visual_prep: Batch Effect Diagnostic Visualization Preparation

Description

Prepare relevant datasets and statistical test results for batch/site effect diagnostic visualization.

Usage

visual_prep(
  type = "lm",
  features,
  batch,
  covariates = NULL,
  interaction = NULL,
  random = NULL,
  smooth = NULL,
  smooth_int_type = NULL,
  df,
  cores = detectCores(),
  mdmr = TRUE
)

Value

visual_prep returns a list containing the following components:

residual_add_df: Residuals that might contain additive and multiplicative joint batch effects
residual_ml_df: Residuals that might contain multiplicative batch effect
pr.feature: PCA results
pca_summary: A dataframe containing the variance explained by Principal Components (PCs)
pca_df: A dataframe contains features in the form of PCs
tsne_df: A dataframe prepared for T-SNE plots
kr_test_df: A dataframe contains Kenward-Roger(KR) test results
fk_test_df: A dataframe contains Fligner-Killeen(FK) test results
mdmr.summary: A dataframe contains MDMR results
anova_test_df: A dataframe contains ANOVA test results
kw_test_df: A dataframe contains Kruskal-Wallis test results
lv_test_df: A dataframe contains Levene's test results
bl_test_df: A dataframe contains Bartlett's test results
red: A parameter to highlight significant p-values in result table
info: A list contains input information like batch, covariates, df etc

Arguments

type: The name of a regression model to be used in batch effect diagnostics stage: "lmer", "lm", "gam".
features: The name of the features to be evaluated.
batch: The name of the batch variable.
covariates: Name of covariates supplied to model.
interaction: Expression of interaction terms supplied to model (eg: "age,diagnosis").
random: Variable name of a random effect in linear mixed effect model.
smooth: Variable name that requires a smooth function.
smooth_int_type: Indicates the type of interaction in gam models. By default, smooth_int_type is set to be "linear", representing linear interaction terms. "categorical-continuous", "factor-smooth" both represent categorical-continuous interactions ("factor-smooth" includes categorical variable as part of the smooth), "tensor" represents interactions with different scales, and "smooth-smooth" represents interaction between smoothed variables.
df: Dataset to be evaluated.
cores: number of cores used for parallel computing.
mdmr: A boolean variable indicating whether to run the MDMR test (default: TRUE).

Examples

Run this code

visual_prep(type = "lm", features = colnames(adni)[43:53], batch = "manufac",
covariates = c("AGE", "SEX", "DIAGNOSIS"), df = head(adni, 500), cores = 1)

Run the code above in your browser using DataLab