Work

scExtract: leveraging large language models for fully automated single-cell RNA-seq data annotation and prior-informed multi-dataset integration

Year: 2025

Type: article

Abstract: Single-cell RNA sequencing has revolutionized cellular heterogeneity research, but analyzing the abundance of unannotated public datasets remains challenging. We present scExtract, a framework leveraging large language models to automate scRNA-seq data analysis from preprocessing to annotation and integration. scExtract extracts information from research articles to guide data processing, outperforming existing reference transfer methods in benchmarks. We introduce scanorama-prior and cellhint-prior, which incorporate prior annotation information for improved batch correction while preserving biological diversities. We demonstrate scExtract's utility by integrating 14 datasets to create a comprehensive human skin atlas of 440,000 cells.

Source: Genome biology

DOI: 10.1186/s13059-025-03639-x

Authors: Yuxuan Wu, Fuchou Tang

Institutions: Peking University, Ministry of Education of the People's Republic of China

Language: English

Topic: Single-cell and spatial transcriptomics

Subfield: Molecular Biology

Field: Biochemistry, Genetics and Molecular Biology

Domain: Life Sciences

Open Access status: gold

Estimated APC paid: $5,490

Funders: Natural Science Foundation of Beijing Municipality, Peking University