SPEAK: Spatial Prompting with Expert Aligned Knowledge for Tissue Domain Identification in Spatial Transcriptomics
August 10 @ 1:00 pm – 2:00 pm
Statistical Bioinformatics Seminar
Speaker: Dr Xiting Yan, Yale University
This is an online event held via Zoom: https://uni-sydney.zoom.us/j/85114748391
Spatially resolved transcriptomic (SRT) data requires spatial domain identification to enable tissue microenvironment-specific downstream analyses. Here we present SPEAK (Spatial Prompting with Expert-Aligned Knowledge), a large language model (LLM)-based method to identify spatial domains from SRT data by taking advantage of the prior knowledge from both LLM and human experts. SPEAK constructs a spatial context prompt for each cell/spot based on cell types and marker genes of its neighboring cells, enabling zero-shot inference, expert-guided fine-tuning, and prototype updating through two-stage prompting. Applications to STARmap, Visium, MERFISH and Xenium datasets showed advantages of SPEAK over existing spatial domain identification methods in domain prediction accuracy, robustness to limited prior knowledge, biological interpretability, and capacity for efficient expert-guided fine-tuning with generalizability to other tissue sections.
Subscribe to our seminar mailing list
Find out more about the Statistical Bioinformatics seminar series
Dr Xiting Yan
Dr Yan is an Associate Professor from the Section of Pulmonary, Critical Care and Sleep Medicine at Yale University School of Medicine, with a secondary appointment at the Department of Biostatistics at Yale University School of Public Health. She is the Director of Data Analysis and Bioinformatics Hub at the Cent for Precision Pulmonary Medicine (P2MED). Dr Yan is a world known computational biologist, bioinformatician and biostatistician, with extensive research experiences in large scale multi-omic data analyses at both bulk and single-cell resolution. Her current research interest focus on two parts: (1) developing novel statistical and computational models to analyze large scale multi-omics and drug perturbation data to better understand disease pathogenesis and facilitate precision medicine development, and (2) understanding the heterogeneity, pathogenesis and progression of pulmonary diseases, such as asthma, idiopathic pulmonary fibrosis (IPF), sarcoidosis, chronic obstructive pulmonary disease (COPD), pediatric cystic fibrosis and so on, by tailoring statistical and computational methods based on existing biological knowledge of the diseases. She is specifically interested in development of novel analytical methods for single-cell RNA sequencing data, spatial transcriptomic data, drug perturbation data and integration of different omics data. In this talk, she will introduce a recently developed method for spatial domain identification using large language model.