Histology imaging is an important tool in medical diagnosis and research, enabling the examination of tissue structure and composition at the microscopic level. Understanding the underlying molecular mechanisms of tissue architecture is critical in uncovering disease mechanisms and developing effective treatments.Gene expression profiling provides insight into the molecular processes underlying tissue architecture, but the process can be time-consuming and expensive. We present BLEEP (Bi-modaL Embedding for Expression Prediction), a bi-modal embedding framework capable of generating spatially resolved gene expression profiles of whole-slide Hematoxylin and eosin (H&E) stained histology images. BLEEP uses contrastive learning to construct a low-dimensional joint embedding space from a reference dataset using paired image and expression profiles at micrometer resolution. With this approach, the gene expression of any query image patch can be imputed using the expression profiles from the reference dataset.

The rapid development of spatial transcriptomics(ST) enables the measurement of gene expression at spatial resolution, making it possible to simultaneously profile the gene expression, spatial locations of spots, and the matched histopathological images. However, the cost for collecting ST data is much higher than acquiring histopathological images, and thus several studies attempt to predict the gene expression on ST by leveraging their corresponding histopathological images. Most of the existing image-based gene prediction models treat the prediction task on each spot of ST data independently, which ignores the spatial dependency among spots. In addition, while the histology images share phenotypic characteristics with the ST data, it is still challenge to extract such common information to help align paired image and expression representations. To address the above issues, we propose a Multi-view Graph Contrastive Learning framework with HSICbottleneck Regularization(ST-GCHB) aiming at learning shared representation to help impute the gene expression of the queried imaging spots by considering their spatial dependency. Specifically, ST-GCHB firstly adopts the intra-modal graph contrastive learning (GCL) to learn meaningful imaging and genomic features of spots by considering their spatial characteristics. Then, to reduce the redundancy for the extracted features of different modalities, we also add a HSIC-bottleneck regularization term in the GCL to enhance the efficiency of our model. Finally, an cross-modal contrastive learning strategy is applied to align the multi-modal data for imputing the spatially resolved gene expression data from the histopathological images.We conduct experiments on the dorsolateral prefrontal cortex (DLPFC) dataset and observe a significant improvement compared to the existing approaches. These results show the viability and effectiveness of our ST-GCHB for predicting molecular signatures of tissues from the histopathological images.