Single-Cell Multi/Spatial-Omics and Precision Technology

1. Advanced AI Solutions for Spatial Omics:

Spatial omics combines omics measurements with spatial context, offering an unprecedented opportunity for scientists to explore tissue complexity and cell-cell interactions. However, this approach also presents challenges in data analysis. Specifically, incorporating spatial information necessitates a spatially aware analytical approach. Dr. Chen has developed several AI-driven tools to address these specific challenges in spatial omics.

SEDR: Spatial transcriptomics is often characterized by high levels of noise and sparsity, with frequent drop-out events. SEDR is a tool designed to address these issues by effectively removing noise and imputing missing data. It combines variational graph autoencoders with masked self-supervised learning, offering a crucial advancement in enhancing the accuracy of spatial transcriptomics data.

STAMP: Spatial transcriptomics produces high-dimensional gene expression measurements with their spatial context. Obtaining a biologically meaningful low-dimensional representation of such data is crucial for effective interpretation and downstream analysis. STAMP integrates topic modeling with deep generative models to provide an interpretable, spatially aware dimensionality reduction tool. It facilitates the identification of biologically meaningful patterns in spatial transcriptomics data. This tool has been applied to various tissues, including the lung, brain, and embryo, and is soon to be published in Nature Methods.

GraphST: A novel approach using graph neural networks and contrastive learning for the spatial integration and clustering of transcriptomics data. This method enhances the understanding of complex spatial relationships within tissues and has been featured in Nature Communications.

SpatialGlue: A cutting-edge tool designed to decode fine-grained spatial domains from multi-omics data using graph attention neural networks. This tool represents a significant advancement in spatial biology and has been published in Nature Methods.

2. AI for drug discovery:

Dr. Chen’s lab is now expanding into drug discovery, focusing on developing large foundational models tailored for single-cell and spatial biology. By fine-tuning these models with specific datasets, particularly from Asian biology contexts such as hepatocellular carcinoma (HCC) and nasopharyngeal cancer (NPC), her research aims to revolutionize marker discovery, drug response predictions, and personalized medicine.

3. The DISCO Data Hub (https://immunesinglecell.org/):

Since its inception in 2009, single-cell RNA-seq (scRNA-seq) techniques have rapidly advanced, leading to increased throughput and reduced costs. As the number of published studies grows, efficient data integration and retrieval become increasingly crucial. DISCO (in-Depth Integration of Single-Cell Omics) was developed as a comprehensive repository for scRNA-seq datasets, covering a wide range of tissues and disease states. Currently, DISCO houses over 100 million single-cell profiles from more than 16,000 samples, complete with curated metadata and refined cell type annotations based on a harmonized reference. By providing a centralized platform for single-cell omics data, DISCO plays a crucial role in advancing precision medicine. Researchers can use the hub to identify novel biomarkers, understand disease heterogeneity, and develop targeted therapies.

4. The ezSingleCell Platform (https://immunesinglecell.org/ezsc/):

ezSingleCell is an interactive, user-friendly application designed for analyzing various single-cell and spatial omics data types without requiring programming knowledge. It integrates top-performing public methods for comprehensive data analysis, integration, and interactive visualization. The platform includes five modules—scRNA-seq, scATAC-seq, single-cell multiomics, batch integration, and spatial transcriptomics—each offering a complete workflow for specific data types or tasks. It also enables seamless interaction between modules within a unified interface. By addressing the inefficiencies and complexities of disparate tools, ezSingleCell provides a comprehensive solution for omics analysis and has been widely recognized in the research community, including being featured in Nature Communications.

• Chengwei Zhong, Kok Siong Ang, Jinmiao Chen, “ Interpretable spatially aware dimension reduction of spatial transcriptomics with STAMP”, Nature Methods (accept in principle), https://www.researchsquare.com/article/rs-3234073/v1, 2024.
• Yahui Long, Kok Siong Ang, Sha Liao, Raman Sethi, Yang Heng, Chengwei Zhong, Hang Xu, Nazihah Husna, Min Jian, Lai Guan Ng, Ao Chen, Nicholas RJ Gascoigne, Xun Xu, Jinmiao Chen, “ Deciphering spatial domains from spatial multi-omics with SpatialGlue”, Nature Methods, 2024.
• Raman Sethi, Ang Kok Siong, Mengwei Li, Yahui Long, Jingjing Ling, Jinmiao Chen, “ezSingleCell: An integrated one-stop single-cell and spatial omics analysis platform for bench scientists”, Nature Communications, 2024.
• Hang Xu, Huazhu Fu, Kelvin Chong, Mengwei Li, Kok Siong Ang, Hong Kai Lee, Jingjing Ling, Ao Chen, Ling Shao, Longqi Liu, Jinmiao Chen, “Unsupervised spatially embedded deep representation of spatial transcriptomics”, Genome Medicine, 2024 (IF=15.3, citation=100).
• Yahui Long, Kok Siong Ang, Mengwei Li, Kian Long Kelvin Chong, Raman Sethi, Chengwei Zhong, Hang Xu, Zhiwei Ong, Karishma Sachaphibulkij, Ao Chen, Zeng Li, Huazhu Fu, Min Wu, Lina Hsiu Kim Lim, Longqi Liu, Jinmiao Chen, “Spatially informed clustering, integration and deconvolution of spatial transcriptomics with GraphST”, Nature Communications, 2023 (IF=17.694, citation=103).
• Mengwei Li, Xiaomeng Zhang, Kok Siong Ang, Jingjing Ling, Raman Sethi, Nicole Yee Shin Lee, Florent Ginhoux, Jinmiao Chen, “DISCO: a database of Deeply Integrated human Single-Cell Omics data”, Nucleic Acids Research, 2022 (IF=19.160, citation=71).
• Hoa Thi Nhu Tran, Kok Siong Ang, Marion Chevrier, Xiaomeng Zhang, Nicole Yee Shin Lee, Michelle Goh, Jinmiao Chen, “A benchmark of batch-effect correction methods for single-cell RNA sequencing data”, Genome Biology, 2020 (IF=17.906, citation=587).