A multimodal cell-free RNA language model for liquid biopsy applications

Abstract

Abstract Cell-free RNA (cfRNA) profiling has emerged as a powerful tool for non-invasive disease detection, but its application is limited by data sparsity and complexity, especially in settings with constrained sample availability. We introduce Exai-1, a multimodal, transformer-based generative foundation model that integrates RNA sequence embeddings with cfRNA abundance data to capture biologically meaningful representations of circulating RNAs. By leveraging both sequence and expression modalities, Exai-1 captures a biologically meaningful latent structure of cfRNA profiles. Pretrained on over 306 billion tokens from 8,339 samples, Exai-1 enhances signal fidelity, reduces technical noise and improves disease detection by generating synthetic cfRNA profiles. We show that self-attention and variational inference are particularly important for the preservation of biological signals and contextual relationships. Additionally, Exai-1 facilitates cross-biofluid translation and assay compatibility through disentangling biological signals from confounders. By uniting sequence-informed embeddings with cfRNA expression patterns, Exai-1 establishes a transfer learning foundation for liquid biopsy, offering a scalable and adaptable framework for next-generation cfRNA-based diagnostics.

Affiliated Institutions

• Palo Alto Institute US
• Arc Research Institute
• University of California, Davis US
• University of California, San Francisco US

Related Publications