QTLBIM (QTL Bayesian Interval Mapping)
an R package for QTL Mapping using Bayesian Inference
an R package for QTL Mapping using Bayesian Inference
The implementation of R/qtlbim includes the full process of a typical Bayesian analysis with significant control in the user's hands. Here is a list that summarizes the major capabilities of the package in aiding such analyses.
Model Setup
- Handles different phenotypes:
- Continuous phenotypes
- Ordinal Phenotypes
- Binary Phenotypes
- Models different interactions:
- Gene - Gene interaction
- Gene x Environment interaction
- Includes arbitrary covariates:
- Random covariates
- Fixed covariates
Model Selection
- A choice of fast, efficient MCMC algorithms:
- Gibbs Sampling
- Metropolis-Hastings
- Handles and updates missing genotypes
Output & Result Summarization
- Mixing behavior and convergence monitoring using trace plots or formal diagnostics
- Extensive graphical/numerical summaries:
- Bayes factors
- Posterior probabilities
- Heritability
- Genetic effects
- Genotypic values etc
Model Fitting
- Deviance information criterion (DIC)
- Bayesian FDR correction
Diagnostics
- MCMC diagnostics using the R/coda package
