harmony

Harmony removes batch effects in single-cell data by applying soft clustering and iterative alignment to align datasets from different sources while preserving biological signals. It enables integrated analyses across conditions and technologies.

You can integrate multiple single-cell datasets in a Seurat workflow by loading your Seurat objects in R and running the RunHarmony function, which supports multiple grouping variables for multi-dataset integration across batches, technologies, and donors.

Yes, Harmony works with Seurat objects in R and can be paired with Python tools such as scanorama or bbknn for cross-language pipelines, allowing flexible integration workflows across different programming environments.

Harmony scales to large single-cell datasets containing millions of cells, applying soft clustering and iterative alignment to remove batch effects while maintaining biological signals across multi-dataset integration scenarios.

Harmony provides tunable parameters including theta, lambda, sigma, and max.iter.harmony for adjusting batch correction in single-cell data, allowing you to control the integration of datasets from different sources while preserving biological variation.

You need single-cell data integration when combining datasets from different sources, conditions, technologies, or donors. Harmony aligns these datasets by removing batch effects while preserving biological signals, enabling accurate integrated analyses across experiments.