Skip to contents

Canonical imuGAP input schema

Source: R/schema.R
IMURUN_SCHEMA.Rd

The column schema that imurun targets for each of its two inputs.

Usage

IMURUN_SCHEMA

Format

A named list of two character vectors (observations, locations), each giving the required column names in schema order.

Details

imurun simplifies imuGAP's inputs: rather than a separate populations sheet, each observation row carries its own loc_id/year/age_min/age_max/dose, and imurun constructs the imuGAP populations from the observations (see build_populations()). This list gives the column names that are expected once header aliases (see IMURUN_HEADER_ALIASES) have been applied and before types and values are enforced.

Each element of IMURUN_SCHEMA describes one sheet/input:

  • observations: One row per observation. Columns: obs_id (a unique identifier – the loader assigns one automatically if you omit it, so it is not a user column), loc_id (must exist in locations), year (positive integer observation year), age_min and age_max (positive integers giving the inclusive age span the count was drawn over, with age_min <= age_max), dose (integer in 1:max_dose), positive (non-negative integer count of positive results), sample_n (positive integer sample size, with positive <= sample_n). Optional: censored (NA or 1).

    A single-age observation may be written with one age column instead of age_min/age_max; the loader expands it to age_min = age_max = age (see expand_obs_age()).

  • locations: The location hierarchy. Required columns: loc_id (unique identifier), parent_id (the parent's loc_id, or NA for the single root).

See also

validate_inputs(), read_inputs(), IMURUN_TARGET_SCHEMA

Examples

IMURUN_SCHEMA$observations
#> [1] "obs_id"   "loc_id"   "year"     "age_min"  "age_max"  "dose"     "positive"
#> [8] "sample_n"
IMURUN_SCHEMA$locations
#> [1] "loc_id"    "parent_id"