DeMixPert: Decomposed Response Modeling with Gaussian Mixtures for OOD Single-Cell Perturbation Prediction
2026-08-24 • Machine Learning
Machine LearningArtificial Intelligence
AI summaryⓘ
The authors developed DeMixPert, a method to predict how single cells respond to genetic changes that haven't been seen before. Their approach breaks down these responses into three parts: a basic cell state, changes specific to the genetic perturbation, and natural differences across cells. By using a special type of math model called a Gaussian mixture, they better capture the variability in these responses. Tests show that DeMixPert predicts gene expression changes in single cells more accurately than previous methods.
single-cell transcriptomicsgenetic perturbationGaussian mixture modelinvertible neural networkbasal cell statedistributional modelingout-of-distribution predictiongene expressionheterogeneous response
Authors
Jiawen Liu, Xuechenxiao Cao, Yutong Li, Bing Liu, Jiaming Liang, Tinghe Zhang, Xiaoqi Sheng, Hongmin Cai
Abstract
Predicting transcriptome-wide responses to unseen genetic perturbations remains a major computational challenge because accurate prediction requires recovering both perturbation-specific transcriptional shifts and heterogeneous cellular responses. Existing methods often entangle deterministic response structure with stochastic population-level variation, causing dominant shared patterns to mask weaker perturbation-specific signals and impair distributional modeling. To address these challenges, we propose \textbf{DeMixPert}, an approach for Decomposed response Modeling with Gaussian Mixtures for Out-Of-Distribution (OOD) single-cell Perturbation prediction. DeMixPert decomposes perturbation-induced changes into a basal-state-dependent systematic response, a perturbation-specific response, and population-level variation. The systematic component is derived from the basal state encoded from control-cell expression, whereas the perturbation-specific component is inferred from pretrained target embeddings for unseen-target generalization. DeMixPert models population-level variation using a Gaussian prototype Invertible Network and adaptively combines reusable Gaussian prototypes according to the basal state and perturbation condition. The resulting mixture is mapped to a condition-specific variation distribution. Sampled variations are integrated with the systematic and perturbation-specific components, followed by joint decoding with the basal state to reconstruct perturbed-cell gene expression. Experimental results show that DeMixPert effectively captures heterogeneous single-cell perturbation responses and achieves superior performance across unseen-perturbation settings. The source code is made publicly available upon publication.