Our lab develops artificial intelligence (AI) and computational genomics methods to bridge the gap between genetic variation and human disease. We design tools that improve both the identification of genetic variants and the interpretation of their biological effects, with the goal of enabling faster and more precise diagnoses and guiding therapeutic strategies for rare diseases and cancer. Our lab is also involved in developing innovative strategies to reconstruct on-target genome editing events by integrating long-read sequencing with advanced computational approaches.
Some recent achievements include:
- A hybrid AI model that jointly processes Illumina (short-read) and Nanopore (long-read) sequencing data to improve germline variant detection, achieving higher accuracy at lower cost (Cell Reports Methods, 2025).
- A lineage tracing and single-cell transcriptomics study in triple-negative breast cancer that identified IGFBP2 as a driver of drug tolerance (Genome Medicine, 2024).
- Computational methods for predicting drug response from single-cell expression profiles, helping to anticipate treatment outcomes and improve therapy selection (BMC Medicine, 2023).
- A single-cell atlas of breast cancer cell lines that revealed how transcriptional heterogeneity influences therapeutic response (Nature Communications, 2022).
By integrating AI, high-dimensional single-cell data, genome editing, and next-generation sequencing technologies, our lab transforms genomic information into actionable insights. Ultimately, our work supports earlier diagnoses, more effective therapies, and new opportunities for personalized medicine.
Quote
Computational machine learning combined with human decision making, I believe, is the road we have to pursue for effective and individualized treatment.
Additional Funding
- Explore therapeutic resistance of triple negative breast cancer with single-cell transcriptomics and lineage tracing (2020-2024), AIRC