Official websites use .gov

A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS

A lock ( 🔒 ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

Need Help

Pathway-Centric AI for Dissecting Colorectal Cancer Heterogeneity Using Single-Cell and Spatial Omics

Award Number: K99CA312801
ORGANIZATION: NATIONAL CANCER INSTITUTE
OPDIV: NIH
AWARD CLASS: DISCRETIONARY
AWARD ACTIVITY TYPE: TRAINING/TRAINEESHIP
PERIOD OF PERFORMANCE START DATE: 07/01/2026
PERIOD OF PERFORMANCE END DATE: 06/30/2038

Group Awards By:

View Award Description

Pathway-Centric AI for Dissecting Colorectal Cancer Heterogeneity Using Single-Cell and Spatial Omics - ABSTRACT Understanding functional pathway activity from single-cell (scRNA-seq) and spatial transcriptomic (ST) data is critical for elucidating disease mechanisms, particularly in complex diseases like colorectal cancer (CRC). Current methods often overlook crucial molecular interactions and spatial context, limiting biological interpretability and the discovery of robust biomarkers, thereby hindering progress in precision oncology for CRC. This project will develop and apply novel interpretable Artificial Intelligence (AI) frameworks, SPACT and SPACT-ST, to infer context-aware pathway activity from scRNA-seq and ST data, addressing these limitations. The K99 mentored phase will focus on methods development and validation: · Aim 1 will develop and validate SPACT, an interpretable AI model that leverages molecular interaction network topology for robust single-cell pathway inference. · Aim 2 will extend this framework to spatial data through SPACT-ST, incorporating tissue architecture such as histology and cellular neighborhoods to produce spatially coherent functional maps. Both tools will be rigorously benchmarked for accuracy, interpretability, and robustness across diverse platforms and tissue types. The R00 independent phase will apply these tools to generate a publicly accessible CRC Pathway Atlas by analyzing large-scale scRNA-seq and ST datasets (>3.7 million cells from >250 patients). This resource will characterize pathway heterogeneity across CRC subtypes and cell states, identify spatially organized functional programs in the tumor microenvironment, and support the development of novel pathway-based CRC classifications. It will also facilitate the discovery of candidate biomarkers linked to clinical and genomic features, with the potential to inform therapeutic targeting and precision treatment strategies. This research will yield open-source, interpretable AI tools (SPACT, SPACT-ST) and a high-resolution CRC Pathway Atlas for the research community. These resources will advance our understanding of CRC and accelerate precision oncology efforts by enabling refined tumor subtyping, biomarker discovery, and development of targeted therapies to improve patient outcomes. This K99/R00 award will support Dr. Yuxia Xie’s transition to an independent research career in computational cancer biology. With a PhD in computer science and a strong track record in interpretable AI (e.g., CPACT, xMINT), Dr. Xie will be mentored by a highly interdisciplinary team at Duke University. This collaborative environment will provide the training and support necessary to lead an AI-driven research program in cancer precision medicine.


Issue Date FY	Funding FY	Legal Entity Name	Legal Entity Address	Legal Entity City	Legal Entity State	Legal Entity Zip Code	Legal Entity COUNTY	Legal Entity COUNTRY	Assistance Listing	Award Code	Budget Year	Action Date	Action Type	Action Amount

Issue Date FY: 2026 ( Subtotal = $109,767 )
2026	2026	DUKE UNIVERSITY	2200 W MAIN ST	DURHAM	NC	27705	DURHAM	USA	Cancer Research Manpower	000	1	6/30/2026	NEW	$109,767
														Subtotal = $109,767

Grand Total All Awards = $109,767

Top

All Categories

About

Search

Reports

Data Submission

Award Information

Pathway-Centric AI for Dissecting Colorectal Cancer Heterogeneity Using Single-Cell and Spatial Omics

Award Number: K99CA312801

ORGANIZATION: NATIONAL CANCER INSTITUTE

OPDIV: NIH

AWARD CLASS: DISCRETIONARY

AWARD ACTIVITY TYPE: TRAINING/TRAINEESHIP

PERIOD OF PERFORMANCE START DATE: 07/01/2026

PERIOD OF PERFORMANCE END DATE: 06/30/2038

Federal Websites

Department of Health & Human Services

HHS Operating Divisions

HHS Staff Divisions

Download A Document Viewer