Integrated nested Laplace approximations (INLA) were introduced by Rue et al. (2009). It is a new approach for inference on latent Gaussian models. It substitutes MCMC sampling with a series of numerical approximations, providing very accurate estimates for the posterior marginals and the parameters of the model in only a fraction of the time needed by MCMC algorithms. The methodolgy is implemented in the R-package INLA which can be downloaded from the INLA webpage



The R-package BAPC implements Bayesian age-period-cohort models with a focus on prediction. Model fitting is performed using integrated nested Laplace approximations (INLA). BAPC generates age-specific and age-standardised projected rates. When interest lies in the predictive distribution, Poisson noise is added automatically.
The BAPC package depends on the INLA package which is available from and we recommend to install it in R using:

install.packages("INLA", repos="")
The BAPC package can be downloaded from R-forge . It is installed in R using the command
install.packages("BAPC", repos="")
A testing datasets representing USA lung cancer mortality counts for females from 1950 to 2007 together with the corresponding population counts is available here: us_data_2014.txt , us_pop_2014.txt .

BAPC-code for predicting the last 10 years retrospectively is provided here: illustrateBAPC.R.


BayMeth is a method for quantifying methylation in affinity capture sequencing data. It is an empirical Bayes approach that uses a fully methylated control sample to transform observed read counts into regional methylation levels. The approach is implemented in the Bioconductor package Repitools. Riebler et al. (2014) describe the methodology and all data files and code used in the paper are available here.


Bayesian inference for bivariate meta-analysis of diagnostic test studies using integrated nested Laplace approximation (INLA). A purpose built graphic user interface is available. The installation of the R package INLA is compulsory for successful usage. The INLA package can be obtained from . We recommend the testing version, which can be downloaded by running:

install.packages("INLA", repos="")
The package meta4diag is available from CRAN. A technical report describing its features and usage is available here.