bpr_bayes_predict_wrap: Predict gene expression from methylation profiles using Gibbs...

Description Usage Arguments Value See Also Examples

Description

bpr_bayes_predict_wrap is a function that wraps all the necessary subroutines for performing predictions on gene expressions. Initially, it computes the posterior distribution, using Gibbs sampling, of the parameters of the basis functions so as to learn the methylation profiles. Then uses the learned parameters / coefficients of the basis functions as input features for performing linear regression in order to predict/regress the corresponding gene expression data.

Usage

1
2
3
4
5
6
bpr_bayes_predict_wrap(formula = NULL, x, y, model_name = "svm", w = NULL,
  basis = NULL, train_ind = NULL, train_perc = 0.7,
  fit_feature = "RMSE", cpg_dens_feat = TRUE, lambda = 1/2,
  w_0_mean = NULL, w_0_cov = NULL, gibbs_nsim = 5000,
  gibbs_burn_in = 1000, opt_method = "CG", opt_itnmax = 50,
  is_parallel = TRUE, no_cores = NULL, is_summary = TRUE)

Arguments

formula

An object of class formula needed when calling the lm function for performing linear regression. If NULL, the SVM regression method is used.

x

The binomial distributed observations, which has to be a list where each element is an L x 3 dimensional matrix.

y

Corresponding gene expression data for each element of the list x

model_name

A charcter denoting the regression model.

w

Optional vector of initial parameter / coefficient values.

basis

Optional basis function object, default is create_rbf_object

train_ind

Optional vector containing the indices for the train set.

train_perc

Optional parameter for defining the percentage of the dataset to be used for training set, the remaining will be the test set.

fit_feature

Additional feature on how well the profile fits the methylation data.

cpg_dens_feat

Additional feature for the CpG density across the promoter region.

lambda

The complexity penalty coefficient for penalized regression.

w_0_mean

The prior mean hyperparameter for w

w_0_cov

The prior covariance hyperparameter for w

gibbs_nsim

Optional argument giving the number of simulations of the Gibbs sampler.

gibbs_burn_in

Optional argument giving the burn in period of the Gibbs sampler.

opt_method

Parameter for defining the method to be used in the optimization procedure, see optim.

opt_itnmax

Optional parameter for defining the max number of iterations of the optimization procedure, see optim.

is_parallel

Logical, indicating if code should be run in parallel.

no_cores

Number of cores to be used, default is max_no_cores - 1.

is_summary

Logical, print the summary statistics.

Value

An mpgex object consisting of the following elements:

See Also

bpr_optim

Examples

1
2
3
4
5
6
obs <- meth_data
y   <- gex_data
basis <- create_rbf_object(M = 5)
out   <- bpr_bayes_predict_wrap(x = obs, y = y, basis = basis,
                                is_parallel = FALSE,
                                gibbs_nsim = 20, gibbs_burn_in = 10)

andreaskapou/BPRMeth-devel documentation built on May 12, 2019, 3:32 a.m.