+
=
+
=
The futurize package allows you to easily turn sequential code
into parallel code by piping the sequential code to the futurize()
function. Easy!
library(futurize)
plan(multisession)
library(sva)
adjusted <- ComBat(dat = dat, batch = batch) |> futurize()
This vignette demonstrates how to use this approach to parallelize the sva functions.
The sva Bioconductor package provides functions for removing
batch effects and other unwanted variation in high-throughput
experiments. The ComBat() function is a widely used method for
batch effect correction using an empirical Bayes framework. It
supports parallelization via BiocParallel's BPPARAM argument.
The ComBat() function adjusts for known batch effects in
microarray or RNA-seq data:
library(sva)
# Create example data with batch effect
set.seed(42)
n_genes <- 200L
n_samples <- 40L
dat <- matrix(rnorm(n_genes * n_samples), nrow = n_genes, ncol = n_samples)
rownames(dat) <- paste0("gene", seq_len(n_genes))
colnames(dat) <- paste0("sample", seq_len(n_samples))
batch <- rep(c(1, 2), each = n_samples / 2L)
dat[, batch == 2] <- dat[, batch == 2] + 2
adjusted <- ComBat(dat = dat, batch = batch)
Here ComBat() runs sequentially, but we can easily make it run in
parallel by piping to futurize():
library(futurize)
adjusted <- ComBat(dat = dat, batch = batch) |> futurize()
This will distribute the work across the available parallel workers, given that we have set up parallel workers, e.g.
plan(multisession)
The built-in multisession backend parallelizes on your local
computer and works on all operating systems. There are other
parallel backends to choose from, including alternatives to
parallelize locally as well as distributed across remote machines,
e.g.
plan(future.mirai::mirai_multisession)
and
plan(future.batchtools::batchtools_slurm)
You can also include a model matrix for biological covariates of interest, which will be protected during batch correction:
mod <- model.matrix(~ group)
adjusted <- ComBat(dat = dat, batch = batch, mod = mod) |> futurize()
The following sva functions are supported by futurize():
ComBat()read.degradation.matrix()