"Recent advances in high-throughput technologies have led to an explosion in the amount of data in the agronomic field," he adds. There is an urgent need to effectively integrate complementary information to understand the biological system as a whole." Indeed, for scientists, locating relevant information in these mountains of data is sometimes like looking for a needle in a haystack... The researchers from the DIADE and LIRMM Units plan to develop original approaches using artificial intelligence to integrate, enrich and analyze biological data with the aim of predicting key genes, as well as their regulatory networks, for agronomic traits of interest. They have already developed AgroLD, a knowledge base for the study of the phenome? of cultivated plants. With a hundred or so data sets from 15 public sources and concerning 12 tropical crop species, it will of course be at the heart of DIG-AI. This powerful tool is based on Semantic Web? technology and relevant biological ontologies? It will facilitate the formulation of new scientific hypotheses linking genotype and phenotype.

What are the key genes for plant adaptation to climate change? How do these genes interact with each other to determine phenotype expression? The DIG-AI members aim to understand how the biological system works. "The difficulty is to identify these interactions, which exist at different molecular levels in the plant and are strongly influenced by environmental factors," explain the resarchers in the proposal. By answering these essential questions, DIG-AI will contribute to achieving two of the Sustainable Development Goals: MDG 2 "Eradicate hunger, ensure food security" and MDG 15 "Conserve and restore terrestrial ecosystems".