CellxPert: Inference-Time MCMC Steering of a Multi-Omics Single-Cell Foundation Model for In-Silico Perturbation

2026-04-30

AI Summary

Existing single-cell foundation models struggle to accurately simulate gene perturbation effects and lack a unified framework for modeling multi-omics and spatial data. This work proposes a scalable, unified foundation model that integrates a multimodal Transformer architecture with LoRA fine-tuning and, for the first time, combines Metropolis-Hastings MCMC sampling with masked conditional distributions to enable continuous, interpretable simulation of transcriptional responses under perturbations—avoiding out-of-distribution artifacts caused by conventional token-based manipulations. The model supports 154 fine-grained cell-type annotations and incorporates spatial multi-omics data such as MERFISH and mass spectrometry imaging. It significantly outperforms current methods across multiple benchmarks—including PBMC68K and Replogle Perturb-seq—in tasks spanning cell annotation, perturbation response prediction, and multi-omics integration.

Abstract

In this work, we introduce CellxPert, a scalable multimodal foundation model that unifies single-cell and spatial multi-omics within a common representation space. CellxPert jointly encodes transcriptomic (scRNA-seq), chromatin-accessibility (ATAC-seq), and surface-proteomic (CITE-seq) measurements, while directly incorporating MERFISH and imaging mass-cytometry data as 2D or 3D spatial-visual layers. CellxPert facilitates four key downstream tasks out of the box: (i) cell-type annotation across a broad ontology of 154 largely overlapping identities -- the largest label space addressed to date and a stringent test of fine-grained discrimination, (ii) efficient fine-tuning using Low Rank Adaptation (LoRA), (iii) genome-wide transcriptomic response prediction to in-silico perturbations (ISP), and (iv) seamless multi-omic integration across various assays and platforms. Unlike current single-cell foundation models, which approximate gene perturbations by deleting or reordering tokenized gene expression ranks, CellxPert employs a Metropolis-Hastings sampler whose proposal kernel uses the model's masked conditional distributions to transition to new transcriptomic states conditioned on the perturbed genes. This Markov-chain procedure mitigates out-of-distribution artifacts introduced by abrupt token manipulation and produces trajectories that are biologically interpretable. Evaluations on PBMC68K, Replogle Perturb-seq, Systema, and BMMC benchmarks show that CellxPert surpasses classical and state-of-the-art baselines in cell-type annotation, perturbation response prediction, and multi-omic integration.

Problem

• in-silico perturbation
• single-cell multi-omics
• transcriptomic response prediction
• foundation model
• out-of-distribution artifacts

Innovation

• in-silico perturbation
• MCMC steering
• multi-omics foundation model
• Metropolis-Hastings sampling
• spatial single-cell integration