Ag Tech Talk: AI’s Growing Role in Agricultural R&D
12 March 2026 Artificial intelligence (AI) is rapidly reshaping how agricultural researchers analyze complex biological data and accelerate discovery. In this episode, Josh Colmer, CEO and co-founder of TraitSeq, discusses how AI-driven tools are helping scientists unlock insights from genomics, metabolomics, and other “omics” datasets to improve crop protection, plant health, and biological product development.
Colmer shares how emerging AI architectures and agentic systems enable deeper analysis, faster decision-making, and new approaches to understanding plant biology across the R&D pipeline.
Podcast Transcript:
*This is an edited and partial transcript.
AgriBusiness Global: AI is becoming a major tool in agriculture R&D. From your perspective, how is AI changing the way companies extract insights from complex biological data like genomics and metabolomics?
Josh Colmer: I think two simultaneous impacts are happening. First, you have purpose-built AI architectures — transformer models trained on biological sequence data that can predict things like protein structure or gene function in ways traditional statistics simply couldn’t.
The second impact is agentic AI, which is what we’re most focused on at TraitSeq. These systems can be guided to perform complex analytical workflows. In practice, that means an AI system can take gene expression or metabolomic data and interpret it with the depth of someone who has read hundreds of research papers on those genes, metabolites, and pathways.
ABG: Many companies are already generating huge amounts of omics data. Why are traditional analysis methods often leaving valuable insights on the table, and how can AI unlock more value from those datasets?
JC: One of the biggest bottlenecks is translating analysis results into something that drives an R&D decision — whether that’s prioritizing leads, positioning a product, or understanding why one candidate performs better than another.
Another challenge is that the annotation information needed to interpret these datasets — gene functions, pathways, metabolite libraries — is scattered across academic repositories online. To get maximum value from the data, you’d need to explore a huge range of sources, which isn’t feasible for most teams.
AI systems can help overcome those barriers by connecting analysis results with the broader body of published research, allowing companies to reach more informed research decisions faster.
Get The Newsletter Today!
