A Quick Start of cola Package

Author: Zuguang Gu ( z.gu@dkfz.de )

Date: 2020-05-06

Package version: 1.4.1

Assume your matrix is stored in an object called mat, to perform consensus partitioning with cola, you only need to run following code:

# code only for demonstration
mat = adjust_matrix(mat)  # optional
rl = run_all_consensus_partition_methods(mat, mc.cores = ...)
cola_report(rl, output_dir = ..., mc.cores = ...)

In above code, there are three steps:

  1. Adjust the matrix. In this step, rows with too many NAs are removed. Rows with very low variance are removed. NA values are imputed if there are less than 50% in each row. Outliers are adjusted in each row.
  2. Run consensus partitioning with several methods. Partitioning methods are hclust (hierarchical clustering with cutree), kmeans (k-means clustering), skmeans::skmeans (spherical k-means clustering), cluster::pam (partitioning around medoids) and Mclust::mclust (model-based clustering). The default methods to extract top n rows are SD (standard deviation), CV (coefficient of variation), MAD (median absolute deviation) and ATC (ability to correlate to other rows).
  3. Generate a detailed HTML report for the complete analysis.

There are examples on real datasets for cola analysis that can be found at https://jokergoo.github.io/cola_collection/.