Dealing with missing values: By looking at Density distribution curves can I decide best data imputation method for missing omics data?

Question

[WUSCHEL]

By looking at Density distribution curves can I decide best imputation method for omics data?

# All possible imputation methods are printed in an error, if an invalid function name is given.
impute(data_norm, fun = "")
## Error in match.arg(fun): 'arg' should be one of "bpca", "knn", "QRILC", "MLE", "MinDet", "MinProb", "man", "min", "zero", "mixed", "nbavg"
# Impute missing data using random draws from a Gaussian distribution centered around a minimal value (for MNAR)
data_imp <- impute(data_norm, fun = "MinProb", q = 0.01)

# Impute missing data using random draws from a manually defined left-shifted Gaussian distribution (for MNAR)
data_imp_man <- impute(data_norm, fun = "man", shift = 1.8, scale = 0.3)

# Impute missing data using the k-nearest neighbour approach (for MAR)
data_imp_knn <- impute(data_norm, fun = "knn", rowmax = 0.9)
The effect of the imputation on the distributions can be visualized.

# Plot intensity distributions before and after imputation
plot_imputation(data_norm, data_imp)

What are the parameters I should look at the decide best imputation method when working with several genotypes?

[Jean-Karim Heriche]

What's the data? There are established procedures for dealing with missing data for some data types. Read the literature related to your data to look for commonly used imputation methods.