Speaker: 林志翔副教授，香港中文大学

Inviter: 王勇研究员

Title: Representation learning for single-cell multi-omics and spatial-omics data

Language: English&Chinese

Time & Venue: 2024.07.18 15:00-16:00 思源楼625会议室

Abstract: In this talk, I will share two of our recent works.

First, I will talk about JSNMF, which is a versatile toolkit for the integrative analysis of transcriptomic and epigenomic data profiled from the same cell. The core of JSNMF is an unsupervised method based on jointly semi-orthogonal nonnegative matrix factorization, where it assumes different latent variables for the two molecular modalities, and integrates the information of transcriptomic and epigenomic data with consensus graph fusion, which better tackles the distinct characteristics and levels of noise across different molecular modalities in single-cell multi-omics data. We applied JSNMF to single-cell multi-omics datasets from different tissues and different technologies. The results demonstrate the superior performance of JSNMF in clustering and data visualization of the cells. JSNMF also allows joint analysis of multiple single-cell multi-omics experiments and single-cell multi-omics data with more than two modalities profiled on the same cell. JSNMF also provides rich biological insight on the markers, cell-type-specific region–gene associations and the functions of the identified cell subpopulation.

Second, I will talk about SGCAST, an efficient auto-encoder framework to identify spatial domains. SGCAST adopts a symmetric graph convolutional auto-encoder to learn aggregated latent embeddings via integrating the gene expression similarity and the proximity of the spatial spots. This framework in SGCAST enables a mini-batch training strategy, which makes SGCAST memory-efficient and scalable to high-resolution spatial transcriptomic data with a large number of spots. SGCAST improves the overall accuracy of spatial domain identification on benchmarking data. We also validated the performance of SGCAST on spatial transcriptomic datasets at various scales across multiple platforms. Our study illustrates the superior capacity of SGCAST on analyzing spatial transcriptomic data.