Single-Cell Analysis and Deep Learning for
Cancer Precision Medicine
My lab is dedicated to advancing Single-Cell Analysis and Deep Learning for Cancer Precision Medicine. Cellular heterogeneity drives diverse functions in multicellular organisms, shaping both normal physiology and disease progression, including cancer. However, biological insights often miss this variability due to population-averaged analyses. By leveraging high-throughput single-cell multi-omics and machine learning, we focus on uncovering cell dynamics and profiling heterogeneity at an unprecedented scale. While microfluidics provides exceptional single-cell tracking, it lacks seamless integration with automated systems. Our research overcomes this by combining user-friendly microfluidics, robotic liquid handling, and autonomous computer vision for efficient single-cell assays. This automation significantly improves speed, accuracy, and reproducibility, minimizing error and human bias to propel scientific discovery. Using high-throughput single-cell data, we employ deep learning to predict cellular responses, iteratively refining treatment strategies. Specifically, our approach enables label-free cell status prediction and in silico virtual compound efficacy assessment, utilizing chemical properties, transcriptomic effects, and literature insights processed through large language models. This integrated approach aims to redefine our understanding and treatment of cancer, ultimately enhancing patient outcomes.
Single-Cell Isolation |
Cancer-Stromal Engulfment |
Single-Cell Retrieval |
Robotic Operation of a Microfluidic Cell Migration Platform
Yu-Chih Chen, Ph.D.
|