How artificial intelligence is boosting crop yield to feed the world 

The GRAIN system aims to be the “holy grail” for genetic science.

Over the last several decades, genetic research has seen incredible advances in gene sequencing technologies. In 2004, scientists completed the Human Genome Project, an ambitious project to sequence the human genome, which cost $3 billion and took 10 years. Now, a person can get their genome sequenced for less than $1,000 and within about 24 hours. 

Scientists capitalized on these advances by sequencing everything from the elusive giant squid to the Ethiopian eggplant. With this technology came promises of miraculous breakthroughs: all diseases would be cured and world hunger would be a thing of the past. 

So, where are these miracles? 

“We need about 60 to 70% more food production by 2050.”

Oliver Peoples

In 2015, a group of researchers founded Yield10 Bioscience, an agriculture biotech company that aimed to use artificial intelligence to start making those promises into reality. 

Two things drove the development of Yield10 Bioscience.

“One, obviously, [the need for] global food security: we need about 60 to 70% more food production by 2050,” explained Dr. Oliver Peoples, CEO of Yield10 Bioscience, in an interview with Freethink. “And then, of course, CRISPR.”

CRISPR: Finally a way to easily edit DNA

It turns out that having the tools to sequence DNA is only step one of manufacturing the miracles we were promised. 

The second step is figuring out what a sequence of DNA actually does. In other words, it’s one thing to discover a gene, and it is another thing entirely to discover a gene’s role in a specific organism. 

In order to do this, scientists manipulate the gene: delete it from an organism and see what functions are lost, or add it to an organism and see what is gained. During the early genetics revolution, although scientists had tools to easily and accurately sequence DNA, their tools to manipulate DNA were labor-intensive and cumbersome.

It’s one thing to discover a gene, and it is another thing entirely to discover a gene’s role in a specific organism.

Around 2012, CRISPR technology burst onto the scene, and it changed everything. Scientists had been investigating CRISPR — a system that evolved in bacteria to fight off viruses — since the ‘80s, but it took 30 years for them to finally understand how they could use it to edit genes in any organism.

Suddenly, scientists had a powerful tool that could easily manipulate genomes. Equipped with DNA sequencing and editing tools, scientists could complete studies that once took years or even decades in mere months. 

Promises of miracles poured back in, with renewed vigor: CRISPR would eliminate genetic disorders and feed the world! But of course, there is yet another step: figuring out which genes to edit.

Which genes are worth editing?

Over the last couple of decades, researchers have compiled databases of millions of genes. For example, GenBank, the National Institute of Health’s (NIH) genetic sequence database, contains 38,086,233 genes, of which only tens of thousands have some functional information. 

For example, ARGOS is a gene involved in plant growth. Consequently, it is a very well-studied gene. Scientists found that genetically engineering Arabidopsis, a fast-growing plant commonly used to study plant biology, to express lots of ARGOS made the plant grow faster. 

Dozens of other plants have ARGOS (or at least genes very similar to it), such as pineapple, radish, and winter squash. Those plants, however, are hard to genetically manipulate compared to Arabidopsis. Thus, ARGOS’s function in crops in general hasn’t been as well studied.

The big crop companies are struggling to figure out what to do with CRISPR.

CRISPR suddenly changed the landscape for small groups of researchers hoping to innovate in agriculture. It was an affordable technology that anyone could use — but no one knew what to do with it. Even the largest research corporations in the world don’t have the resources to test all the genes that have been identified. 

“I think if you talk to all the big crop companies, they’ve all got big investments in CRISPR. And I think they’re all struggling with the same question, which is, ‘This is a great tool. What do I do with it?’” said Dr. Peoples.

The algorithm can identify genes that act at a fundamental level in crop metabolism.

GRAIN: The holy grail of crop sciences

The “holy grail of crop science,” according to Dr. Peoples, would be a tool that could identify three or four genetic changes that would double crop production for whatever you’re growing.

With CRISPR, those changes could be made right now. However, there needs to be a way to identify those changes, and that information is buried in the massive databases.

To develop the tool that can dig them out, Dr. Peoples’ team merged artificial intelligence with synthetic biology, a field of science that involves redesigning organisms to have useful new abilities, such as increasing crop yield or bioplastic production. 

This union created Gene Ranking Artificial Intelligence Network (GRAIN), an algorithm that evaluates scientific databases like GenBank and identifies genes that act at a fundamental level in crop metabolism. 

That “fundamental level” aspect is one of the keys to GRAIN’s long-term success. It identifies genes that are common across multiple crop types, so when a powerful gene is identified, it can be used across multiple crop types.

For example, using the GRAIN platform, Dr. Peoples and his team identified four genes that may significantly impact seed oil content in Camelina, a plant similar to rapeseed (true canola oil). When the researchers increased the activity of just one of those genes via CRISPR, the plants had a 10% increase in seed oil content.

It’s not quite a miracle yet, but with more advances in gene editing and AI happening all the time, the promises of the genetic revolution are finally starting to pay off.

We’d love to hear from you! If you have a comment about this article or if you have a tip for a future Freethink story, please email us at [email protected].

OpenAI’s GPT-4 outperforms doctors in another new study
OpenAI’s most powerful AI model, GPT-4, outperformed junior doctors in deciding how to treat patients with eye problems.
Watch the first AI vs. human dogfight using military jets
An AI fighter pilot faced off against a human pilot in a “dogfight” using actual planes — a huge milestone in military automation.
AI can help predict whether a patient will respond to specific tuberculosis treatments
Instead of a one-size-fits-all treatment approach, AI could help personalize treatments for each patient to provide the best outcomes.
New AI music generator makes songs from text prompts
AI music generators — AIs that create new music based on users’ text prompts — are lowering the bar for music creation, for better or worse.
Meta’s AI assistant just got a major upgrade — here’s how you can use it
Meta has upgraded its AI assistant, Meta AI, with a new LLM (Llama 3) and made it available in far more places.
Up Next
Subscribe to Freethink for more great stories