MIMIC: A Generative Multimodal Foundation Model for Biomolecules

2026-04-27

AI Summary

Biological macromolecule function is jointly governed by multimodal factors—including sequence, structure, regulation, evolution, and cellular context—yet existing foundation models are often confined to single modalities or fixed tasks. To address this limitation, we propose MIMIC, a generative multimodal foundation model built upon the newly curated, aligned dataset LORE. MIMIC employs a split-track encoder–decoder architecture that accepts any subset of observed modalities as conditional input, enabling unified generative modeling across nucleic acids, proteins, structures, evolutionary profiles, regulatory signals, and semantic context. The model supports conditional prediction, allele- and isoform-aware inference, and constrained sequence design, while dynamically incorporating experimental context as semantic conditioning. MIMIC achieves state-of-the-art performance on downstream tasks such as RNA splicing prediction, successfully designs non-reverting editing strategies to rescue HBB splicing mutations, and generates high-confidence protein sequences binding PD-L1 and hACE2.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: Generation
• Computer Vision: Multi-modal Vision

Application Category

• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Biological function emerges from coupled constraints across sequence, structure, regulation, evolution, and cellular context, yet most foundation models in biology are trained within one modality or for a fixed forward task. We present MIMIC, a generative multimodal foundation model trained on our newly curated and aligned dataset, LORE, linking nucleic acid, protein, evolutionary, structural, regulatory, and semantic/contextual modalities within partially observed biomolecular states. MIMIC uses a split-track encoder-decoder architecture to condition on arbitrary subsets of observed modalities and reconstruct or generate missing components of molecular state across the genome, transcriptome, and proteome. Multimodal conditioning consistently improves MIMIC's sequence reconstruction relative to sequence-only inputs, while its learned representations enable state-of-the-art performance on RNA and protein downstream tasks. MIMIC achieves state-of-the-art splicing prediction, and its joint generative formulation enables isoform-aware inference that further improves performance. Beyond prediction, the same generative framework supports constrained design. For RNA, MIMIC identifies corrective edits in a clinically relevant HBB splice-disrupting mutation without reverting it by using evolutionary and structural signals. For proteins, jointly conditioning on shape and surface chemistry of PD-L1 and hACE2 binding sites produces diverse, high-confidence sequences with strong in silico support for target binding. Finally, MIMIC uses experimental context as semantic conditioning to model assay-dependent RNA chemical probing, rather than treating context as a fixed output. Together, these results position MIMIC's aligned multimodal generative modeling as a strong foundation for unifying representation learning, conditional prediction, and constrained biomolecular design within a single model.

Problem

• multimodal
• biomolecules
• foundation model
• generative modeling
• molecular representation

Innovation

• multimodal foundation model
• generative modeling
• biomolecular design
• conditional reconstruction
• aligned biological modalities