Christopher I. Li

This study identifies subtype-specific molecular and immune features associated with recurrence risk in breast cancer using multi-omic analysis of 340 tumors and validates predictive models in an external cohort.

PhenoBIC is a deep learning model that automates cell phenotyping in multiplex imaging data, outperforming manual methods and other ML approaches with an F1-score of ~0.88.

This study identifies Leptotrichia wadei as a pro-tumorigenic bacterium in CRC by linking metabolic specialization to microbial niches and tumor progression.

ROSIE generates multiplex immunofluorescence staining from H&E images by imputing protein expression and localization using deep learning.

This study identifies obesity-driven inflammatory and metabolic transcriptomic signatures in mouse and human colon tumors through integrated transcriptomic analyses.

This study identifies 13 new genetic loci specific to proximal or distal colorectal cancer subtypes, revealing partly distinct genetic architectures for these tumor locations.

This study uses Mendelian randomization to show that higher physical activity levels are associated with lower risks of breast and colorectal cancer.