hd_cluster() takes a dataset and returns the same dataset ordered
according to the clustering method of the rows and columns. This dataset
can then be used to plot a heatmap with ggplot2 that is not having clustering functionality.
Usage
hd_cluster(
dat,
distance_method = "euclidean",
clustering_method = "ward.D2",
cluster_rows = TRUE,
cluster_cols = TRUE,
normalize = TRUE
)Arguments
- dat
An HDAnalyzeR object or a dataset in wide format and sample ID as its first column.
- distance_method
The distance method to use. Default is "euclidean". Other options are "maximum", "manhattan", "canberra", "binary" or "minkowski".
- clustering_method
The clustering method to use. Default is "ward.D2". Other options are "ward.D", "single", "complete", "average" (= UPGMA), "mcquitty" (= WPGMA), "median" (= WPGMC) or "centroid" (= UPGMC)
- cluster_rows
Whether to cluster rows. Default is TRUE.
- cluster_cols
Whether to cluster columns. Default is TRUE.
- normalize
A logical value indicating whether to normalize the data. Z-score normalization is applied using the
hd_normalize()function. Default is TRUE.
Value
A list with the dataset ordered according to the clustering of the rows and columns and the hierarchical clustering object for rows and columns.
Details
You can read more about the distance and clustering methods in the documentation
of the dist() and hclust() functions in the stats package.
Examples
# Create the HDAnalyzeR object providing the data and metadata
hd_object <- hd_initialize(example_data, example_metadata)
# Clustered data
hd_cluster(hd_object)
#> Warning: attributes are not identical across measure variables; they will be dropped
#> $cluster_res
#> # A tibble: 586 × 101
#> DAid AARSD1 ARHGAP1 AKR1B1 ATOX1 ANXA11 AKT1S1 ARHGEF12 AXIN1 ANXA3
#> <fct> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 DA00029 4.04 -1.42 -2.24 -1.86 -3.11 -0.0624 -1.51 -0.960 -2.46
#> 2 DA00273 2.36 -0.923 -0.801 -2.66 -3.30 -0.580 -2.15 -2.38 -2.22
#> 3 DA00511 1.88 -1.56 0.144 -1.07 -1.39 -2.37 -2.08 -2.05 -1.78
#> 4 DA00016 1.79 -0.835 -0.734 -2.36 -2.03 -2.03 -1.41 -1.33 -1.97
#> 5 DA00267 NA -0.281 0.0621 -1.29 -2.73 -1.47 -2.03 -1.20 -1.16
#> 6 DA00494 1.98 0.0550 -1.09 -1.26 -1.58 -1.60 -0.568 -1.97 -1.18
#> 7 DA00250 2.92 -0.0653 -0.676 -1.18 -3.54 -1.26 -1.50 -1.56 -2.04
#> 8 DA00271 1.40 -0.0407 0.0906 -1.09 -1.50 -0.256 -1.19 -1.17 -0.228
#> 9 DA00418 1.79 -0.476 -0.387 -0.241 -0.473 -0.261 -1.38 -0.886 -1.64
#> 10 DA00557 2.83 0.557 -0.963 0.119 -0.486 0.442 -0.601 -0.436 -0.155
#> # ℹ 576 more rows
#> # ℹ 91 more variables: ANXA4 <dbl>, AIF1 <dbl>, ATP6V1F <dbl>, AHCY <dbl>,
#> # ATXN10 <dbl>, ACAA1 <dbl>, ACOX1 <dbl>, AKT3 <dbl>, ARSB <dbl>,
#> # AIFM1 <dbl>, ATP5IF1 <dbl>, ANGPT1 <dbl>, APP <dbl>, ANKRD54 <dbl>,
#> # AK1 <dbl>, ATG4A <dbl>, ADAM23 <dbl>, AMIGO2 <dbl>, ADAMTS8 <dbl>,
#> # AGER <dbl>, ADGRG2 <dbl>, AOC3 <dbl>, AMBN <dbl>, ADCYAP1R1 <dbl>,
#> # AOC1 <dbl>, ANGPTL3 <dbl>, APOH <dbl>, AGR3 <dbl>, ACP6 <dbl>, AMN <dbl>, …
#>
#> $cluster_rows
#>
#> Call:
#> stats::hclust(d = stats::dist(x, method = distance), method = method)
#>
#> Cluster method : ward.D2
#> Distance : euclidean
#> Number of objects: 586
#>
#>
#> $cluster_cols
#>
#> Call:
#> stats::hclust(d = stats::dist(x, method = distance), method = method)
#>
#> Cluster method : ward.D2
#> Distance : euclidean
#> Number of objects: 100
#>
#>
#> attr(,"class")
#> [1] "hd_cluster"