Type: Package
Title: Subsetting and Plotting Optimized for Admixture Data
Version: 1.0.0
Description: A few functions which provide a quick way of subsetting genomic admixture data and generating customizable stacked barplots.
License: GPL (≥ 3)
URL: https://github.com/SpaceCowboy-71/xadmix
BugReports: https://github.com/SpaceCowboy-71/xadmix/issues
Depends: R (≥ 2.10)
Imports: dplyr, forcats, ggplot2, magrittr, methods, rlang, stringr, tidyr, viridis
Suggests: knitr, rmarkdown
VignetteBuilder: knitr
Encoding: UTF-8
Language: en-US
LazyData: true
RoxygenNote: 7.2.0
NeedsCompilation: no
Packaged: 2022-07-07 15:08:33 UTC; Austin
Author: Lukas Schönmann ORCID iD [aut, cre]
Maintainer: Lukas Schönmann <lukas.schoenmann@outlook.com>
Repository: CRAN
Date/Publication: 2022-07-08 15:20:02 UTC

Admixture Data Stacked Barplot

Description

Stacked barplot optimized for admixture data.

Usage

admix_barplot(
  data,
  K = 2:ncol(data),
  individuals = 1,
  sortkey = NULL,
  grouping = NULL,
  palette = "default",
  names = TRUE,
  xlab = "Individuals",
  ylab = "Ancestry",
  main = "Admixture Plot",
  noclip = FALSE
)

Arguments

data

Data frame containing the admixture data.

K

Positions of the columns containing the ancestry percentages in the provided data frame; default is second to last column.

individuals

Position of the column with the names for the x-axis; default is the first column.

sortkey

Name of the column containing ancestry percentages to sort the stacked barplot with.

grouping

Name of the column by which the stacked bars are to be grouped.

palette

Either a color palette object, or a string to use one of the predefined color palettes ("viridis", "turbo", "alternating"); default is a modified ggplot palette.

names

Whether to show the x-axis bar labels or not; default is "TRUE".

xlab

A label for the x-axis.

ylab

A label for the y-axis.

main

A main title for the plot.

noclip

Directly draw the plot, with clipping removed from elements. Then function does not return an object; default is set to "FALSE". Setting to "TRUE" may require launching a new R graphics device.

Value

A ggplot object of the stacked barplot.

Examples

# load simulated admixture data
data("xadmixture")

# for data frame with ancestries (K) in fourth to last column,
# without showing bar labels
admix_barplot(xadmixture,
    K = 4:ncol(xadmixture),
    names = FALSE
)

# grouping data by column "country",
# and sorting each group by ancestry column "K1"
admix_barplot(xadmixture,
    K = 4:ncol(xadmixture),
    grouping = "country",
    sortkey = "K1",
    names = FALSE
)

# changing color palette to "turbo" from package 'viridis',
admix_barplot(xadmixture,
    K = 4:ncol(xadmixture),
    palette = "turbo",
    names = FALSE
)

# removing title and changing axis labels text
admix_barplot(xadmixture,
    K = 4:ncol(xadmixture),
    main = "",
    xlab = "Accessions",
    ylab = "Ancestry [%]",
    names = FALSE
)

# directly output grouped plot with clipping removed from elements
# (useful if there are groups with a low number of observations)
# create a subset of the data
xadmixture_sub <- admix_subset(xadmixture,
                              anc = c("K3", "K4"), 
                              pct = c(0.3, 0.2))
# generate a grouped & sorted stacked barplot 
# setting "noclip" to "TRUE" may require opening a new graphics device
dev.new()
admix_barplot(xadmixture_sub, 
             K = 4:ncol(xadmixture),          
             sortkey = "K5",
             grouping = "country", 
             palette = "viridis", 
             names = FALSE, 
             noclip = TRUE)
dev.off()

Admixture Data Subsetting

Description

Subset function optimized for admixture data. Filters for the percentages of any number of ancestry (K) columns and prints progress. Also allows passing additional arguments to filter columns with.

Usage

admix_subset(
  data,
  anc = NULL,
  pct = NULL,
  comparison = "greater",
  quiet = FALSE,
  ...
)

Arguments

data

Data frame containing the admixture data.

anc

Vector of ancestry column names to use for pairwise subsetting with percentage vector. Must be of same length as the supplied percentage vector.

pct

Vector of percentage values to use for pairwise subsetting with ancestry column name vector. Only ancestries with values above the percentage are kept.

comparison

What comparison operator to use for the subsetting. Can either be "greater" or "less"; default is "greater". Also accepts "gt", "lt", ">" and "<".

quiet

Whether to print progress or not; default is "FALSE".

...

Variable number of additional vectors for subsetting. Looking at the column with argument name, keeps only those observations with values which are elements of the argument vector.

Value

A subset of the provided data frame.

Examples

# load simulated admixture data
data("xadmixture")

# keep only observations with K1 > 0.1 and K2 > 0.01
subset1 <- admix_subset(xadmixture, 
                        anc = c("K1", "K2"), 
                        pct = c(0.1, 0.01))

# keep only observations with K2 < 0.4 and K3 < 0.1
subset2 <- admix_subset(xadmixture, 
                        anc = c("K2", "K3"), 
                        pct = c(0.4, 0.1), 
                        comparison = "less")

# keep only observations with values "GBR" or "FRA" in column 
# "country" and values "lorem" or "dolor" in column "species"
subset3 <- admix_subset(xadmixture, 
                        country = c("GBR", "FRA"), 
                        species = c("lorem", "dolor"))

# keep only observations with K1 > 0.1 and K4 < 0.3, 
# without printing progress; subsets can be chained 
# using the pipe operator from package `magrittr` 
library(magrittr)
subset4 <- admix_subset(xadmixture, 
                        anc = "K1", 
                        pct = 0.1, 
                        quiet = TRUE) %>% 
 admix_subset(anc = "K4", 
              pct = 0.3, 
              comparison = "less", 
              quiet = TRUE)

Simulated Admixture Data

Description

A dataset containing simulated admixture data of 600 observations.

Usage

xadmixture

Format

A data frame with 600 rows and 8 variables:

acc

Accession identifier

country

Country where plant material was collected

species

Name of species

K1,K2,K3,K4,K5

Admixture coefficients; expresses the proportions of the respective ancestries. Sum up to 1.

Source

Data simulated for this package; for code see: https://github.com/SpaceCowboy-71/xadmix/blob/main/data-raw/xadmixture.R

Examples

# load simulated admixture data
data("xadmixture")

# create a subset of the data
xadmixture_sub <- admix_subset(xadmixture, 
                       country = c("GBR", "FRA"),
                       anc = c("K1", "K2"), 
                       pct = c(0.02, 0.2))
                       
# generate a grouped & sorted stacked barplot
admix_barplot(xadmixture_sub, 
         K = 4:ncol(xadmixture),
         sortkey = "K1",
         grouping = "country", 
         palette = "turbo")