Harnessing the deep learning power of foundation models in single-cell omics

IF 81.3 1区生物学 Q1 CELL BIOLOGY

Nature Reviews Molecular Cell Biology Pub Date : 2024-06-26 DOI:10.1038/s41580-024-00756-6

Qin Ma, Yi Jiang, Hao Cheng, Dong Xu

引用次数: 0

Abstract

Foundation models hold great promise for analyzing single-cell omics data, yet various challenges remain that require further advancements. In this Comment, we discuss the progress, limitations and best practices in applying foundation models to interrogate data and improve downstream tasks in single-cell omics. This Comment discusses the progress, limitations and best practices in applying foundation models to single-cell omics data.

查看原文本刊更多论文

在单细胞全息研究中利用基础模型的深度学习能力。

基础模型在分析单细胞组学数据方面大有可为，但仍存在各种挑战，需要进一步改进。在本评论中，我们将讨论应用基础模型查询数据和改进单细胞组学下游任务的进展、局限性和最佳实践。本评论讨论了将基础模型应用于单细胞组学数据的进展、局限性和最佳实践。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Nature Reviews Molecular Cell Biology 生物-细胞生物学

CiteScore

173.60

自引率

0.50%

发文量

118

审稿时长

6-12 weeks

期刊介绍： Nature Reviews Molecular Cell Biology is a prestigious journal that aims to be the primary source of reviews and commentaries for the scientific communities it serves. The journal strives to publish articles that are authoritative, accessible, and enriched with easily understandable figures, tables, and other display items. The goal is to provide an unparalleled service to authors, referees, and readers, and the journal works diligently to maximize the usefulness and impact of each article. Nature Reviews Molecular Cell Biology publishes a variety of article types, including Reviews, Perspectives, Comments, and Research Highlights, all of which are relevant to molecular and cell biologists. The journal's broad scope ensures that the articles it publishes reach the widest possible audience.