Fast, sensitive and flexible integration of single-cell data
Harmony is one of the most widely used algorithms for integrating single-cell datasets, iteratively removing technical batch differences while preserving subtle biological cell states.
Iterative maximum diversity clustering to align batch-specific distributionsExtremely fast execution scaling linearly with millions of cellsAvailable in both R and Python (harmonypy) packagesIntegrated natively into Seurat and Scanpy pipelines
Harmony is one of the most widely used algorithms for integrating single-cell datasets, iteratively removing technical batch differences while preserving subtle biological cell states.
Information checked against an official source; not a hands-on test. Source · Last reviewed: 20/09/2026, 11:10:15
Key Features
Iterative maximum diversity clustering to align batch-specific distributions
Extremely fast execution scaling linearly with millions of cells
Available in both R and Python (harmonypy) packages
Integrated natively into Seurat and Scanpy pipelines
Academic Context & Research Evidence
Biological & Workflow Fit
Biological Application
Cross-tissue cell atlas integration, patient cohort harmonization, and multi-donor studies
Research Workflow
Uncorrected PCA matrix -> Harmony batch alignment -> corrected embedding for clustering and UMAP
Use this citation format when referencing Harmony in scientific publications and benchmark papers.
@software{harmony_2026,
title = {{Harmony}},
author = {{Broad Institute / Harvard Medical School (Raychaudhuri Lab)}},
year = {2026},
url = {https://github.com/immunogenomics/harmony},
note = {Indexed on aibioatlas - AI for Biology and Drug Discovery}
}
Peer-Reviewed Literature & Preprints
Live scientific citations streamed from Europe PMC and PubMed for Harmony.
⏳ Fetching real-time literature from Europe PMC & PubMed...
Technical / Product Information
Missing values mean the catalog has no recorded information. They do not mean a feature is absent.
Entry typeSoftware Library (R / Python)
Access modeOpen Source
AI roleIterative Dimensionality Alignment
Input dataNot recorded
Output dataNot recorded
Licence conditionsGPL-3.0
Commercial eligibilityPermissive open source release