AROMA: Augmented Reasoning Over a Multimodal Architecture for Virtual Cell Genetic Perturbation Modeling

2026-04-22

AI Summary

Existing virtual cell modeling approaches suffer from unconstrained inference, limited interpretability of predictions, and weak alignment between retrieval signals and regulatory topology. To address these limitations, this work proposes a knowledge-driven, multimodal reasoning framework that integrates textual evidence, protein sequences, and regulatory graph topology through a two-stage optimization strategy. The authors also introduce PerturbReason, a large-scale dataset comprising 498k samples, along with two domain-specific knowledge graphs. The proposed method substantially improves the accuracy, interpretability, and generalization of gene perturbation effect prediction, outperforming current state-of-the-art methods across multiple cell lines. Notably, it demonstrates robust performance in zero-shot settings on unseen cell lines and in long-tail scenarios characterized by sparse biological knowledge.

Technology Category

• Knowledge Representation and Reasoning: Computational Complexity of Reasoning
• Reasoning under Uncertainty: Graphical Models
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Virtual cell modeling predicts molecular state changes under genetic perturbations in silico, which is essential for biological mechanism studies. However, existing approaches suffer from unconstrained reasoning, uninterpretable predictions, and retrieval signals that are weakly aligned with regulatory topology. To address these limitations, we propose AROMA, an Augmented Reasoning Over a Multimodal Architecture for virtual cell genetic perturbation modeling. AROMA integrates textual evidence, graph-topology information, and protein sequence features to model perturbation-target dependencies, and is trained with a two-stage optimization strategy to yield predictions that are both accurate and interpretable. We also construct two knowledge graphs and a perturbation reasoning dataset, PerturbReason, containing more than 498k samples, as reusable resources for the virtual cell domain. Experiments show that AROMA outperforms existing methods across multiple cell lines, and remains robust under zero-shot evaluation on an unseen cell line, as well as in knowledge-sparse, long-tail scenarios. Overall, AROMA demonstrates that combining knowledge-driven multimodal modeling with evidence retrieval provides a promising pathway toward more reliable and interpretable virtual cell perturbation prediction. Model weights are available at https://huggingface.co/blazerye/AROMA. Code is available at https://github.com/blazerye/AROMA.

Problem

• virtual cell modeling
• genetic perturbation
• unconstrained reasoning
• uninterpretable predictions
• regulatory topology

Innovation

• multimodal reasoning
• genetic perturbation modeling
• knowledge graph
• interpretable AI
• virtual cell