Computomics Podcast - In Conversation

In the first episode of Season 8, Felix Sattler, our very own ;) Bioinformatics Analyst at Computomics, talks about predictive breeding, environmental modeling, and the role of data in improving breeding decisions under challenging growing conditions.

Drawing on his pearl millet research in West Africa, Felix reflects on how breeding analytics have evolved from simpler statistical approaches to models that combine genomic, phenotypic, and environmental data.

The conversation also looks at genotype-by-environment interactions, field trial design, data quality, and the importance of setting breeding targets that reflect local conditions rather than focusing on yield alone.

A broader theme runs through the episode: advanced analytics can only create value when they are paired with the right infrastructure, local knowledge, and practical understanding of how crops are grown and used.

Listen to Season 8, Episode 1: Can AI Make Plant Breeding More Accessible?

In this episode
Using rainfall, humidity, soil, radiation, and other environmental variables in predictive models
Why breeding for resilience can be more important than breeding for maximum yield
The challenges of applying AI and AGtech solutions across different regions and cultures
Why local knowledge and trust matter when entering new agricultural markets
The role of data infrastructure in making AI accessible to smaller breeding programs
How to work with messy and inconsistent breeding data
Computomics’ work with Chapter 54 and its exploration of opportunities across African markets

Mentioned in the episode
Chapter 54
An accelerator supporting European scale-ups exploring opportunities in African markets.

ICRISAT
The International Crops Research Institute for the Semi-Arid Tropics, an agricultural research organization with extensive work in dryland regions.

Dorcus C. Gemenet
Works with CGIAR-NARS crop breeding programs and was highlighted by Felix as someone working to make new breeding technologies more accessible, including for smallholder farmers and breeding programs in regions such as the Sahel.

If you are interested in the future of agri-food innovation, this episode offers a clear view of where the sector stands today: more practical, more disciplined, and more focused on technologies that can prove their value in the real world.

In the final episode of Season 7 of the Computomics Podcast, we speak with Louisa Burwood-Taylor, Chief Communications and Intelligence Officer at AgFunder, a global venture capital investor working at the intersection of agri-food, AI, biology, climate, and deep tech.
Louisa explains why agriculture remains one of the most challenging sectors for startups. A technology may be scientifically strong and technically impressive, but it still has to work across different soils, crops, climates, production systems, and growing cycles. In agriculture, innovation only matters if it can create real value for growers, supply chains, and the wider food system.
The episode also looks at the growing role of multi-sector platforms, corporate partnerships, biologicals, robotics, physical AI, and the connection between agriculture, nutrition, and human health. From precision robotics to self-driving labs, Louisa shares a practical and optimistic perspective on where agri-food innovation is heading next.

Companies and people mentioned:
AgFunder, Verdant Robotics, Gabe Sibley (Verdant Robotics), Reservoir Farms, Atinary, Bayer, John Deere, and Bear Flag Robotics.

Louisa Burwood-Taylor on LinkedIn
Agfunder
Agfunder News

In this Computomics Podcast episode, we launch the new “Make Sense of Science” explainer series, a format designed to break down complex scientific and technical topics into clear, accessible conversations.
In this first episode, Computomics machine learning scientist Alaukik Saxena introduces deep learning for genomic prediction in plant breeding. He explains how breeders use DNA marker data, field data, and environmental information to predict important plant traits such as yield, flowering time, or plant height, often before new plants are even grown. The episode also looks at why this is such a challenging task: breeding data is complex, environments differ greatly, and models can easily give overly optimistic results if they are not tested carefully.

Alaukik also explains the difference between classical statistical models, machine learning, and deep learning, and where each approach has its strengths. A key message of the episode is that good models depend on good data. Better phenotyping, smarter experimental design, and realistic validation are just as important as the algorithms themselves.

In this episode of the Computomics podcast, Dr. Rex Bernardo discusses the evolution of predictive plant breeding, from early molecular markers to today’s machine learning and AI approaches. He explains how these tools can support not only trait prediction, but also more complex breeding decisions, such as whether a line is likely to be selected or become a successful variety. The conversation also highlights community-focused work on leafy African vegetables in Minnesota and reflects on how plant breeding education must evolve to prepare future breeders for a more interdisciplinary, data-driven field.

In this episode, Darryn Keiller, Futurist, Entrepreneur, and founder of WayBeyond, discusses how digital agronomy helps growers make better decisions with data. He explains how farms can combine sensor, environmental, plant-response, and satellite data to improve daily decisions, predict risks, and raise yield and fruit quality. The conversation also covers the challenges of introducing new technology into agriculture, especially around trust and data ownership. Darryn argues that AI will not replace growers, but support them by closing knowledge gaps, and helping younger professionals adapt faster. It also enables farming systems to respond better to climate change, labor shortages, and growing production pressures.
More:
Darryn Keiller on LinkedIn
WayBeyond Homepage