This function extracts a rectangular block from a dataframe using user-specified top/bottom row indices and left/right column identifiers (numeric or Excel-style letters). It ensures the block contains only numeric values or NAs, and returns a cleaned matrix.
Usage
extract_data(
data,
bottom_row,
right_col,
top_row = 1,
left_col = 1,
feature_name_columns = NA,
use_row_index = FALSE
)Arguments
- data
A dataframe containing the full input, including annotation columns and the numeric block to extract.
- bottom_row
Integer. Specifies the last (bottom) row of the numeric data block. Must be >= `top_row`.
- right_col
Same format as `left_col`. Specifies the right-most column of the numeric block. Must be >= `left_col` after conversion.
- top_row
Integer. Specifies the first (top) row of the numeric data block. Row indexing is 1-based.
- left_col
Column specifier for the left-most column of the data block. Can be either: - An integer index (e.g., 2 for the second column), or - A character string using Excel-style letters (e.g., "A", "AB").
Column names (e.g., "age") are **not** allowed here. Only letters or numeric indices are accepted.
- feature_name_columns
Optional character vector specifying columns in `data` to be used as row (feature) names in the output. If `NA`, generic feature names are used.
- use_row_index
Logical. If
TRUE, prepend the row index to each combined feature name to ensure uniqueness. Defaults toFALSE.