Introducing Heritable Agriculture

Agriculture is a modern miracle—our food systems have been able to accommodate the world’s ever-growing population because of the beautiful nature of plants. Plants are solar powered, carbon negative, self-assembling machines that feed on sunlight and water. It’s hard to imagine a technology that would be cheaper to scale.

But ironically, even though agriculture is the only industry that’s fundamentally based on nature, it’s one of the most taxing for the planet. Agriculture accounts for nearly half of habitable land use, a quarter of anthropogenic greenhouse gas emissions, and 70 percent of the world’s groundwater withdrawals.

It doesn’t have to be that way. Several years ago, I set out with a small team of scientists, researchers, and technologists to explore how we might address the challenges facing the agriculture industry by harnessing the diversity of our natural systems. Our work ultimately led us to build new technology to advance the way plants are bred.

Today, I’m excited to share that our moonshot has evolved into Heritable Agriculture, an independent company with funded from FTW Ventures, Mythos Ventures, and SVG Ventures.

_{To validate our models, we grew plants at the moonshot factory inside a specialized growth chamber.}

A wealth of data ripe for the parsing

Although we have a wealth of biological information, this data is messy and hard to parse. Our team decided to tackle this data challenge first. What if we could use that information to breed plants that can deliver higher crop yields, improve their nutrition, and boost their resilience?

My years as a synthetic biologist studying and working at organizations like the Synthetic Biology Research Consortium, the Harvard Medical School and the U.S. Department of Energy taught me that advancements in biology require space and patience to run tons of different experiments and test many approaches that might not work.

The flexible, iterative work environment at the moonshot factory, coupled with Google’s AI expertise, ultimately led our team to create a machine learning platform that can uncover critical new details about how plants grow, so we can use these breakthroughs to breed any type of crop to be more sustainable.

_{To generate AI-ready datasets to train models, the team focused on capturing multiple types of information, in multiple crops, in multiple tissues.}

'Programming' plants with machine learning

Our team inside X used advanced computational biology techniques to decode the natural adaptations in plant biodiversity, developing machine learning models that can identify and understand the function of specific genomes in plants. By understanding those genomes, the crops can then be bred with climate-friendly traits for increased yields, lower water requirements, and higher carbon storage capacity in roots and soil. 

To validate the models, our team grew thousands of plants at X inside a specialized growth chamber. The chamber was fitted with automated photography that took hourly pictures of test plants to measure them as they grew, which allowed us to track how accurately we could predict when a plant might flower, and understand how specific changes might affect that budding time.

We also spent time getting our hands dirty out in the field. In partnership with researchers and industry leaders, we ran studies across a variety of crops at agricultural sites in California, Wisconsin, and Nebraska, gathering information about everything from the number of kernels in an ear of corn to the bitterness of certain vegetable crops.

Amidst dust, debris, and sweltering heat, we developed systems to collect and freeze hundreds of samples in liquid nitrogen, and we dug and power-washed thousands of plant roots.

_{Preparing the liquid nitrogen needed to collect and freeze plant samples out in the field.}

Food, forestry, and beyond

From our tests, we see a path towards using our technology platform to identify and predict genetic determinants of plant traits like color, taste, water requirements, nitrogen requirements, and even the ability to absorb carbon from the air.

Now, at Heritable Agriculture, we’re working to help farmers and agriculture companies better understand the specific genomes of plants they will need to yield the healthiest, most abundant crops in their given environments. We’re also excited about the potential for our platform to enable sustainable forestry by re-populating native tree species in deforested areas. Our initial research suggests we can improve the health and resilience of trees 400 times faster than existing processes.

_{The team rapidly moved from growth chamber experiments to generating data from commercial crops, in real fields.}

Bringing Heritable to more partners around the world

A “heritable” trait is something that has been inherited by its genome. We chose the name “Heritable Agriculture” as a nod to the power of plants: if you get the genomes of these self-assembling, carbon-negative machines just right, you’ve built an intervention that can scale like nothing else. By optimizing these heritable traits, our vision is to use our technology to bring about a more diverse, abundant, resilient, and nutritious food system across the globe.

We’re currently collaborating with partners across the agriculture and forestry industries to bring Heritable Agriculture’s technology to life and advance our mission. If you’re working on advanced crop improvement in the seed, food, or forestry spaces and are interested in learning more, we’d love to talk—reach out to contact@heritable.ag.