLNPCA fits (a PLNPCAfamily)
getBestModel(): Best model extraction from a collection of models
getModel(): Model extraction from a collection of models

Mixture of Poisson Lognormal fits

Description of the PLNmixturefit and PLNmixturefamily objects and methods for their manipulation.

PLNmixturefit: An R6 Class to represent a PLNfit in a mixture framework
PLNmixture_param(): Control of a PLNmixture fit
plot(<PLNmixturefit>): Mixture visualization of a PLNmixturefit object
PLNmixturefamily: An R6 Class to represent a collection of PLNmixturefit
plot(<PLNmixturefamily>): Display the criteria associated with a collection of PLNmixture fits (a PLNmixturefamily)
coef(<PLNmixturefit>): Extract model coefficients
predict(<PLNmixturefit>): Prediction for a PLNmixturefit object
sigma(<PLNmixturefit>): Extract variance-covariance of residuals 'Sigma'
fitted(<PLNmixturefit>): Extracts model fitted values from objects returned by PLNmixture() and its variants
getBestModel(): Best model extraction from a collection of models
getModel(): Model extraction from a collection of models

Sparse Poisson lognormal fit and network, w/o Zero Inflated component

Description of the (ZI)PLNnetworkfit and (ZI)PLNnetworkfamily objects and methods for their manipulation.

PLNnetworkfit: An R6 Class to represent a PLNfit in a sparse inverse covariance framework
PLNnetwork_param(): Control of PLNnetwork fit
ZIPLNnetwork_param(): Control of ZIPLNnetwork fit
plot(<PLNnetworkfit>): Extract and plot the network (partial correlation, support or inverse covariance) from a PLNnetworkfit object
plot(<ZIPLNfit_sparse>): Extract and plot the network (partial correlation, support or inverse covariance) from a ZIPLNfit_sparse object
Networkfamily: An R6 Class to virtually represent a collection of network fits
ZIPLNnetworkfamily: An R6 Class to represent a collection of ZIPLNnetwork
PLNnetworkfamily: An R6 Class to represent a collection of PLNnetworkfits
plot(<Networkfamily>) plot(<PLNnetworkfamily>) plot(<ZIPLNnetworkfamily>): Display various outputs (goodness-of-fit criteria, robustness, diagnostic) associated with a collection of network fits (either PLNnetworkfamily or ZIPLNnetworkfamily)
getBestModel(): Best model extraction from a collection of models
getModel(): Model extraction from a collection of models
coefficient_path(): Extract the regularization path of a PLNnetwork fit
stability_selection(): Compute the stability path by stability selection
extract_probs(): Extract edge selection frequency in bootstrap subsamples

Other functions and objects

prepare_data(): Prepare data for use in PLN models
compute_offset(): Compute offsets from a count data using one of several normalization schemes
PLNfamily: An R6 Class to represent a collection of PLNfit
plot(<PLNfamily>): Display the criteria associated with a collection of PLN fits (a PLNfamily)
rPLN(): PLN RNG
ICL(): Integrated Classification Likelihood
compute_PLN_starting_point(): Helper function for PLN initialization.
compute_ZIPLN_starting_point(): Helper function for ZIPLN initialization.

Data sets

trichoptera: Trichoptera data set
microcosm: Cow microbiome data set
oaks: Oaks amplicon data set
barents: Barents fish data set
mollusk: Mollusk data set
scRNA: Single cell RNA-seq data