R Coding Conventions

A shared style guide for pharmaceutical sciences practicals

Following these conventions makes your code readable, reproducible, and consistent across exercises.

1. Variable and object naming

Use `snake_case` for all variable names

Words should be separated by underscores. Do not use dots or camelCase.

Important

This is required and used throughout all course materials.

# Do this
fill_mass_mg      <- 174.2
ibu_per_mg_powder <- 0.43
label_claim_mg    <- 75.0
X_bar_s1          <- mean(stage1$pct_label)

# Avoid this
fillMassMg <- 174.2     # camelCase
ibu.per.mg <- 0.43      # dots
lc         <- 75.0      # unclear
x1         <- mean(stage1$pct_label)

Include units in variable names where relevant

Append units such as _mg, _pct, _ug_ml, _min to avoid ambiguity.

# Do this
powder_mass_mg      <- 174.2
concentration_ug_ml <- 7.56
pct_label           <- 98.4
time_min            <- 45

# Avoid this
powder <- 174.2   # mg or g?
conc   <- 7.56    # ug/mL or mg/mL?
label  <- 98.4    # % or mg?
t      <- 45      # minutes or hours?

2. Assignment operator

Always use `<-` for assignment, never `=`

Use = only for function arguments.

Important

Using <- consistently avoids confusion between assignment and argument naming.

# Do this
label_claim_mg <- 75.0
std_slope      <- 0.04478
data           <- readr::read_csv("file.csv")

# Avoid this
label_claim_mg = 75.0
std_slope      = 0.04478
data           = readr::read_csv("file.csv")

Put spaces around `<-` and operators

# Do this
C2           <- (absorbance - std_intercept) / std_slope
AV           <- k * S
fill_mass_mg <- mass_filled_mg - mass_empty_mg

# Avoid this
C2<-(absorbance-std_intercept)/std_slope
AV<-k*S
fill_mass_mg<-mass_filled_mg-mass_empty_mg

3. Printing results

Use `cat()` to print labelled results

Always include a descriptive label and a unit.

# Do this
cat("Ibuprofen content:",
        signif(ibu_per_mg_powder, 3),
        "mg Ibuprofen per mg powder\n")

cat("Mean content (X-bar):",
        signif(X_bar, 4),
        "% of label claim\n")

# Avoid this
ibu_per_mg_powder
print(ibu_per_mg_powder)
X_bar

Use `view()` to inspect data frames

Use view(data) for interactive inspection and head(data, 5) for short previews.

# Do this
view(data)
head(data, 5)

# Avoid this
print(data)
data

4. Rounding and significant figures

Use `signif()` for final results, `round()` for fixed decimals

Keep full precision in intermediate calculations and round only when printing.

# Store full precision
C2         <- (absorbance - std_intercept) / std_slope
ibu_per_mg <- C2 * 20 / 0.1 * 50 / 1000 / powder_mass_mg

# Round only when printing
cat("Ibuprofen per mg powder:", signif(ibu_per_mg, 3), "mg/mg\n")
cat("Label claim %:", round(pct_label, 2), "%\n")

Never round inside a `mutate()` column used later

# Do this
data %>%
    dplyr::mutate(
        fill_mass_mg = mass_filled_mg - mass_empty_mg,
        ibu_mg       = fill_mass_mg * ibu_per_mg_powder
    )

# Avoid this
data %>%
    dplyr::mutate(
        fill_mass_mg = round(mass_filled_mg - mass_empty_mg, 1),
        ibu_mg       = fill_mass_mg * ibu_per_mg_powder
    )

5. Pipes and data manipulation

Use `%>%` with each step on its own line

One function per line, with clear indentation.

# Do this
data <- data %>%
    dplyr::mutate(
        fill_mass_mg = mass_filled_mg - mass_empty_mg,
        ibu_mg       = fill_mass_mg * ibu_per_mg_powder,
        pct_label    = ibu_mg / label_claim_mg * 100
    )

# Avoid this
data <- dplyr::mutate(data, fill_mass_mg=mass_filled_mg-mass_empty_mg, ibu_mg=fill_mass_mg*ibu_per_mg_powder, pct_label=ibu_mg/label_claim_mg*100)

Group constants at the top of a chunk

# Do this
vol_flask1_mL  <- 50.0
vol_aliquot_mL <- 0.1
vol_flask2_mL  <- 20.0

C1 <- C2 * vol_flask2_mL / vol_aliquot_mL

# Avoid this
C1 <- C2 * 20 / 0.1

6. Plotting with ggplot2

Use `theme_classic()` as the base theme

ggplot2::ggplot(data, ggplot2::aes(x = time_min, y = mean_pct)) +
    ggplot2::geom_line() +
    ggplot2::geom_point() +
    ggplot2::labs(
        x = "Time (min)",
        y = "Dissolved (%)",
        title = "Dissolution profile - Ibuprofen 75 mg capsules"
    ) +
    ggplot2::theme_classic()

Always label axes with units in parentheses

# Do this
ggplot2::labs(
    x = "Time (min)",
    y = "Ibuprofen released (mg)"
)

# Avoid this: no labs() call, defaults to raw variable names

Put each `+` layer on its own line

# Do this
ggplot2::ggplot(data, ggplot2::aes(x = time_min, y = mean_mg)) +
    ggplot2::geom_line() +
    ggplot2::geom_point() +
    ggplot2::geom_errorbar(
        ggplot2::aes(ymin = mean_mg - sd_mg, ymax = mean_mg + sd_mg)
    ) +
    ggplot2::theme_classic()

# Avoid this
ggplot2::ggplot(data, ggplot2::aes(x=time_min,y=mean_mg))+ggplot2::geom_line()+ggplot2::geom_point()+ggplot2::theme_classic()

Use `"steelblue"` as the default single-series color

For multi-group plots, map color in aes(color = group) and let ggplot choose.

# Single series
ggplot2::geom_line(color = "steelblue")
ggplot2::geom_point(color = "steelblue")

# Multiple groups
ggplot2::ggplot(
    data,
    ggplot2::aes(x = time_min, y = mean_pct, color = formulation)
) +
    ggplot2::geom_line() +
    ggplot2::geom_point()

7. Code comments

Comment why, not what, and always comment units

# Useful comments
label_claim_mg <- 75.0  # mg - labelled dose per capsule
k_s1           <- 2.4   # EP 2.9.40, Table 2.9.40-1, n = 10
L1             <- 15.0  # acceptance limit, Stage 1 (%)

# Unhelpful comments
label_claim_mg <- 75.0  # set label claim to 75
k_s1           <- 2.4   # k value
L1             <- 15.0  # L1

Use section separator comments in long scripts

# --- Assay: Ibuprofen concentration from UV absorbance ---
C2 <- (absorbance - std_intercept) / std_slope
C1 <- C2 * vol_flask2_mL / vol_aliquot_mL

# --- EP 2.9.40 Stage 1 acceptance value ---
AV_s1 <- k_s1 * S_s1

8. Quick reference summary

Topic	Convention	Level
Variable names	`snake_case`, include units (`_mg`, `_pct`, `_mL`)	Required
Assignment	Always `<-` with spaces; `=` only inside function arguments	Required
Printing values	`cat("label:", signif(x, 3), "unit\n")`	Required
Printing tables	`view(data)` to inspect; `head(data, 5)` to preview in output	Required
Rounding	`signif(x, n)` or `round(x, n)` only at print-time; full precision in calculations	Required
Pipes	`%>%` with each step on its own indented line	Required
ggplot theme	Always `theme_classic()`	Required
Axis labels	`labs(x = "Name (unit)", y = "Name (unit)")`	Required
Plot layers	One `geom_*()` or modifier per line, `+` at end of line	Required
Default color	`"steelblue"` for single series; `aes(color = group)` for multi-series	Recommended
Constants	Define at top of chunk with inline unit comment	Recommended
Comments	Explain why and state units, not just what the code does	Required

Use the Coding Conventions

1. Variable and object naming

Use snake_case for all variable names

Include units in variable names where relevant

2. Assignment operator

Always use <- for assignment, never =

Put spaces around <- and operators

3. Printing results

Use cat() to print labelled results

Use view() to inspect data frames

4. Rounding and significant figures

Use signif() for final results, round() for fixed decimals

Never round inside a mutate() column used later

5. Pipes and data manipulation

Use %>% with each step on its own line

Group constants at the top of a chunk

6. Plotting with ggplot2

Use theme_classic() as the base theme

Always label axes with units in parentheses

Put each + layer on its own line

Use "steelblue" as the default single-series color

7. Code comments

Comment why, not what, and always comment units

Use section separator comments in long scripts

8. Quick reference summary

Use `snake_case` for all variable names

Always use `<-` for assignment, never `=`

Put spaces around `<-` and operators

Use `cat()` to print labelled results

Use `view()` to inspect data frames

Use `signif()` for final results, `round()` for fixed decimals

Never round inside a `mutate()` column used later

Use `%>%` with each step on its own line

Use `theme_classic()` as the base theme

Put each `+` layer on its own line

Use `"steelblue"` as the default single-series color