Generate a Stacked Bar Plot of Assay Values Colored by a Variable

This function creates a stacked bar plot using ggplot2, where the x-axis represents MIC (Minimum Inhibitory Concentration) or disk values, the y-axis indicates their frequency, and the bars are colored by a variable (by default, colours indicate whether the assay value is expressed as a range or not). Plots can optionally be faceted on an additional categorical variable. If breakpoints are provided, or species and drug are provided so we can extract EUCAST breakpoints, vertical lines indicating the S/R breakpoints and ECOFF will be added to the plot.

Usage

assay_by_var(
  pheno_table,
  antibiotic,
  measure = "mic",
  facet_var = NULL,
  species = NULL,
  bp_site = NULL,
  bp_S = NULL,
  bp_R = NULL,
  ecoff = NULL,
  guideline = "EUCAST 2025",
  marker_free_strains = NULL,
  colour_by = NULL,
  cols = NULL
)

Arguments

pheno_table: Phenotype table in standard format as per import_ast().
antibiotic: Name of the antibiotic.
measure: Field name containing the assay measurements to plot (default "mic").
facet_var: (optional) Field name containing a field to facet on (default NULL).
species: (optional) Name of species, so we can retrieve breakpoints to print at the top of the plot to help interpret it.
bp_site: (optional) Breakpoint site to retrieve (only relevant if also supplying species to retrieve breakpoints, and not supplying breakpoints via bp_S, bp_R, ecoff).
bp_S: (optional) S breakpoint
bp_R: (optional) R breakpoint
ecoff: (optional) ECOFF breakpoint
guideline: (optional) Guideline to use when looking up breakpoints (default 'EUCAST 2025')
marker_free_strains: (optional) Vector of sample names to select to get their own plot. Most useful for defining the set of strains with no known markers associated with the given antibiotic, so you can view the distribution of assay values for strains expected to be wildtype, which can help to identify issues with the assay.
colour_by: (optional) Field name containing a field to colour bars by (default NULL, which will colour each bar to indicate whether the value is expressed as a range or not)
cols: (optional) Manual colour scale to use for plot. If NULL, colour_by variable is of class 'sir', bars will by default be coloured using standard SIR colours.

Value

A list containing

plot: Main plot with all samples that have assay data for the given antibiotic
plot_nomarkers: Additional plot showing only those samples listed in marker_free_strains

Examples

# plot MIC distribution, highlighting values expressed as ranges
assay_by_var(pheno_table=ecoli_ast, antibiotic="Ciprofloxacin", 
                measure="mic")
#> $plot_nomarkers
#> NULL
#> 
#> $plot

#> 

# colour by SIR interpretion recorded in column 'pheno_clsi'
assay_by_var(pheno_table=ecoli_ast, antibiotic="Ciprofloxacin", 
                measure="mic", colour_by = "pheno_clsi")
#> $plot_nomarkers
#> NULL
#> 
#> $plot

#> 

# look up ECOFF and CLSI breakpoints and annotate these on the plot
assay_by_var(pheno_table=ecoli_ast, antibiotic="Ciprofloxacin", 
                measure="mic", colour_by = "pheno_clsi", 
                species="E. coli", guideline="CLSI 2025")
#> Error in executing command: object of type 'builtin' is not subsettable
#>   MIC breakpoints determined using AMR package: S <= 0.25 and R > 1
#> $plot_nomarkers
#> NULL
#> 
#> $plot
#> Warning: Removed 26 rows containing missing values or values outside the scale range
#> (`geom_vline()`).

#> 

# facet by method
assay_by_var(pheno_table=ecoli_ast, antibiotic="Ciprofloxacin", 
                measure="mic", colour_by = "pheno_clsi", 
                species="E. coli", guideline="CLSI 2025", 
                facet_var ="method")
#> Error in executing command: object of type 'builtin' is not subsettable
#>   MIC breakpoints determined using AMR package: S <= 0.25 and R > 1
#> $plot_nomarkers
#> NULL
#> 
#> $plot
#> Warning: Removed 208 rows containing missing values or values outside the scale range
#> (`geom_vline()`).

#>