Deciphering plant genotype-phenotype Interactions using knowledge Graphs and AI – DIG-AI

The demand for food is expected to grow substantially in the future. To meet this challenge in a context of climate change, a better understanding of genotype-phenotype relationships is crucial to improve crop production capacities. Agronomic research is witnessing an unprecedented revolution in the acquisition of various data. The understanding of genotype-phenotype interactions is one of the most critical research areas in agronomy. However, these interactions are complex to identify because they are expressed at different molecular levels and are strongly influenced by environmental factors. The challenges consist in identifying these interactions by integrating information from different levels in a global model using a systemic approach in order to understand the real functioning of the biological system.
The semantic Web offers the methods and technologies to transform big data into knowledge. We developed AgroLD, a Knowledge Graph powered by Semantic Web technologies to integrate heterogeneous agronomic data. DIG-AI’s main objective is to develop data science methods to leverage actionable knowledge from biological and identify real world candidate genes for plant improvement. We will focus on the following specific objectives: 1) Exploit knowledge engineering methods through the use of ontologies to formulate research hypotheses that link genotype to phenotype, 2) Manage efficiently large volumes of biological data to extract knowledge, 3) Use the AgroLD knowledge base to build networks of molecular interactions between genes and phenotypes from scattered data (scientific articles, public databases, experimental data), 4) Identify key genes for plant improvement among hundreds of potential results. The ambition of DIG-AI will be to build upon the AgroLD KG to develop several research directions. They are grouped into three research areas: dynamic data integration, knowledge enrichment and candidate genes prioritization.