Artificial Intelligence

Artificial intelligence helps scientists design better food


Combating hunger and maintaining complete nutrition remains a challenge around the world, particularly in underdeveloped regions. Efforts to combat those challenges are ongoing and it’s a big job, one which gets bigger as the population increases, and the effects of climate change alter the agricultural landscape.

Sometimes finding the solution to big problems requires detailed knowledge of very small things, like the genome of staple plants like chickpeas. A recent study carried out by Rajeev Varshney from the Center of Excellence in Genomics and Systems Biology at the International Crops Research Institute for Semi-Arid Tropics (ICRISAT) and colleagues looked at the genome of chickpeas in hopes of finding a way to build a better food crop. Their findings were published in the journal Nature.

They didn’t just look at one kind of chickpea. Instead, the work analyzed more than 3,366 varieties of the plant — 3,171 cultivated species and 195 wild species — in order to get a full picture of genetic diversity. They created a pan-genome which describes genetic diversity across cultivated species and their wild source plants.

“It was a long journey from inception in 2014,” Varshney told SYFY WIRE. “This was the first effort of its kind across any crop. It took about three years for us to generate all of the data and then three to four years for data analysis and interpretation.”

The work, though daunting, resulted in the identification of 29,870 total genes, including 1,582 which had not been reported before. This analysis identified beneficial genes as well as detrimental mutations which result in less successful plants and lower crop yields. That data was then delivered to the University of Queensland where it was analyzed by an artificial intelligence called FastStack, which is specialized for designing new varieties of plants and crops with an eye toward optimal output.

Productivity of pulse crops, of which chickpeas are one type, has been stagnant for the last half-century. As populations increase, that has resulted in low per-capita food availability and contributes to malnutrition. Improving yields through enhancements could help to lessen some of that load.

“Chickpeas are an important legume crop, cultivated in more than 50 countries and are a rich source of protein. The chickpea is a key crop toward nutritional security, especially in developing countries,” Varshney said.

Scientists identified genes and gene families called haplotypes — groupings of genes which are all inherited together — which play potential roles in controlling seed size and development. Next steps are to take the genetic and artificial intelligence data and breed new, more robust varieties of chickpea in the real world. Importantly, the team identified potential enhancements with an eye toward maintaining genetic diversity. They estimate an ability to improve seed weight, an important yield metric, by up to 23%

“We propose three breeding approaches based on the genomic predictions that aim to improve 16 traits and enhance production,” Varshney said. “Our findings can be utilized to develop improved chickpea varieties with better yield, nutrition, and higher resistance to several biotic and abiotic factors. We have a plan to use the AI approach to combine haplotypes of our choosing for optimal output in elite varieties of chickpea.”

While this genetic and AI research has provided scientists with the ability to begin breeding novel species for improved crop yields, it has also opened a door for similar research in other staple crops. ICRISAT is reviewing other legumes and cereals, crops which have been identified as key stability foods, to identify their next target. Moreover, the work has the potential for deploying benefits not just across different species of chickpea, but to entirely different plants.

Previous work sequencing the genome of the pigeonpea isolated a gene which could be deployed in soybeans, making them resistant to Asian soybean rust, a destructive fungal disease. If the movies tell us anything about artificial intelligence, it’s that they will inevitably see us as a threat and seek to destroy us, but before they do, they’re busy working to design us more and better food.

We can’t wait to try some of that intelligently designed hummus.



READ SOURCE

This website uses cookies. By continuing to use this site, you accept our use of cookies.