MAD: Microenvironment-Aware Distillation -- A Pretraining Strategy for Virtual Spatial Omics from Microscopy

TL;DR

MAD is a novel pretraining strategy that learns cell and microenvironment embeddings from microscopy images, enabling accurate virtual spatial omics and biological insights without extensive labels.

Contribution

MAD introduces a dual-view self-distillation approach to encode single-cell identity and tissue context, outperforming larger models on various biological prediction tasks.

Findings

State-of-the-art performance on cell subtyping and transcriptomic prediction

Outperforms larger foundation models with similar parameters

Effectively captures cellular diversity within tissues

Abstract

Bridging microscopy and omics would allow us to read molecular states from images-at single-cell resolution and tissue scale-without the cost and throughput limits of omics technologies. Self-supervised pretraining offers a scalable approach with minimal labels, yet how to encode single-cell identity within tissue environments-and the extent of biological information such models can capture-remains an open question. Here, we introduce MAD (microenvironment-aware distillation), a pretraining strategy that learns cell-centric embeddings by jointly self-distilling the morphology view and the microenvironment view of the same indexed cell into a unified embedding space. Across diverse tissues and imaging modalities, MAD achieves state-of-the-art prediction performance on downstream tasks including cell subtyping, transcriptomic prediction, and bioinformatic inference. MAD even outperforms foundation models with a similar number of model parameters that have been trained on substantially larger datasets. These results demonstrate that MAD's dual-view joint self-distillation effectively captures the complexity and diversity of cells within tissues. Together, this establishes MAD as a general tool for representation learning in microscopy, enabling virtual spatial omics and biological insights from vast microscopy datasets.