A team of researchers led by María José Aranzana, a researcher at the Institute of Agri-Food Research and Technology (IRTA) and head of the Genomics and Biotechnology program at the Agrigenomics Research Center (CRAG), has developed an innovative tool based on generative artificial intelligence to predict complex traits in plants.
The tool, called GenoDrawing, introduces a new methodology that predicts the physical traits of an apple variety from its genetic information, generating images that closely resemble the real thing. The study has been published in the journal Plant Phenomics.
The researchers used a deep learning - or machine learning - approach to train the GenoDrawing model on a dataset of more than 10,000 images of apples and their genetic information (SNP markers). Then, they tested its effectiveness on a separate dataset. The results show that GenoDrawing can very accurately predict and create images of the shape of apples based solely on information from SNP markers.
One of the benefits of GenoDrawing is that it can be used to predict a complex trait like shape. In addition, in subsequent studies, it could predict other characteristics, such as color, leaf morphology, or plant architecture. This is important because traditional plant phenotyping methods, such as visual inspection, are slow and expensive and may not deliver accurate results.
GenoDrawing also allows researchers to quickly and easily predict plant traits without having to grow them or wait for them to bear fruit, which could lead to more efficient and effective breeding programs.
A tool with multiple applications
The researchers stressed the potential that artificial intelligence (AI) has to improve processes in the scientific field. "We believe that this tool could have a wide range of applications, not only in plant cultivation and agriculture, but also in different areas of scientific research, and we hope to explore its potential in future research," stated Dr. Aranzana.
The use of generative AI in the GenoDrawing tool represents a major step forward in crop improvement. The project's scientific team hopes that their work will inspire further research and lead to further advances in agriculture and plant breeding.
Source: irta.cat