Class for symmetric correlation matrices

The class allows to store the upper triangular matrix of a symmetric matrix (think correlation matrix, distance matrix, etc.) in an memory-efficient form and return a data.table or dense (or sparse) R matrix if need be.

Methods

Public methods

• upper_triangular_sym_mat$new()

• upper_triangular_sym_mat$print()

• upper_triangular_sym_mat$get_data_table()

• upper_triangular_sym_mat$get_sym_matrix()

• upper_triangular_sym_mat$get_sparse_matrix()

• upper_triangular_sym_mat$get_data()

• upper_triangular_sym_mat$clone()

---

upper_triangular_sym_mat$new()

Initialises the R6 class.

Usage

upper_triangular_sym_mat$new(values, features, shift)

Arguments

values

Numerical vector. The correlation coefficients of the upper triangular correlation matrix stored as a row-major vector

features

String vector. The features of the correlation matrix.

shift

Bollean Was a shift applied during the calculation of the upper triangular matrix. If FALSE, the diagonal was included, if TRUE, the diagonal was removed.

Returns

Returns the initialised class.

---

upper_triangular_sym_mat$print()

Print the class

Usage

upper_triangular_sym_mat$print()

Returns

Returns the initialised class

---

upper_triangular_sym_mat$get_data_table()

Returns the data in form of a data.table.

Usage

upper_triangular_sym_mat$get_data_table(factor = FALSE, .verbose = TRUE)

Arguments

factor

Boolean. Shall the string columns be transformed into factors. Reduces size of the object; however, takes longer to generate.

.verbose

Boolean. Controls verbosity.

Returns

A data.table with three columns:

• feature_a: The name of the first feature in the correlation matrix.

• feature_b: The name of the second feature in the correlation matrix.

• cor: The correlation coefficients between these two features.

---

upper_triangular_sym_mat$get_sym_matrix()

Return the full correlation matrix.

Usage

upper_triangular_sym_mat$get_sym_matrix(.verbose = TRUE)

Arguments

.verbose

Boolean. Controls verbosity.

Returns

Returns the correlation matrix as a dense R matrix.

---

upper_triangular_sym_mat$get_sparse_matrix()

Return a sparse version of the correlation matrix

Usage

upper_triangular_sym_mat$get_sparse_matrix(.verbose = TRUE)

Arguments

.verbose

Boolean. Controls verbosity

Returns

The sparse matrix.

---

upper_triangular_sym_mat$get_data()

Return the correlation data and shift

Usage

upper_triangular_sym_mat$get_data()

Returns

A list with

• cor_data - Numeric vector. The values.

• features - String. The feature names.

• n_features - Integer. Number of initial features.

• shift - Integer. The applied shift.

---

upper_triangular_sym_mat$clone()

The objects of this class are cloneable with this method.

Usage

upper_triangular_sym_mat$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.

Examples

# store the off-diagonal of a correlation matrix and get it back
set.seed(42)
cor_mat <- cor(matrix(rnorm(80), nrow = 20, ncol = 4))
object <- upper_triangular_sym_mat$new(
  values = cor_mat[lower.tri(cor_mat)],
  features = sprintf("gene_%i", 1:4),
  shift = TRUE
)
object$get_data_table(.verbose = FALSE)
#>    feature_a feature_b         sim
#>       <char>    <char>       <num>
#> 1:    gene_1    gene_2  0.43694936
#> 2:    gene_1    gene_3  0.04324370
#> 3:    gene_1    gene_4 -0.03960938
#> 4:    gene_2    gene_3  0.03976509
#> 5:    gene_2    gene_4 -0.06555478
#> 6:    gene_3    gene_4  0.33289052
dim(object$get_sym_matrix(.verbose = FALSE))
#> [1] 4 4