LIFT:用潜空间流匹配从单细胞快照预测细胞命运
LIFT: Latent Flow Matching for Probabilistic Cell Fate Dynamics with Single Cell Snapshots
做单细胞或谱系示踪研究的朋友可以看看这篇。它把条件流匹配用到了细胞命运预测上,一次前向就能采样,蒙特卡洛跑大批量很划算。比传统只拟合终端群体分布的方法多了概率化输出,方法细节写得也比较完整。
一篇来自 arXiv 的新论文提出 LIFT,一个基于潜空间流匹配的概率性细胞状态生成框架,用于从谱系示踪数据中学习细胞命运动力学。核心问题在于:谱系示踪只能通过克隆祖先连接细胞状态,但通常无法给出唯一的一对一对应关系,因此模型必须学习条件化的结局分布,而不是只拟合终端群体分布。
技术机制上,LIFT 通过变分推断把多种可能的结局编码进潜空间,并设计条件先验使模型仅凭初始状态即可采样;解码器采用 Mean Flow 结构,每次采样只需一次解码器前向计算,从而支持大规模蒙特卡洛采样来估计细胞命运概率,论文还附带了条件预测误差的分析。
在合成数据与真实数据集上的实验显示,LIFT 能同时恢复群体分布和细胞命运概率。对做细胞动力学建模、谱系追踪分析的研究者来说,这种一对多关系建模加上高效采样的组合,与以往只输出终端分布的方法相比更适合做概率化的命运推断。
LIFT: Latent Flow Matching for Probabilistic Cell Fate Dynamics with Single Cell Snapshots
Lineage tracing links cell states through clonal ancestry, but does not generally identify unique source-descendant pairings. Learning from such data therefore requires a model of conditional outcomes, rather than only the terminal population distribution. We introduce LIFT, a latent flow framework for probabilistic cell-state generation from lineage-informed couplings. Variational inference encodes alternative outcomes in a latent space, while a conditional prior enables sampling from the initial state alone. A Mean Flow decoder predicts future cell state directly, requiring one decoder evaluation per sampling, which makes it convenient to predict cell fate through large scale Monte-Carlo sampling. This construction combines a representation of one-to-many relationships with efficient sampling and an analysis of conditional prediction error. Experiments on synthetic and real datasets demonstrate LIFT's ability to recover both population distribution and cell fate probabilities.
本条情报归属 AITOP「论文」核心追踪栏目。涉及核心实体与技术议题包括:流匹配、单细胞测序、变分推断、细胞命运预测,读者可通过下方标签进入对应专题页查阅历史脉络与关联对比。
信源核验说明:本文原始线索采编自「arXiv cs.LG」(@Bowen Yang, Junda Ying, Yuxuan Wang, Peijie Zhou, Lei Zhang),经 AITOP 多维评分管线结构化提炼。