近日，美国陈-扎克伯格倡议Theofanis Karaletsos及其小组研究出TranscriptFormer：跨越15亿年进化的生成细胞图谱。2026年5月7日出版的《科学》发表了这项成果。

研究团队开发了转录前（TranscriptFormer），这是一个基于12个物种15.3亿年进化过程中多达1.12亿个细胞的生成基础模型家族。小组展示了最先进的细胞类型分类性能，即使是在6.85亿年的进化中分离的物种，以及人类细胞的零射击疾病状态识别。发育轨迹、系统发育关系和细胞层次自然地出现在TranscriptFormer的表示中，而无需对这些注释进行任何明确的训练。这项工作为定量单细胞分析和比较细胞生物学建立了一个强大的框架，它们证明了细胞组织的普遍原理可以通过生命之树来学习和预测。

研究人员表示，单细胞转录组学正在彻底改变他们对细胞多样性的理解，但比较整个生命树的转录程序仍然具有挑战性。

附：英文原文

Title: TranscriptFormer: A generative cell atlas across 1.5 billion years of evolution

Author: James D. Pearce, Sara E. Simmonds, Gita Mahmoudabadi, Lakshmi Krishnan, Giovanni Palla, Ana-Maria Istrate, Alexander Tarashansky, Benjamin Nelson, Omar Valenzuela, Donghui Li, Stephen R. Quake, Theofanis Karaletsos

Issue&Volume: 2026-05-07

Abstract: Single-cell transcriptomics is revolutionizing our understanding of cellular diversity, yet comparing transcriptional programs across the tree of life remains challenging. We developed TranscriptFormer, a family of generative foundation models trained on up to 112 million cells spanning 1.53 billion years of evolution across 12 species. We demonstrate state-of-the-art performance on cell type classification, even for species separated over 685 million years of evolution, and zero-shot disease state identification in human cells. Developmental trajectories, phylogenetic relationships and cellular hierarchies emerge naturally in TranscriptFormer’s representations without any explicit training on these annotations. This work establishes a powerful framework for quantitative single-cell analysis and comparative cellular biology, thus demonstrating that universal principles of cellular organization can be learned and predicted across the tree of life.

DOI: aec8514

Source: https://www.science.org/doi/10.1126/science.aec8514