spinToPlots: Take anansi output and wrangle it to a list of plottable...

In thomazbastiaanssen/anansi: Annotation-Based Analysis of Specific Interactions

Take anansi output and wrangle it to a list of plottable objects.

Description

Take anansi output and wrangle it to a list of plottable objects.

Usage

spinToPlots(
  anansiYarn,
  target = NULL,
  translate = FALSE,
  Y_translation = NULL,
  X_translation = NULL
)

Arguments

anansiYarn	The output of the main anansi function.
target	A boolean matrix. Determines which associations should be prepared for plotting. Default is the dictionary.
translate	Boolean, default is FALSE Toggles whether to translate the names of the features in tableX and tableY to human readable names.
Y_translation	data.frame, a lookup table with featureY names as the first column and human readable names as the second. See cpd_translation for an example file.
X_translation	data.frame, a lookup table with featureX names as the first column and human readable names as the second. See KO_translation for an example file.

Value

a list of ready to plot data.frames and their names.

Examples

# Starting off from the example in ?anansi

data(dictionary)
data(FMT_data)

# Clean and prepare the example data.
# In the example dataset, the metabolites are already cleaned.

KOs <- floor(FMT_KOs)
KOs <- apply(KOs, c(1, 2), function(x) as.numeric(as.character(x)))
KOs <- KOs[apply(KOs == 0, 1, sum) <= (ncol(KOs) * 0.90), ]
KOs.exp <- clr_c(KOs[row.names(KOs) %in% sort(unique(unlist(anansi_dic))), ])

t1 <- t(FMT_metab)
t2 <- t(KOs.exp)

# Run anansi pipeline.

web <- weaveWebFromTables(
  tableY = t1,
  tableX = t2,
  dictionary = anansi_dic
)

anansi_out <- anansi(
  web = web,
  formula = ~Legend,
  groups = "Legend",
  metadata = FMT_metadata,
  adjust.method = "BH",
  verbose = TRUE
)

# Let's look at all canonical interactions that also have a sufficiently well-fitting model:

outPlots <- spinToPlots(anansi_out)

# load ggplot2 and patchwork for plotting

library(ggplot2)
library(patchwork)

plotted <- lapply(outPlots, FUN = function(p) {
  # Main ggplot call
  ggplot(data = p$data, aes(x = X, y = Y, fill = groups)) +

    # Define geoms:
    geom_point(shape = 21) +
    geom_smooth(method = "lm") +
    theme_bw() +

    # Improve annotation:
    scale_fill_manual(values = c(
      "Young yFMT" = "#2166ac",
      "Aged oFMT" = "#b2182b",
      "Aged yFMT" = "#ef8a62"
    )) +
    ylab(p$name[1]) +
    xlab(p$name[2]) +
    ggtitle(paste(p$name[1], "vs", p$name[2]))
})

# Call patchwork to unify and arrange the first 6 plots

wrap_plots(plotted[1:6]) + plot_layout(guides = "collect")

thomazbastiaanssen/anansi documentation built on Feb. 9, 2025, 2:07 p.m.