Read text and csv files

read_text() reads a delimited text file (.txt, .tab, .data) into an AnnData() object and read_csv() is the same with the delimiter ",", as in Python's read_text() and read_csv(). Blank lines are skipped and lines starting with # are comments. If the first line holds names (its last field is not a number) they become the var names, otherwise the last comment line is used or the names are "0", "1" and so on. If a line starts with a field that is not a number (or first_column_names is TRUE) the first column holds the obs names, and double quotes around names are removed. A first line with one name fewer than the second line's values is read as integer names for both axes.

Usage

read_text(
  filename,
  delimiter = NULL,
  first_column_names = NULL,
  dtype = "float32"
)

read_csv(
  filename,
  delimiter = ",",
  first_column_names = NULL,
  dtype = "float32"
)

Arguments

filename

A path (a .gz or .bz2 suffix is decompressed), a connection, or a character vector of lines. A single line is passed as I("..."), because a single string is a path.

delimiter

The field delimiter. With NULL lines are split at any run of white space, which differs from " ".

first_column_names

Whether the first column stores the obs names, detected when NULL.

dtype

A numpy dtype string for X, such as "float32" (the default), "float64" or "int32". The values are R doubles or integers that carry the On-disk dtype.

Value

An AnnData() object with a dense X.

Examples

lines <- c("cell,g1,g2", "c1,1,2", "c2,3,4")
read_csv(lines)
#> AnnData object with n_obs × n_vars = 2 × 2
#>     layers: None (.X)