Mural AgriOmics Lab

We study the genetic basis of complex, polygenic traits using genome-wide association studies (GWAS), QTL mapping, and genomic selection models. Our work identifies genetic loci and pathways driving variation in traits like yield, drought tolerance, and nutrient use efficiency across diverse crop populations.

We leverage large-scale genomic datasets — including SNP arrays, whole-genome sequencing, and transcriptomics — to understand gene function, population structure, and evolutionary history in major crops like maize, soybean, and sorghum. We are particularly interested in the genomic basis of local adaptation and stress tolerance.

Measuring plant traits quickly and accurately is a bottleneck in modern breeding. We develop and apply sensor-based, drone-based, and image-driven phenotyping approaches to measure hundreds of traits across large field populations. By coupling HTP data with genomic information, we aim to dramatically improve genetic mapping and breeding selection efficiency.

We develop bioinformatics pipelines and statistical models to handle the large datasets generated by modern genomics and phenotyping experiments. We use R, Python, and HPC to build reproducible workflows, and explore machine learning for genomic prediction, image-based trait extraction, and integrative multi-omics analysis.

We apply cutting-edge artificial intelligence and deep learning methods to agricultural genomics challenges. This includes training neural networks for genomic prediction, developing convolutional models for image-based phenotyping, applying natural language processing to mining biological literature, and building transformer-based models that integrate multi-modal data — genomic, phenotypic, and environmental — for precision crop improvement.