Harmony

Harmony is a general-purpose R package with an efficient algorithm for integrating multiple data sets. It is especially useful for large single-cell datasets such as single-cell RNA-seq.

Harmony is:

• Fast: Analyze thousands of cells on your laptop.
• Sensitive: Different cell types may be present or absent in each batch.
• Accurate: Integrate cells from multiple donors, tissues – even different technologies.

Getting started

See how to use Harmony with your data and integrate it into your analysis pipeline.

Read →

Advanced tutorial

Find out more about the internal data structures and algorithm details in this tutorial.

Read →

Animation

Visualize how Harmony aligns single-cell RNA-seq datasets from three different donors.

Read →

Installation

The easiest way to get Harmony is to install it from Github:

# install.packages("devtools")
devtools::install_github("immunogenomics/harmony")

Harmony has been tested on R versions >= 3.4 on Linux, macOS, and Windows.

Usage

Run the HarmonyMatrix() function on your PCs from principal component analysis:

library(harmony)

harmonized_pcs <- HarmonyMatrix(
  data_mat  = pcs,       # Matrix with coordinates for each cell (row) along many PCs (columns)
  meta_data = meta_data, # Dataframe with information for each cell (row)
  vars_use  = "dataset", # Column in meta_data that defines dataset for each cell
  do_pca    = FALSE      # Since we are providing PCs, do not run PCA
)

Citation

If you use Harmony for published work, please cite our manuscript:

Fast, sensitive, and accurate integration of single cell data with Harmony

Ilya Korsunsky, Jean Fan, Kamil Slowikowski, Fan Zhang, Kevin Wei, Yuriy Baglaenko, Michael Brenner, Po-Ru Loh, Soumya Raychaudhuri

bioRxiv 2019. doi.org/10.1101/461954

We will share the code needed to reproduce results from the manuscript at https://github.com/immunogenomics/harmony2019.