Multi Omics Clustering with SNF integration and Hierarchical Clustering

Performs the integration of log R ratio matrices (mat_list) using Similarity Network Fusion (SNF) followed by hierarchical clustering, on the integrated SNF matrix, to identify clusters of cells for each specified k number of cluster (k_range). For more than 1 omic, the integration and clustering are performed only on common cells between omics.

Usage

cluster_hclust(
  mat_list,
  k_range = seq(2, 10, 1),
  dist_method = "euclidean",
  hclust_method = "ward.D",
  weights = rep(1, length(mat_list)),
  knn_affinity = 40,
  var_affinity = 1,
  knn_SNF = 40,
  iter_SNF = 50,
  knn_umap = 20,
  quiet = FALSE
)

Arguments

mat_list

A named list of log R ratio matrices (cells x features), one per omic layer (list).

k_range

A numeric vector of integers (≥2) specifying the cluster numbers (k) to extract from hierarchical clustering (numeric vector). Default is from 2 to 10.

dist_method

A string specifying the distance method for Rfast::Dist() (e.g., "euclidean", "manhattan", "cosine") (character string). Default is "euclidean".

hclust_method

A string specifying the hierarchical clustering linkage method for fastcluster::hclust() (e.g., "ward.D", "average") (character string). Default is "ward.D".

weights

A numeric vector of non-negative values of length equal to the number of omic (internally normalized to sum to 1) (numeric vector). It specifies the relatives weights of each omic for SNF with weightedSNF(). Omics with a weight of 0 will not contribute to the clustering. If NULL (default), weights are uniform.

knn_affinity

Integer specifying the number of nearest neighbors used when building affinity matrices with SNFtool::affinityMatrix() (integer). Default is 40.

var_affinity

Numeric value for the variance parameter (Gaussian kernel width sigma) when building affinity matrix with SNFtool::affinityMatrix() (numeric). Default is 1.

knn_SNF

Integer specifying the number of nearest neighbors used during the Similarity Network Fusion (SNF) with weightedSNF() (integer). Default is 40.

iter_SNF

Integer specifying the number of iterations for SNF with weightedSNF() (integer). Default is 50.

knn_umap

Integer specifying the number of nearest neighbors used for manifold approximation (UMAP) (see uwot::umap()). Default is 20.

quiet

Logical. If TRUE, suppresses informative messages during execution. Default is FALSE.

Value

A list containing:

params

List of parameters used for clustering (list).

SNF

Fused similarity matrix computed with SNF (matrix).

dist

Distance matrix derived from the SNF similarity (matrix).

hclust

Hierarchical clustering object from fastcluster::hclust() (hclust).

umap

UMAP coordinates (matrix).

clusters

A named list of clustering results (vectors of cluster labels) for each value in k_range (list).

Details

The function calculates pairwise distances between cells within each omic dataset using the specified dist_method (using only common cells between omics).

It constructs affinity matrices based on these distances, applies SNF to generate a fused similarity matrix.

weights can be assigned to each omic dataset to prioritize certain data types over others, allowing users to tailor the analysis based on the characteristics and importance of each dataset.

It then performs a hierarchical clustering using the specified hclust_method to assign clusters to each common cells.

Results are given as cluster assignments for each number of cluster specified by k_range.

See also

Similarity Network Fusion: weightedSNF().

Examples

if (FALSE) { # \dontrun{
# Load example muscadet object
data(muscadet_obj)

# Format input
# transpose matrices to: cells x features matrices
mat_list <- lapply(muscadet::matLogRatio(muscadet_obj), t)

# Run integration & clustering
result <- cluster_hclust(mat_list, k_range = 2:4)

# View results
lapply(result$clusters, table)
} # }