Semi-supervised integration of single-cell transcriptomics data.

Autor: Andreatta M; Department of Oncology, Lausanne Branch, Ludwig Institute for Cancer Research, CHUV and University of Lausanne, 1011, Lausanne, Switzerland.; AGORA Cancer Research Center, 1005, Lausanne, Switzerland.; Swiss Institute of Bioinformatics, 1015, Lausanne, Switzerland., Hérault L; Department of Oncology, Lausanne Branch, Ludwig Institute for Cancer Research, CHUV and University of Lausanne, 1011, Lausanne, Switzerland.; AGORA Cancer Research Center, 1005, Lausanne, Switzerland.; Swiss Institute of Bioinformatics, 1015, Lausanne, Switzerland., Gueguen P; Department of Oncology, Lausanne Branch, Ludwig Institute for Cancer Research, CHUV and University of Lausanne, 1011, Lausanne, Switzerland.; AGORA Cancer Research Center, 1005, Lausanne, Switzerland.; Swiss Institute of Bioinformatics, 1015, Lausanne, Switzerland., Gfeller D; Department of Oncology, Lausanne Branch, Ludwig Institute for Cancer Research, CHUV and University of Lausanne, 1011, Lausanne, Switzerland.; AGORA Cancer Research Center, 1005, Lausanne, Switzerland.; Swiss Institute of Bioinformatics, 1015, Lausanne, Switzerland., Berenstein AJ; Laboratorio de Biología Molecular, División Patología, Instituto Multidisciplinario de Investigaciones en Patologías Pediátricas (IMIPP), CONICET-GCBA, Buenos Aires, C1425EFD, Argentina., Carmona SJ; Department of Oncology, Lausanne Branch, Ludwig Institute for Cancer Research, CHUV and University of Lausanne, 1011, Lausanne, Switzerland. santiago.carmona@unil.ch.; AGORA Cancer Research Center, 1005, Lausanne, Switzerland. santiago.carmona@unil.ch.; Swiss Institute of Bioinformatics, 1015, Lausanne, Switzerland. santiago.carmona@unil.ch.
Jazyk: angličtina
Zdroj: Nature communications [Nat Commun] 2024 Jan 29; Vol. 15 (1), pp. 872. Date of Electronic Publication: 2024 Jan 29.
DOI: 10.1038/s41467-024-45240-z
Abstrakt: Batch effects in single-cell RNA-seq data pose a significant challenge for comparative analyses across samples, individuals, and conditions. Although batch effect correction methods are routinely applied, data integration often leads to overcorrection and can result in the loss of biological variability. In this work we present STACAS, a batch correction method for scRNA-seq that leverages prior knowledge on cell types to preserve biological variability upon integration. Through an open-source benchmark, we show that semi-supervised STACAS outperforms state-of-the-art unsupervised methods, as well as supervised methods such as scANVI and scGen. STACAS scales well to large datasets and is robust to incomplete and imprecise input cell type labels, which are commonly encountered in real-life integration tasks. We argue that the incorporation of prior cell type information should be a common practice in single-cell data integration, and we provide a flexible framework for semi-supervised batch effect correction.
(© 2024. The Author(s).)
Databáze: MEDLINE
Externí odkaz: