The Fernandes Lab is part of the Center for Agricultural Data Analytics (CADA) and the Department of Crop, Soil, and Environmental Sciences at the University of Arkansas, led by Dr. Samuel B. Fernandes, Assistant Professor of Agricultural Statistics and Quantitative Genetics.
We sit at the meeting point of quantitative genetics, statistics, and machine learning. Our work spans genomic prediction, genome-wide association studies, high-throughput phenotyping, and enviromics — integrating genomic and environmental information to predict how crops will perform across diverse environments. We build and freely share the software that makes these methods usable, and we partner with breeding programs in soybean, rice, maize, sorghum, and specialty crops to translate methods into real-world genetic gain.
Our mission is to connect genotype with phenotype: identifying the genes that shape complex traits and predicting how plants perform across diverse environments. By integrating quantitative genetics, statistics, and artificial intelligence, we develop predictive models and open-source software that advance plant breeding, biology, and physiology.
We envision a future in which the genome can be read not only as a sequence of DNA, but as a guide to how plants grow, adapt, and perform. In this future, researchers can identify the genes that shape complex traits and predict how different genotypes will respond across environments, transforming genomic information into tools for discovery, breeding, and crop design. As we learn to guide plant performance with greater precision, we also open the door to agriculture in the most challenging settings — from a changing Earth to future habitats beyond it.
By making plant biology more predictable and actionable, we aim to shorten breeding cycles, accelerate discovery, and help build agricultural systems that are more productive, sustainable, and resilient. Ultimately, our vision is to strengthen food security on Earth while expanding the role of plants in sustaining life wherever people go.
Concrete commitments that guide the projects we take on and the people we train.
Develop multi-omics prediction models that integrate genomic, environmental (enviromics), and high-throughput phenotyping data to forecast genotype-by-environment performance.
Push the statistical theory behind multi-trait, multi-environment analysis — from GWAS to genomic selection — so breeders can act on more of their data.
Create and maintain free, well-documented tools (such as our R packages) that the wider breeding and genetics community can use, trust, and extend.
Apply machine learning and AI to extract more value from the data breeding programs already collect — answering the question, "Can we better use the data we have?"
Mentor quantitative geneticists, statisticians, and data scientists for careers in industry and academia.
Collaborate with breeding programs in soybean, rice, maize, sorghum, and specialty crops to turn new methods into measurable genetic gain in farmers' fields.
If our mission resonates with your background in statistics, genetics, engineering, computer science, or agronomy, we'd love to hear from you.
Get in touchType to search · Esc to close · ⌘K to open anywhere
