Oral Presentation Multi-Omics Conference 2024

CellClique:dissecting single-cell spatial interactions using generative AI and spatially resolved transcript analysis (110047)

Sachini Weerasekara 1 , Natasha Darras 2 , Jacqueline Isaacs 1 , Sagar Kamarthi 1 , Colles Price 2
  1. Northeastern University, Cambridge, MA, United States
  2. Takeda Pharmaceutical, Cambridge, MA, USA

Novel cellular therapy has the potential to change cancer treatment.  However, to design new therapies requires a better understanding of the tumor microenvironment (TME) and the cellular dynamics that occur within the tumor.  This requires a high resolution and subcellular view of cancer cells, immune cells while identifying the neighborhood architecture that exists between them. 

We introduce CellClique, a tool that assists in dissecting TMEs using spatial transcriptomics and Hematoxylin and Eosin (H&E) stains. CellClique integrates both generative and analytical capabilities to provide a comprehensive view of TMEs. On the generative end, CellClique uses generative AI to model both gene expressions and cell-neighborhood interactions for cell annotation, gene expression prediction, and gene imputation. On the analytical end, CellClique offers a comprehensive four-way spatial analysis scheme aimed at 1) understanding the effects of neighboring cells on a target cell, 2) dissecting cancer-normal cell boundaries, 3) observing immune invasion mechanisms, and 4) exploring cellular lifecycles with the TME.

We applied CellClique to study the cellular landscape in lung and colon cancers. By generating proximal (10 microns), local (60 microns), and distal neighborhoods (120 microns) for each cell within the tissue, we find distinct expression profiles of tumor-infiltrated lymphocytes, which are functionally different from lymphocytes excluded from the tumor. Additionally, we present expression variation patterns in cancer-normal boundary cell communities with various cell compositions. We also compare lymphocytes at different stages of life by analyzing expression differences among lymphocytes classified with over 80% confidence, those with a 50% probability of being lymphocytes, and those with below 20% probability. We further compare how these neighborhoods change between lung cancer and colon cancer. 

Together, these methods paint a complex picture of the TME but also identifies numerous avenues for future cellular therapy and development of new treatments.