Menu
Sign In Search Podcasts Charts Entities Add Podcast API Pricing
Podcast Image

聊聊Sci

083-基于表型特征和度量学习的细胞谱系树重建

31 Aug 2025

Description

这项研究介绍了 CellTreeQM,这是一种基于深度学习的新方法,旨在利用单细胞的表型特征(如基因表达)重构细胞谱系树。文章指出,传统的基因工程谱系追踪方法在许多生物体中不可行或受到伦理限制,而现代单细胞技术能够测量丰富的高维分子数据。CellTreeQM 将谱系重构视为一个度量学习问题,通过学习一个能够反映树状几何特性的嵌入空间来实现。研究人员在合成数据和秀丽隐杆线虫的真实谱系数据集上对该方法进行了基准测试,结果表明即使在有限监督和数据约束下,CellTreeQM 也能有效恢复谱系结构,并且优于其他对比学习基线方法。References: Kuang D, Qiu G, Kim J. Reconstructing Cell Lineage Trees from Phenotypic Features with Metric Learning[J]. arXiv preprint arXiv:2503.13925, 2025.

Audio
Featured in this Episode

No persons identified in this episode.

Transcription

No transcription available yet

Help us prioritize this episode for transcription by upvoting it.

0 upvotes
🗳️ Sign in to Upvote

Popular episodes get transcribed faster

Comments

There are no comments yet.

Please log in to write the first comment.