Import and process antimicrobial phenotype data from WHONET files

This function imports antimicrobial susceptibility testing (AST) data from WHONET software output files (wide CSV format) and converts it to the standardised long-format used by AMRgen.

Usage

import_whonet_pheno(
  input,
  sample_col = NULL,
  source = NULL,
  species = NULL,
  ab = NULL,
  interpret_eucast = FALSE,
  interpret_clsi = FALSE,
  interpret_ecoff = FALSE,
  include_patient_info = FALSE
)

Arguments

input

A dataframe or path to a CSV file containing WHONET AST output data

sample_col

Column name for sample identifiers. If NULL (default), the function auto-detects from known WHONET column names: "Identification number", "Identification", "laboratory", "patient_id". Supply a column name explicitly to override.

source

Optional string value to record in the source column for all data points (e.g., dataset name or study identifier)

species

Optional string indicating a single species to use for phenotype interpretation (otherwise this is inferred per-sample from the input)

ab

Optional string indicating a single antibiotic to use for phenotype interpretation (otherwise this is inferred per-sample from the input)

interpret_eucast

A logical value (default is FALSE). If TRUE, the function will re-interpret the susceptibility phenotype (SIR) for each observation based on the MIC or disk diffusion values, against EUCAST human breakpoints. These will be reported in a new column pheno_eucast, of class sir.

interpret_clsi

A logical value (default is FALSE). If TRUE, the function will re-interpret the susceptibility phenotype (SIR) for each observation based on the MIC or disk diffusion values, against CLSI human breakpoints. These will be reported in a new column pheno_clsi, of class sir.

interpret_ecoff

A logical value (default is FALSE). If TRUE, the function will re-interpret the wildtype vs nonwildtype status for each observation based on the MIC or disk diffusion values, against epidemiological cut-off (ECOFF) values. These will be reported in a new column ecoff, of class sir and coded as NWT (nonwildtype) or WT (wildtype).

include_patient_info

Include patient demographic columns in output

Value

Standardised AST data frame

Examples

# Built-in AMR package WHONET example dataset (standard uppercase format)
result <- import_whonet_pheno(AMR::WHONET)
#> Warning: There were 2 warnings in `mutate()`.
#> The first warning was:
#> ℹ In argument: `mic = as.mic(...)`.
#> Caused by warning:
#> ! in `as.mic()`: 1311 results in column mic truncated (9%) that were invalid
#> MICs: "S", "I", and "R"
#> ℹ Run `dplyr::last_dplyr_warnings()` to see the 1 remaining warning.
head(result)
#> # A tibble: 6 × 31
#>   id     drug   mic  disk guideline method  platform disk_potency pheno_provided
#>   <chr>  <ab> <mic> <dsk> <chr>     <chr>   <chr>    <chr>        <sir>         
#> 1 fe41d… AMP     NA    NA CLSI      disk d… NA       10             S           
#> 2 fe41d… AMC     NA    NA EUCAST    disk d… NA       20             S           
#> 3 fe41d… TZP     NA    NA EUCAST    disk d… NA       30             S           
#> 4 fe41d… FEP     NA    NA EUCAST    disk d… NA       30             NA          
#> 5 fe41d… CTX     NA    NA EUCAST    disk d… NA       5              NA          
#> 6 fe41d… FOX     NA    NA EUCAST    disk d… NA       30             NA          
#> # ℹ 22 more variables: spp_pheno <mo>, `Specimen number` <int>, Organism <chr>,
#> #   Country <chr>, Laboratory <chr>, collection_date <date>,
#> #   `Specimen type` <chr>, `Specimen type (Numeric)` <dbl>, Reason <chr>,
#> #   `Isolate number` <int>, `Organism type` <chr>, Serotype <chr>,
#> #   `Beta-lactamase` <lgl>, ESBL <lgl>, Carbapenemase <lgl>,
#> #   `MRSA screening test` <lgl>, `Inducible clindamycin resistance` <lgl>,
#> #   Comment <chr>, `Date of data entry` <date>, ab_col <chr>, …

if (FALSE) { # \dontrun{
# WHONET file with lowercase columns and _SIR suffix (e.g. amc_nd20, amc_nd20_SIR)
# import_whonet_pheno("path/to/whonet_export.csv", sample_col = "patient_id")

# Include patient demographics in output
import_whonet_pheno("path/to/whonet_export.csv",
  sample_col = "patient_id",
  include_patient_info = TRUE
)

# Interpret against EUCAST breakpoints for a specific species
result_interp <- import_whonet_pheno(AMR::WHONET,
  interpret_eucast = TRUE,
  species = "Escherichia coli"
)
} # }