Research
Our laboratory is at the forefront of DNA Large Language Models (LLMs) and their applications in plant genomics. We develop foundation models for DNA sequence understanding and apply them to diverse biological questions, from regulatory element prediction to genome engineering. Below are our key research areas:
DNA Large Language Models
We develop and apply large language models for DNA sequence analysis. This includes PDLLMs (Plant DNA LLMs) — a suite of foundation models tailored for plant genomes published in Molecular Plant — and DNALLM-Suite, a comprehensive toolkit for fine-tuning and inference with DNA Language Models featuring CLI, Web UI, and MCP protocol support.
AI-Driven Genomics
We apply machine learning and deep learning to solve fundamental questions in genomics. This includes developing predictive models for gene regulation, chromatin accessibility, and genome evolution. Our work bridges the gap between cutting-edge AI methods and biological discovery.
Cis-regulatory Elements
Cis-regulatory elements (CRMs) control gene expression during specific developmental stages or under various biotic and abiotic stresses. We identify and characterize these elements based on their unique molecular signatures associated with open chromatin, leveraging LLM-based approaches for improved prediction accuracy.
Oligo-FISH Probe Design
Oligo probes designed from conserved DNA sequences can be used among genetically related species, enabling comparative cytogenetic mapping. We develop computational pipelines for genome-scale oligonucleotide-based probe design for fluorescence in situ hybridization (FISH), significantly expanding the applications of FISH in non-model plant species.
CRISPR/Cas Genome Editing
We develop and optimize CRISPR/Cas-based genome editing systems for plants, including base editors, prime editors, and multiplex editing strategies. Our work includes gRNA design algorithms, efficiency prediction models, and the development of CrisprStitch for the research community.
Plant Genomics & Comparative Genomics
We study the structure, function, and evolution of plant genomes using large-scale sequencing and comparative approaches. Our recent work includes telomere-to-telomere genome assemblies and the application of large language models for DNA sequence analysis in plants.
Epigenetics & Chromatin Biology
We investigate the epigenetic regulation of gene expression in plants, focusing on DNA methylation, histone modifications, and chromatin accessibility. Our research explores how epigenetic changes contribute to plant development and stress responses.
Bioinformatics Tool Development
We develop and maintain open-source bioinformatics software for the plant science community, including tools for probe design (Chorus2), CRISPR analysis (CrisprStitch), and plant DNA language models (PDLLMs, DNALLM-Suite). All tools are freely available.