Data selection

The optional [data_selection] block excludes records from the dataset at read time without modifying the CSV. It is the ferx equivalent of NONMEM’s $DATA IGNORE= / $DATA ACCEPT=.

Syntax

[data_selection]
  ignore = <expression>
  accept = <expression>
  ignore_subjects = [<id>, ...]

All three keys are optional. Omitting the block entirely means “use all records.”

ignore

A record is excluded when the expression is true.

[data_selection]
  ignore = DV < 0.001

Multiple ignore lines are independent: a record is excluded when any one of them matches. Each line is a separate reason to drop the record.

Within a single line, join sub-conditions with &&:

[data_selection]
  ignore = EVID == 0 && DV < 0.001
Note

|| within a single expression is not supported. Use multiple lines instead — each line already acts as an independent OR condition.

accept

A record is kept only when the expression is true; excluded otherwise.

[data_selection]
  accept = BW >= 30 && BW < 48

Multiple accept lines are independent: a record is excluded when any accept condition fails.

ignore_subjects

Exclude all records for one or more subjects by ID value:

[data_selection]
  ignore_subjects = [3, 17]

Single-subject shorthand (no brackets):

[data_selection]
  ignore_subjects = 3

Evaluation order

For each record:

  1. ignore_subjects — exclude immediately if the ID is listed.
  2. ignore — exclude if any clause matches.
  3. accept — exclude if any clause fails.

A record must pass all three stages to be included.

Supported columns

Column names are case-insensitive.

Column Type
ID String — only == and != (no ordered comparisons)
TIME, DV, EVID, AMT, CMT, RATE, MDV, CENS, II, SS Numeric
Any covariate column Numeric

Missing values: a comparison against a column whose value is missing (., blank, NA) never fires. For example, ignore = DV < 0.001 does not exclude dose rows — their DV is missing. Only observation rows with a real DV below the threshold are dropped.

ADDL and OCC are not filter targets.

Exclusion summary

After reading the data, ferx prints and stores exclusion counts:

DATA SELECTION
  Records read: 120    Obs excl.: 2    Doses excl.: 0    Other excl.: 0
  Fired ignore: DV < 1.0

The same information is in fit$exclusions:

fit$exclusions$n_records_total   # 120
fit$exclusions$n_obs_excluded    # 2
fit$exclusions$fired_ignore      # "ignore: DV < 1.0"

R companion: ferx_selection()

ferx_selection() applies the same filtering logic in pure R before committing to a fit, so you can inspect which records would be excluded:

library(ferx)
ex  <- ferx_example("warfarin_data_selection")

sel <- ferx_selection(ex$data, ignore = "DV < 1.0")
nrow(sel)                           # retained records
sel$exclusions$n_obs_excluded       # number of excluded obs

# Inspect excluded rows and the matched rule:
ferx_selection_excluded(sel)[, c("ID", "TIME", "DV", ".exclude_reason")]

The returned ferx_data object can be passed directly to ferx_fit():

fit <- ferx_fit(model = ex$model,
                data  = ferx_selection(ex$data, ignore = "DV < 1.0"))

When both a [data_selection] block in the model file and a ferx_selection() call supply conditions, they are merged: duplicate expressions are deduplicated automatically.

NONMEM equivalents

NONMEM ferx
$DATA IGNORE=C ignore = C == 1
$DATA IGNORE=(BW.GT.80) ignore = BW > 80
$DATA ACCEPT=(DV.GE.0.001) accept = DV >= 0.001
$DATA IGNORE=(ID.EQ.3) IGNORE=(ID.EQ.17) ignore_subjects = [3, 17]

Example

library(ferx)
ex  <- ferx_example("warfarin_data_selection")
fit <- ferx_fit(ex$model, ex$data)
print(fit)

The warfarin_data_selection model drops observations below a surrogate LLOQ of 1.0 mg/L (ignore = DV < 1.0). The print output shows:

DATA SELECTION
  Records read: 120    Obs excl.: 2    Doses excl.: 0    Other excl.: 0
  Fired ignore: DV < 1.0

See also Example: data selection for a full worked example including the R-side preview with ferx_selection().

See also