constrained.loglikelihood

Vector of optimized parameters with length p+q, where p is the number of covariates for count model (e.g., beta-binomial), q is the number of covariates for zero model. The first p elements are betas which are the effects/coefficients for the count model. The last q elements are etas which are the effects/coefficients for the zero model.

para

The design matrix (n by p, p is the number of covariates) for the count model (e.g., beta-binomial), and intercept is included.

Vector of counts corresponding to an OTU, with length n.

Y.col

Vector of coefficients in the polynomial mean-overdispersion relationship in constrained approach.

coeff

Vector of library size with length n.

The design matrix (n by q) for the zero model, and intercept is included.

ziMatrix

The objective function is the negative of log likelihood function.

Microbiome count data (Operational Taxonomic Unit, OTUs) is usually overdispersed and has excessive zero counts. The 'ZIBBSeqDiscovery' assumes a zero-inflated beta-binomial model for the distribution of the count data, and employes link functions to adjust interested covariates. To fit the model, two approaches are proposed (i) a free approach which treats the overdispersion parameters for OTUs as independent, and (ii) a constrained approach which proposes a mean-overdispersion relationship to the count data. This package can be used to test the association between the composition of the microbiome counts and the interested covariates.

constrained.loglikelihood: Define the objective function in optimization procedure for estimating parameters with constrained approach

Description

Usage

Arguments