eBIC_allmodels {mlmm.gwas}

Compute eBIC and BIC criteria

Description

Compute log likelihood, BIC and eBIC.

The model with the smallest eBIC should be selected.

Usage

eBIC_allmodels(Y, selec_XX, KK, nb.tests, cofs = NULL, female = NULL,
  male = NULL, lambda=NULL)

Arguments

Y

A numeric named vector where the names are individuals names and the values their phenotype.

selec_XX

A list of length one, two or three matrices depending on the models. Use helper function frommlmm_toebic to get this argument.

KK

a list of one, two or three matrices depending on the models

- additive: a n by n matrix, where n=number of individuals, with rownames()=colnames()=individual names

- additive+dominance: two n by n matrices, where n=number of individuals, with rownames()=colnames()=individual names

- female+male: a n.female by n.female matrix, with rownames()=colnames()=female names and a n.male by n.male matrix, with rownames()=colnames()=male names

- female+male+interaction: the same two matrices as the model female+male and a n by n matrix, where n=number of individuals, with rownames()=colnames()=individual names

nb.tests

number of computed tests (total number of SNPs)

cofs

A n by q matrix, where n=number of individuals, q=number of fixed effect,

female

A factor of levels female names and length n, only for the last two models

male

A factor of levels male names and length n, only for the last two models

lambda

penalty used in the computation of the eBIC; if NULL, the default will be 1 - 1/(2k) with L=n^k where L=total number of SNPs (see function "lambda.calc")

Value

A matrix with a line for each mlmm step and 4 columns : BIC, ajout, eBIC_0.5 and LogL.

Examples

### Additive model ###
## Not run: 
data("mlmm.gwas.AD")

XX = list(Xa)
KK = list(K.add)

# GWAS
res_mlmm <- mlmm_allmodels(floweringDateAD, XX, KK)
manhattan.plot(res_mlmm)

# Model selection
sel_XX <- frommlmm_toebic(XX, res_mlmm)
res.eBIC <- eBIC_allmodels(floweringDateAD, sel_XX, KK, ncol(Xa))

# Effects estimations with the selected model
sel_XXclass <- fromeBICtoEstimation(sel_XX, res.eBIC)
eff.estimations <- Estimation_allmodels(floweringDateAD, sel_XXclass, KK)
genotypes.boxplot(Xa, floweringDateAD, effects = eff.estimations)

## End(Not run)

### Additive + dominance model
## Not run: 
data("mlmm.gwas.AD")

XX = list(Xa, Xd)
KK = list(K.add, K.dom)

# GWAS
res_mlmm <- mlmm_allmodels(floweringDateAD, XX, KK)
manhattan.plot(res_mlmm)

# Model selection
sel_XX <- frommlmm_toebic(XX, res_mlmm)
res.eBIC <- eBIC_allmodels(floweringDateAD, sel_XX, KK, ncol(Xa))
#the selected model is the null model

## End(Not run)


### Female+Male model
## Not run: 
data("mlmm.gwas.FMI")

XX = list(Xf, Xm)
KK = list(K.female, K.male)

# GWAS
res_mlmm <- mlmm_allmodels(floweringDateFMI, XX, KK, female = female, male = male)
manhattan.plot(res_mlmm)

# Model selection
sel_XX <- frommlmm_toebic(XX, res_mlmm)
res.eBIC <- eBIC_allmodels(floweringDateFMI, sel_XX, KK, ncol(Xf), female = female, male = male)
#the selected model is the null model

## End(Not run)



### Female+Male+Interaction model
## Not run: 
data("mlmm.gwas.FMI")

XX = list(Xf, Xm, Xfm)
KK = list(K.female, K.male, K.hybrid)

# GWAS
res_mlmm <- mlmm_allmodels(floweringDateFMI, XX, KK, female = female, male = male)
manhattan.plot(res_mlmm)

# Model selection
sel_XX <- frommlmm_toebic(XX, res_mlmm)
res.eBIC <- eBIC_allmodels(floweringDateFMI, sel_XX, KK, ncol(Xf), female = female, male = male)
#the selected model is the null model

## End(Not run)