This function runs multi-omics integration analysis using a specified clustering algorithm. Users can choose from a variety of algorithms to perform data integration on multiple modalities.
Arguments
- data
A list of matrices where each element represents a different modality (e.g., RNA, protein, methylation). Each matrix should have rows as features and columns as samples.
- algorithm
Character. The integration algorithm to use. Options include "cpca", "iclusterbayes", "intnmf", "lracluster", "mcia", "nemo", "pinsplus", "rgcca", "sgcca", "snf", "cimlr", "bcc".
- N.clust
Integer. Number of clusters to create (recommended).
- ...
Additional algorithm-specific arguments passed to the underlying functions.
Details
This function provides a unified interface to multiple multi-omics integration algorithms. Each algorithm has its own characteristics:
SNF: Similarity Network Fusion
CPCA: Consensus PCA
iClusterBayes: Bayesian integrative clustering
IntNMF: Integrative Non-negative Matrix Factorization
LRAcluster: Low-Rank Approximation clustering
MCIA: Multiple Co-inertia Analysis
NEMO: Neighborhood based multi-omics clustering
PINSPlus: Perturbation clustering for data integration
RGCCA: Regularized Generalized CCA
SGCCA: Sparse Generalized CCA
CIMLR: Cancer Integration via Multi-kernel Learning
BCC: Bayesian Consensus Clustering
Examples
if (FALSE) { # \dontrun{
# Create example data
data1 <- matrix(rnorm(5000), nrow = 50, ncol = 100)
data2 <- matrix(rnorm(5000), nrow = 50, ncol = 100)
colnames(data1) <- colnames(data2) <- paste0("Sample", 1:100)
data_list <- list(data1, data2)
# Run integration clustering using SNF
result <- RunIntegration(data = data_list, algorithm = "snf", N.clust = 3)
} # }