Building biological foundation models, scientific AI agents and AI products people use — from genomes to the App Store.
- 01
Doover: an AI photo-art app, from first commit to the App Store in ten days
An iOS app that gives ordinary photos a second life: 15 hand-written visual skills and 41 looks rendered with gpt-image-2, an on-device Core Image color pipeline, and…
App Store/Swift / SwiftUI/Core Image · Vision/gpt-image-2 - 02
RallyAI: an on-device AI tennis coach for iPhone, Apple Watch and WeChat
A tennis-training companion that runs swing detection, ball-speed estimation, placement and technique scoring entirely on the iPhone Neural Engine, so video never leaves the phone.
in-progress/Swift / SwiftUI · watchOS/Core ML · YOLO-pose · GCN/PyTorch → ONNX → Core ML
- 01
Subset-based method for cross-tissue transcriptome-wide association studies improves power and interpretability
Integrating results from genome-wide association studies (GWASs) and studies of molecular phenotypes such as gene expressions can improve our understanding of the biological…
2024/Human Genetics and Genomics Advances - 02
From G1 to M: a comparative study of methods for identifying cell cycle phases
Accurate identification of cell cycle phases in single-cell RNA-sequencing (scRNA-seq) data is crucial for biomedical research.
2024/Briefings in Bioinformatics - 03
ERK hyperactivation serves as a unified mechanism of escape in intrinsic and acquired CDK4/6 inhibitor resistance in acral lentiginous melanoma
Patients with metastatic acral lentiginous melanoma (ALM) suffer worse outcomes relative to patients with other forms of cutaneous melanoma (CM), and do not benefit as well to…
2024/Oncogene - 04
Methylation of the chromatin modifier KMT2D by SMYD2 contributes to therapeutic response in hormone-dependent breast cancer
Activating mutations in PIK3CA are frequently found in estrogen-receptor-positive (ER+) breast cancer, and the combination of the phosphatidylinositol 3-kinase (PI3K) inhibitor…
2024/Cell Reports
- 01
DNA Foundation Models Are Not Interchangeable: A Practical Routing Guide
A survey of 40 genomic models and a small hands-on comparison of three, on how training objective, context, outputs, and operational cost shape which model suits a given genomics…
completed/Python/PyTorch/AWS SageMaker - 02
Beyond Consequence Labels: Cohort-Scale Sequence-to-Function Interpretation
A dual-model workflow used AlphaGenome and Enigma to add functional evidence to variants missed by conventional high/moderate consequence labels, expanding the reviewable space…
completed/Python/AlphaGenome/Enigma - 03
Teaching Sequence-to-Function Models to Read Structural Variants
A structural-variant encoding study translated deletions, duplications, inversions, breakends, and gene fusions into matched REF/ALT DNA windows for AlphaGenome and Enigma.
completed/Python/AlphaGenome/Enigma - 04
When Sequence Is Not Enough: Modeling FFPE Artifact Risk
A group-held-out study across public multi-caller data and a de-identified matched FF/FFPE cohort showed that read-level technical evidence—not DNA sequence alone—is the key signal…
completed/Python/scikit-learn/Gradient Boosting
Building something at the edge of AI and biology? I'd like to hear about it.
Get in touch