predict_expr: Predict gene expression from methylation profiles

In BPRMeth: Model higher-order methylation profiles

Description

bpr_predict_wrap is a function that wraps all the necessary subroutines for performing prediction on gene expression levels. Initially, it optimizes 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 regression in order to predict the corresponding gene expression levels.

Usage

predict_expr(
  formula = NULL,
  prof_obj,
  expr,
  anno,
  model_name = "lm",
  train_ind = NULL,
  train_perc = 0.7,
  fit_feature = "RMSE",
  cov_feature = TRUE,
  is_summary = TRUE
)

Arguments

formula	An object of class formula, e.g. see lm function. If NULL, the simple linear regression model is used.
prof_obj	Inferred profiles object. This in general will be the output of 'infer_profiles_'(inference_meth) function.
expr	Gene expression data with two columns in data.frame or data.table format. 1st column will have gene IDs and should have column name "id", 2nd column will have expression levels.
anno	Annotation data as a GRanges object.
model_name	A string denoting the regression model. Currently, available models are: "svm", "randomForest", "rlm", "mars", "gp", and "lm".
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	Use additional feature of how well the profile fits the methylation data. Either NULL for ignoring this feature or one of the following: 1) "RMSE" or 2) "NLL" which will be used as input features for predicting expression.
cov_feature	Logical, whether to use coverage as input feature for predictions.
is_summary	Logical, print the summary statistics.

Value

A 'predict_expr' object which consists of the following variables:

• train: The training data.

• test: The test data.

• model: The fitted regression model.

• train_pred The predicted values for the training data.

• test_pred The predicted values for the test data.

• train_errors: The training error metrics.

• test_errors: The test error metrics.

Author(s)

C.A.Kapourani [e-mail]

See Also

infer_profiles_mle, infer_profiles_vb, infer_profiles_gibbs, create_basis,

Examples

# Fit methylation profiles using 5 RBFs
basis <- create_rbf_object(M = 5)
prof <- infer_profiles_vb(X = encode_met$met, model = "binomial",
    basis = basis, is_parallel = FALSE, vb_max_iter = 5)
# Predict expression
pred_obj <- predict_expr(prof_obj = prof, expr = encode_expr,
  anno = encode_met$anno, model_name = "lm", is_summary = FALSE)

BPRMeth documentation built on Nov. 8, 2020, 5:54 p.m.