Brad Jakeman
Founding Managing Partner, Rethink Food
At Rethink Food, we are particularly excited about AI disruption in the area of plant genomics.
To develop the next generation of tailored plant breeding that could transform food security and climate change adaptation, scientists and startups are exploring new horizons by using AI and big data to decipher the intricate world of plant genomes.1
Understanding the Science Behind AI-Driven Genetic Innovation
AI can help to advance the field of plant genomics by accelerating the identification and understanding of genetic traits that contribute to plant growth, disease resistance, and environmental adaptability. Through machine learning algorithms, AI can analyze vast genomic datasets to identify patterns and correlations that might otherwise be difficult for humans to detect. This allows for more accurate predictions of how certain genes affect plant characteristics, enabling the development of crops that are more resilient to climate change, pests, and diseases. AI can also optimize the process of gene editing by identifying the most effective targets for modification, thereby speeding up the development of new plant varieties with enhanced nutritional value, higher yields, and better resistance to environmental stressors. By automating complex data analysis and prediction tasks, AI is transforming plant genomics into a more efficient and powerful tool for sustainable agriculture and food security.
To most people, a tomato or potato are simply the ingredients in some of our favorite dishes. However, to scientists, they are a challenging riddle that, if worked out, may reveal solutions that could help us produce more food with less negative impact on the environment, mass produce new biofuel sources, and improve human longevity. These findings are embedded in the plant DNA that scientists have begun to unlock, thanks to the help of cutting-edge AI technologies. AI’s ability to analyze vast data allows for a deeper understanding of plant genomes, helping researchers develop resilient strains that can better withstand stressors like pest infestation, climate change, and pesticide resistance.
Because plants have evolved over a much longer time span than other forms of life, their genomes – even in “simple” plants like sugarcane – are significantly larger than those of humans or animals. Capturing the interactions between genes and alleles (different versions of the same gene that cause trait variations) across different ploidies (the number of chromosome sets in a call) can be challenging. This is because some of these ploidies may reflect orphan genes from older plant strains that are no longer active. Plants are often polyploid, meaning they have duplicated genes or entire genomes, resulting in more than two sets of chromosomes. Polyploidy is common in plants and is linked to speciation and the evolution of new species. Many crop plants and wild species are polyploid, such as wheat (tetraploid or hexaploid), potatoes (tetraploid), and bananas (triploid). Other examples include canola, cotton, and certain apple varieties. This genetic condition often leads to plants with larger fruits, seeds, or other desirable traits, making it a significant focus in plant breeding and agriculture.2
Single nucleotide polymorphisms (SNPs), or common DNA sequences, are a key research focus because they reveal how plants interact with their environment. By understanding each gene’s function, scientists can develop plants better suited to human needs. For example, to create drought-resistant wheat, researchers would seek genes that promote growth in arid conditions. This gene might be inactive in some samples, but machine learning could identify its potential by analyzing gene-environment interactions, aiding in AI-driven breeding techniques.
While genetic engineering is one approach to developing crops with desired traits, traditional breeding methods have been around for thousands of years. AI enhances this process by identifying the most compatible strains and predicting which breeding strategies – such as hybridization, extensive crossbreeding, or chromosomal building – will be most effective. With comprehensive genetic data, machine learning can link genes to optimal conditions, potentially extending growing seasons or enabling cultivation into previously unsuitable areas. This could significantly increase food production for a growing population, allowing crops to thrive in urban or desert regions and adapt to climate change.
AI can also help breed crops resistant to particular pests or diseases. By using machine learning to detect traits that attract pests, such as scent, color, etc., researchers could develop plants that are less appealing to insects or other pests, reducing the need for pesticides. This could lead to more eco-friendly pest control methods and a shift toward “personalized agriculture,” where pesticides are tailored to specific plants and regions.
Before AI, understanding plant genomes was nearly impossible. Now, AI technologies enable scientists and startups to unravel these complexities, creating solutions to help plants survive threats like pollution, urbanization, climate change, and other problems that compromise their quality and growth. Advanced machine learning will help researchers solve the puzzles of plant biology, offering innovations that improve the future of agriculture and human wellbeing.
How Phytoform is Transforming Plant Genomics with AI
Earlier this year, we invested in Phytoform under our AgTech & Precision Farming investment vertical. The company uses a patented machine learning technique, CRE.AI.TIVE, to pinpoint minute modifications in DNA sequences that will have the greatest effect on crops. For example, cultivating plant varieties that are more resilient and less susceptible to crop losses. Phytoform develops traits in under two years and $2MM versus the current industry standard of eight years and $115MM.
The platform employs high throughput techniques to sift through millions of sequences in the lab using a combination of AI and genomic data, providing it with previously unheard-of control over gene expression. The outcomes can then be used to iterate and gradually enhance the model. Once beneficial features have been identified, they employ footprint-free genome editing, fueled by knowledge of tissue culture and microfluidics. Phytoform has the ability to quickly deploy the feature across a wide number of species without the need for transgenes, allowing it to achieve scalability and accelerate time to market. Commercially relevant features can then be produced at a fraction of the effort and expense.
Phytoform’s current approach uses genome editing technology in conjunction with AI-driven trait discovery to target many attributes across several plant kinds. By minimizing waste in the supply chain, the company is developing features for crops (like potatoes and tomatoes) that will limit environmental impact. In addition, Phytoform is developing characteristics for newly emerging agriculture industries like vertical farming and substitute plant proteins. These examples are a handful of the commercially feasible and environmentally minimal projects the company has in the works.
Future Directions for Agriculture
At Rethink, we believe that the future of agriculture will increasingly reply on AI and advanced technologies to drive innovation. Integrating AI into plant genomics will enhance our ability to understand and improve crop traits, resulting in plants that are more resilient and adaptable. Investments in research and development will be important for refining these technologies and broadening their applications, as will advocating for their use across the industry. As these innovations like Phytoform continue to scale, we can advance towards more sustainable agriculture and improved food security.
Have ideas or opinions on this topic? Reach out to share them with us: food@rethink-capital.com
The information contained herein is confidential and should not be distributed, published or reproduced, in whole or in part, nor should its contents be disclosed by recipients to any other person without the prior written approval of Rethink Food. This material is for general informational purposes only and is not intended as investment advice, as an offer or solicitation of an offer to sell or buy, or as an endorsement, recommendation, or sponsorship of any company, security, advisory service, or product. This information should not be used as the sole basis for investment decisions. All content is presented as of the date published or indicated only, and may be superseded by subsequent market events or for other reasons. Past performance is no guarantee of future results. Investing involves risk including the loss of principal and fluctuation of value.