Generative AI
We develop generative models that reconstruct, enhance, and infer molecular profiles — turning sparse and incomplete measurements into rich, usable signal.
CBIO LAB · RESEARCH
Artificial intelligence for spatial omics.
We develop generative and representation-learning methods that expand what spatial omics can measure, reconstruct, and reveal.
Our approach
We sit at the intersection of artificial intelligence and the life sciences. We build intelligent tools that turn complex omics data into interpretable signals — and use them to uncover new biological insights.
Spatial omics
We work across single-cell and spatial platforms to recover tissue architecture, cell-cell interactions, and 3D molecular context at single-cell resolution.
Bioinformatics analysis
We design computational pipelines and cross-modal inference methods that extract biological insight from complex, heterogeneous omics data.
How we work
Three principles guide everything we build.
Biological grounding
Start from a real biological limitation; validate with independent evidence.
Generalization
Test across tissues, platforms, and resolutions — not a single benchmark.
Open & reproducible
Release code, models, and data so results can be examined and extended.