Advanced Spatial Omics with Quantum and Generative AI

Group Photo of Advanced Spatial Omics with Quantum and Generative AI members

Research

The overarching goal of our research is to leverage advanced artificial intelligence (AI) methodologies—particularly Generative AI and quantum computing—to advance cancer immunology. Motivated by the growing success of spatial omics technologies in revealing previously inaccessible insights into the complexity of the disease microenvironment, our work is driven by two core objectives: (1) to identify novel spatial biomarkers with improved predictive power for treatment response and patient prognosis; and (2) to develop AI models that enable the extraction of such spatial biomarkers directly from routinely collected histological images, an approach we term AI4HE.

The AI4HE framework, together with its associated web-based platform—the Digital Immune Reporter (DIR)—addresses several critical limitations in current spatial biomarker discovery pipelines. These include: (i) the high cost and limited accessibility of advanced spatial omics technologies; (ii) the requirement for specialized computational expertise to analyze high-plex imaging data and derive clinically meaningful biomarkers; and (iii) the typically small cohort sizes of spatial omics studies, which constrain statistical power and biomarker robustness.

Specifically, our approach offers: (i) an affordable and scalable strategy to infer additional molecular and immunological information from standard H&E images; (ii) automated reporting of clinically relevant biomarkers through a user-friendly, interactive visualization and download interface; and (iii) improved robustness and generalizability of biomarker discovery through large-scale histology-based modelling. As illustrated in the Figure 1 below.

 

Overview of the H&E-based AI molecular prediction framework, termed AI4HE

Figure 1. Overview of the H&E-based AI molecular prediction framework, termed AI4HE. Molecular labels derived from multiple sources (including pathologist annotations and molecular staining) are used to train predictive models. The trained models are then applied to unseen H&E images to infer molecular signals and cell types, enabling tissue-based multi-omics biomarker discovery by leveraging routinely collected clinical histology and large-scale public H&E datasets.

Members

 Senior Scientist LAU Mai Chan   |    [View Bio]  
 Research officer  TAN Wei Kit 
 Research Officer CHEONG Jiasheng Isaac
 Research Officer ZHANG Qiao Yan Marcia

 

Selected Publications

Spatial omics-derived biomarker studies

  1. Lau MC, Borowsky J, Väyrynen JP, Haruki K, Zhao M, Dias Costa A, Gu S, da Silva A, Ugai T, Arima K, Nguyen MN, Takashima Y, Yeong J, Tai D, Hamada T, Lennerz JK, Fuchs CS, Wu CJ, Meyerhardt JA, Ogino S, Nowak JA. Tumor-immune partitioning and clustering algorithm for identifying tumor-immune cell spatial interaction signatures within the tumor microenvironment. PLoS Comput Biol. 2025 Feb 18;21(2):e1012707.

  2. Elomaa H, Härkönen J, Väyrynen SA, Ahtiainen M, Ogino S, Nowak JA, Lau MC, Helminen O, Wirta EV, Seppälä TT, Böhm J, Mecklin JP, Kuopio T, Väyrynen JP. Quantitative multiplexed analysis of IDO and ARG1 expression and myeloid cell infiltration in colorectal cancer. Mod Pathol. 2024 Feb 16:100450.

  3. Hong JH, Yong CH, Heng HL, Chan JY, Lau MC, Chen J, Lee JY, Lim AH, Li Z, Guan P, Chu PL, Boot A, Ng SR, Yao X, Wee FYT, Lim JCT, Liu W, Wang P, Xiao R, Zeng X, Sun Y, Koh J, Kwek XY, Ng CCY, Klanrit P, Zhang Y, Lai J, Tai DWM, Pairojkul C, Dima S, Popescu I, Hsieh S-Y, Yu M-C, Yeong J, Kongpetch S, Jusakul A, Loilome W, Tan P, Tan J, Teh BT. Integrative multiomics enhancer activity profiling identifies therapeutic vulnerabilities in cholangiocarcinoma of different etiologies. Gut.

  4. Meng J, Tan JYT, Joseph CR, Ye J, Lim JCT, Goh D, Xue Y, Lim X, Koh VCY, Wee F, Tay TKY, Chan JY, Ng CCY, Iqbal J, Lau MC, Lim EH, Toh HC, Teh BT, Dent RA, Tan PH, Joe Yeong YPS. The prognostic value of CD39 as a marker of tumor-specific T cells in triple-negative breast cancer in Asian women. Lab. Invest. 2023. 100303.

AI4HE studies

  1. Azam AB, Wee F, Väyrynen JP, Yim WW, Xue YZ, Chua BL, Lim JCT, Somasundaram AC, Tan DSW, Takano A, Chow CY, Khor LY, Lim TKH, Yeong J, Lau MC, Cai Y. Training immunophenotyping deep learning models with the same-section ground truth cell label derivation method improves virtual staining accuracy. Front. Immunol. 2024 Jun 28;15:1404640.

  2. Lehner Laurens, Wei Kit Tan, Isaac Jiasheng Cheong, Wei Lin Tang, Raymond Chong Wei Liang, Olaf Rötzschke, Juha P. Väyrynen, Shuji Ogino, and Mai Chan Lau. Abstract 1102 Image-based unsupervised learning of tumor regions in spatial omics. (2025). Journal for ImmunoTherapy of Cancer 2024;12. DOI: 10.1136/jitc-2025-SITC2025.1102

  3. Colwyn Jia Kang Lai, Wei Kit Tan, Marcia Zhang, Timothy Wang, Felicia Wee, Ai Lin Wang, Solomonraj Wilson, Tomotaka Ugai, Joe Sheng Yeong Poh, Jonathan A. Nowak, Shuji Ogino, Juha P. Väyrynen, Mai Chan Lau.  Abstract 6316: Predictive performance comparison of foundational and CNN models for single-cell immune profiling. Cancer Res 15 April 2025; 85 (8_Supplement_1): 6316. https://doi.org/10.1158/1538-7445.AM2025-6316

  4. Ho ZY, Yim WW, Wee F, et al1221 Deciphering cellular features: StyleGAN for enhancing H&E cell classification. Journal for ImmunoTherapy of Cancer 2024;12 DOI: 10.1136/jitc-2024-SITC2024.1221

     

LAU MAI CHAN

close

 

 BII - Lau Mai ChanLAU Mai Chan 
Senior Scientist
Email:  Lau_Mai_Chan@a-star.edu.sg
Research Group: Advanced Spatial Omics with Quantum and Generative AI

Mai Chan Lau earned her PhD from the National University of Singapore, focusing on high-performance GPU research. She began her postdoctoral training at the Singapore Immunology Network, focusing on single-cell immunology bioinformatics. She then moved to Boston, USA, where she leveraged Al for Molecular Pathological Epidemiology studies, using tissue-based multi-marker technology at Harvard Medical School’s teaching hospitals. In 2021, she returned to Singapore and joined the Institute of Molecular and Cell Biology, expanding her research to spatial transcriptomics and Al. In 2022, she established her own laboratory at the Bioinformatics Institute, concentrating on advanced spatial multi-omics studies using Al and quantum computing. Concurrently, she leads the Computational Immunology Platform at the Singapore Immunology Network.

 

GROUP MEMBERS

 Research officer  TAN Wei Kit 
 Research Officer CHEONG Jiasheng Isaac
 Research Officer ZHANG Qiao Yan Marcia

 

A*STAR celebrates International Women's Day

A*STAR Women In Science

From groundbreaking discoveries to cutting-edge research, our researchers are empowering the next generation of female science, technology, engineering and mathematics (STEM) leaders.

Get inspired by our #WomeninSTEM