diff --git a/nexus/apple-touch-icon-120x120.png b/nexus/apple-touch-icon-120x120.png index fa1686576..6a9a50595 100644 Binary files a/nexus/apple-touch-icon-120x120.png and b/nexus/apple-touch-icon-120x120.png differ diff --git a/nexus/apple-touch-icon-152x152.png b/nexus/apple-touch-icon-152x152.png index ae32de691..0738cd90b 100644 Binary files a/nexus/apple-touch-icon-152x152.png and b/nexus/apple-touch-icon-152x152.png differ diff --git a/nexus/apple-touch-icon-180x180.png b/nexus/apple-touch-icon-180x180.png index 2c44765a8..c9e055bc4 100644 Binary files a/nexus/apple-touch-icon-180x180.png and b/nexus/apple-touch-icon-180x180.png differ diff --git a/nexus/apple-touch-icon-60x60.png b/nexus/apple-touch-icon-60x60.png index 99f2a28e8..21a25a16a 100644 Binary files a/nexus/apple-touch-icon-60x60.png and b/nexus/apple-touch-icon-60x60.png differ diff --git a/nexus/apple-touch-icon-76x76.png b/nexus/apple-touch-icon-76x76.png index 84ef01a13..a4a428621 100644 Binary files a/nexus/apple-touch-icon-76x76.png and b/nexus/apple-touch-icon-76x76.png differ diff --git a/nexus/apple-touch-icon.png b/nexus/apple-touch-icon.png index f14c12a9a..28105ec1b 100644 Binary files a/nexus/apple-touch-icon.png and b/nexus/apple-touch-icon.png differ diff --git a/nexus/articles/bibliography.html b/nexus/articles/bibliography.html index f91f08b9e..0553cd065 100644 --- a/nexus/articles/bibliography.html +++ b/nexus/articles/bibliography.html @@ -89,7 +89,7 @@
vignettes/bibliography.Rmd
bibliography.Rmd
vignettes/nexus.Rmd
nexus.Rmd
-library(nexus)
-#>
-#> Attaching package: 'nexus'
-#> The following objects are masked from 'package:stats':
-#>
-#> biplot, screeplot
library(nexus)
nexus provides a set of S4 classes that represent different special types of matrix. The most basic class represents a compositional data matrix, i.e. quantitative (nonnegative) descriptions @@ -162,7 +157,7 @@
## Compositional bar plot
plot(coda, order = "A")
-
+
The amount of every part is limited to the whole. The absolute amount of the whole is noninformative since it is typically due to artifacts on the measurement procedure. Thus only relative changes are relevant. If the relative amount of one part increases, the amounts of other parts must decrease, introducing spurious anticorrelation.
+## Data from Day et al. 2011
+data("kommos", package = "folio")
+
+## Remove rows with missing values
+kommos <- remove_NA(kommos, margin = 1)
+
+## Coerce to a compositional matrix
+coda <- as_composition(kommos)
+#> 2 qualitative variables were removed: type, date.
+
+## Set groups
+set_groups(coda) <- kommos$type
+## Compositional bar plot
+plot(coda, order = "Ca")
+## CLR
+clr <- transform_clr(coda)
+plot(clr, border = c("#EE7733", "#0077BB", "#33BBEE", "#EE3377"))
+## PCA
+clr_pca <- pca(clr, scale = FALSE)
+
+viz_individuals(clr_pca, highlight = get_groups(coda), pch = 16,
+ col = c("#EE7733", "#0077BB", "#33BBEE", "#EE3377"))
+viz_variables(clr_pca)