data.imputeData {CTD} | R Documentation |
Impute missing values
Description
Impute missing values as lowest observed value in a reference population
Usage
data.imputeData(data, ref)
Arguments
data |
- Normalized data with some missingness. Data matrix with features as rows, samples as columns. |
ref |
- Reference sample data with features as rows, samples as columns. Can include some missingness. |
Value
imputed.data - Imputed data.
Examples
data(Thistlethwaite2020)
data_mx = Thistlethwaite2020
# Data with missing values
dt_w_missing_vals = data_mx[1:25,-seq_len(8)]
# Reference data can also have missing values
ref_data = data_mx[1:25,grep("EDTA-REF", colnames(data_mx))]
fil.rate = apply(ref_data, 1, function(i) sum(is.na(i))/length(i))
# Can only impute data that are found in reference samples
dt_w_missing_vals = dt_w_missing_vals[which(fil.rate<1.0),]
ref_data = ref_data[which(fil.rate<1.0),]
imputed.data = data.imputeData(dt_w_missing_vals, ref_data)
print(any(is.na(imputed.data)))
[Package CTD version 1.2 Index]