Cross-modality representation and multi-sample integration of spatially resolved omics data

News

2024-06-15: PRESENT v1.0.0 is released and souce code is available!

PRESENT is now available on GitHub 2024-06-15

Overview

Spatially resolved sequencing technologies have revolutionized our understanding of biological regulatory processes within the microenvironment by accessing the states of genomic regions, genes and proteins as well as spatial coordinates of cells. However, discrepancies between different modalities and samples hinder the analysis of spatial omics data, necessitating the development of advanced computational methods. In this article, we propose PRESENT, an effective and scalable contrastive learning framework, for the cross-modality representation and multi-sample integration of spatial multi-omics, epigenomics and transcriptomics data. Through comprehensive experiments on spatial datasets, PRESENT demonstrates superior performance across various species, tissues, and technologies. Specifically, PRESENT effectively integrates spatial dependency and omics information simultaneously, facilitating the detection of spatially functional domains and the exploration of biological regulatory mechanisms. Furthermore, PRESENT can be extended for the integrative analysis of tissue samples across different dissected regions or developmental stages, promoting the identification of hierarchical structures from systematic and spatiotemporal perspectives.

Citation

Zhen Li, Xuejian Cui, Xiaoyang Chen, Zijing Gao, Yuyao Liu, Yan Pan, Shengquan Chen and Rui Jiang. “Cross-modality representation and multi-sample integration of spatially resolved omics data.” Preprint at bioRxiv https://doi.org/10.1101/2024.06.10.598155 (2024).