CNVassoc: Association analysis between a CNV and phenotype

Description

This function performs an association analysis between a CNV and a dependent variable (phenotype) using a latent class model that incorporates the uncertainty arising from calling procedure. The phenotype may be quantitative or categorical. In the second case (e.g. case-control studies) this variable must be coded as 1 (for cases) and 0 (for controls). The association can be adjusted for other covariates (e.g. clinical covariates, stratification, ...)

Usage

CNVassoc(formula, data, subset, na.action, model = "multiplicative", family = "binomial", tol = 1e-06, max.iter = 30, emsteps = 0, verbose = FALSE, coef.start, sigma.start, alpha.start=1)

Arguments

formula

an object of class "formula" (or one that can be coerced to that class): a symbolic description of the model to be fitted. Right side of ~ should have an object of class 'cnv'.

data

an optional data frame, list or environment (or object coercible by 'as.data.frame' to a data frame) containing the variables in the model. If not found in 'data', the variables are taken from 'environment(formula)'.

subset

an optional vector specifying a subset of observations to be used in the fitting process.

na.action

a function which indicates what should happen when the data contain 'NA's. The default is set by the 'na.action' setting of 'options', and is 'na.fail' if that is unset. The 'factory-fresh' default is 'na.omit'. Another possible value is 'NULL',

model

Genetic model to be tested. Possible values are "multiplicative" (model free, e.g. co-dominant) or "additive", partial matching allowed. Default value is "multiplicative".

family

a description of the error distribution and link function to be used in the model. This must be a character string naming a family function. Possible values are "binomial", "gaussian", "poisson" or "weibull". Default value is "binomial

tol

Tolerance for convergence in fitting model. Default value is 1e-06.

max.iter

Maximum number of iterations in fitting model. Default value is 30.

emsteps

Number of iterations using Expectation Maximization (EM) alghorithm to set initial values before using Newton-Rapson (NR) in fitting model. Default value is zero, that means that EM step is not performed

verbose

logical. If TRUE parameter values for each iteration are shown in the console. Default value is FALSE

coef.start

initial values for coefficients in NR procedure

sigma.start

initial values for scale parameter (only for "gaussian") in NR procedure

alpha.start

initial values for shape parameter (only for "weibull") in NR procedure

Value

An object of class 'CNVassoc'.
'print' returns model parameter estimates
'summary' returns a summary table similar to summary.glm
'anova' performs a Likelihood Ratio Test comparing two nested models fitted using CNVassoc
'logLik' returns the log-likelihood of a model fitted using CNVassoc See examples for further illustration about all previous issues.

References

Gonzalez JR, Subirana I, Escaramis G, Peraza S, Caceres A, Estivill X and Armengol L. Accounting for uncertainty when assessing association between copy number and disease: a latent class model. BMC Bioinformatics, 2009;10:172.

Examples

Run this code

data(dataMLPA)
CNV  <-  cnv(x  =  dataMLPA$Gene2,  threshold.0  =  0.01,  mix.method  =  "mixdist")
modmul  <-  CNVassoc(casco  ~  CNV,  data  =  dataMLPA,  model  =  "mul")
modmul
summary(modmul)
anova(modmul,update(modmul,model="add"))
logLik(modmul)