Calculate the percentage of NAs in each row of the dataset

calc_na_percentage_row() calculates the percentage of NAs in each row of the input dataset. It filters out the rows with 0% missing data and returns the rest in descending order.

Usage

calc_na_percentage_row(dat, sample_id, chunk_size = 500)

Arguments

dat

The input dataset.

sample_id

The name of the column containing the sample IDs.

chunk_size

The number of columns to scan at a time. Default is 500.

Value

A tibble with the DAids and the percentage of NAs in each row.

Details

The counts are accumulated with rowSums() over blocks of columns instead of rowwise(), which evaluates once per row and becomes the dominant cost on datasets with thousands of samples and features. Chunking keeps the temporary logical matrix small regardless of how many features there are.