The blueprint for better crops isn’t hiding in a Silicon Valley lab or in a fancy drone flying over farmland — it’s buried deep in the DNA of the plants themselves. The challenge? Finding the right genes in a sea of billions of nucleotides. That’s where artificial intelligence comes in.

“Gene discovery used to be like looking for a needle in a haystack,” says Christian Dreischer, scientific project manager at Computomics. “Now, it’s more like using a metal detector — with AI guiding us, we know exactly where to dig.”

Based in Tübingen, Germany, Computomics is at the cutting edge of what some are calling the fourth agricultural revolution. The company uses machine learning to sift through massive genomic datasets, identifying genetic markers that influence everything from drought tolerance to disease resistance to yield.

The key, Dreischer explains, is combining genomics with deep environmental data: soil composition, climate history, microbial activity, to see not just which genes exist, but which ones actually matter in real-world plant breeding. “What we’re doing is connecting genotype to phenotype with unprecedented precision,” he says.

This kind of insight is helping plant breeders skip years of field trials. In the past, it could easily take more than a decade to develop a new variety of wheat or soybean. Now, breeders using Computomics' tech can zero in on the most promising candidates almost immediately.

It’s not just about speed. It’s also about sustainability. “We’re in a race against a changing climate,” says Dreischer. “Crops need to do more with less: less water, fewer inputs, more resilience. That means finding and deploying the right genes, faster than ever.”

One of the company’s flagship platforms, xSeedScore®, runs millions of predictive simulations that tell breeders how likely a genetic combination is to succeed, in all sorts of climate conditions, before they ever plant a seed. Think of it as a weather forecast, but for gene performance.

And now, Computomics is pushing the frontier even further with Pantograph. While xSeedScore® specializes in decoding and predicting the value of complex genetic traits, Pantograph is focused on interpreting how those genes behave under specific environmental conditions.

“Pantograph gives us an incredibly detailed picture of how genes are expressed and regulated in different contexts,” Dreischer says. “Together, we’re not just asking what genes are present; we’re asking what they do, and when.”

This layered approach opens the door to context-aware breeding: selecting the best genetic profiles for specific geographies, soil types, or even climate scenarios. For breeders, that could mean seed varieties tailor-made and future-proof for local conditions.

“AI won’t replace plant breeders,” Dreischer is quick to point out. “But it will absolutely empower them. We’re giving them a turbocharged toolbox to tackle the toughest problems in agriculture.” 

As global food demand rises and ecosystems grow more fragile, gene discovery isn’t just a scientific curiosity, it’s a survival strategy. And with AI as the guide and tools like Pantograph lighting the way, the genetic frontier is opening faster than ever before.

Christian Dreischer
Data Science & Operations Manager

 

Originally posted on 3 November, 2025 on SeedWorld US