to avoid spurious visualization problems, it is useful in a heatmap visualization to reorder the samples within each sample class. This function uses hierarchical clustering and dendsort to sort entries in a distance matrix.
Usage
similarity_reorderbyclass(
similarity_matrix,
sample_classes = NULL,
transform = "none",
hclust_method = "complete",
dendsort_type = "min"
)Arguments
- similarity_matrix
matrix of similarities between objects
- sample_classes
data.frame or factor denoting classes
- transform
a transformation to apply to the data
- hclust_method
which method for clustering should be used
- dendsort_type
how should dendsort do reordering?
Value
a list containing the reordering of the matrix in a:
dendrogram
numeric vector
character vector (will be NULL if rownames are not set on the matrix)
Details
The similarity_matrix should be either a square matrix of similarity values
or a distance matrix of class dist. If your matrix does not encode a "true"
distance, you can use a transform to turn it into a true distance
(for example, if you have correlation, then a distance would be 1 - correlation,
use "sub_1" as the transform argument).
The sample_classes should be either a data.frame or factor argument. If
a data.frame is passed, all columns of the data.frame will be pasted together
to create a factor for splitting the data into groups. If the rownames of the
data.frame do not correspond to the rownames or colnames of the matrix, then
it is assumed that the ordering in the matrix and the data.frame are identical.
Examples
library(visualizationQualityControl)
set.seed(1234)
mat <- matrix(rnorm(100, 2, sd = 0.5), 10, 10)
rownames(mat) <- colnames(mat) <- letters[1:10]
neworder <- similarity_reorderbyclass(mat)
mat[neworder$indices, neworder$indices]
#> a d g b c f e h
#> a 1.3964671 2.5511488 2.328294 1.761404 2.067044 1.096984 2.724748 2.003446
#> d 0.8271511 1.7493710 1.665183 2.032229 2.229795 1.492519 1.859688 2.324143
#> g 1.7126300 0.9099802 1.430696 1.744495 2.287378 2.823909 1.446341 1.304650
#> b 2.1387146 1.7622035 3.274496 1.500807 1.754657 1.708962 1.465679 1.772266
#> c 2.5422206 1.6452800 1.982620 1.611873 1.779726 1.445555 1.572318 1.816738
#> f 2.2530279 1.4161904 2.888542 1.944857 1.275898 2.281528 1.515743 1.923301
#> e 2.2145623 1.1854533 1.996198 2.479747 1.653140 1.918845 1.502830 3.035135
#> h 1.7266841 1.3295034 2.683914 1.544402 1.488172 1.613323 1.374007 1.638209
#> i 1.7177740 1.8528531 2.664782 1.581414 1.992431 2.802955 1.738086 2.129131
#> j 1.5549811 1.7670512 2.168236 3.207918 1.532026 1.421096 1.751575 1.841470
#> i j
#> a 1.911105 1.973421
#> d 1.913106 2.500757
#> g 2.274999 1.432696
#> b 1.915003 2.127598
#> c 1.313849 2.852982
#> f 2.348804 2.177775
#> e 2.425116 1.752208
#> h 1.798634 2.439102
#> i 1.904203 2.486458
#> j 1.402736 3.060559
sample_class <- data.frame(grp = rep(c("grp1", "grp2"), each = 5), stringsAsFactors = FALSE)
rownames(sample_class) <- rownames(mat)
neworder2 <- similarity_reorderbyclass(mat, sample_class[, "grp", drop = FALSE])
# if there is a class with only one member, it is dropped, with a warning
sample_class[10, "grp"] = "grp3"
neworder3 <- similarity_reorderbyclass(mat, sample_class[, "grp", drop = FALSE])
#> Warning: Removing groups: grp3
neworder3$indices # 10 should be missing
#> [1] 1 4 5 2 3 6 8 7 9
mat[neworder2$indices, neworder2$indices]
#> a d e b c i j g
#> a 1.3964671 2.5511488 2.724748 1.761404 2.067044 1.911105 1.973421 2.328294
#> d 0.8271511 1.7493710 1.859688 2.032229 2.229795 1.913106 2.500757 1.665183
#> e 2.2145623 1.1854533 1.502830 2.479747 1.653140 2.425116 1.752208 1.996198
#> b 2.1387146 1.7622035 1.465679 1.500807 1.754657 1.915003 2.127598 3.274496
#> c 2.5422206 1.6452800 1.572318 1.611873 1.779726 1.313849 2.852982 1.982620
#> i 1.7177740 1.8528531 1.738086 1.581414 1.992431 1.904203 2.486458 2.664782
#> j 1.5549811 1.7670512 1.751575 3.207918 1.532026 1.402736 3.060559 2.168236
#> g 1.7126300 0.9099802 1.446341 1.744495 2.287378 2.274999 1.432696 1.430696
#> f 2.2530279 1.4161904 1.515743 1.944857 1.275898 2.348804 2.177775 2.888542
#> h 1.7266841 1.3295034 1.374007 1.544402 1.488172 1.798634 2.439102 2.683914
#> f h
#> a 1.096984 2.003446
#> d 1.492519 2.324143
#> e 1.918845 3.035135
#> b 1.708962 1.772266
#> c 1.445555 1.816738
#> i 2.802955 2.129131
#> j 1.421096 1.841470
#> g 2.823909 1.304650
#> f 2.281528 1.923301
#> h 1.613323 1.638209
cbind(neworder$names, neworder2$names)
#> [,1] [,2]
#> [1,] "a" "a"
#> [2,] "d" "d"
#> [3,] "g" "e"
#> [4,] "b" "b"
#> [5,] "c" "c"
#> [6,] "f" "i"
#> [7,] "e" "j"
#> [8,] "h" "g"
#> [9,] "i" "f"
#> [10,] "j" "h"