In genomics, natural sequences resulting from billions of years of evolution represent an immense reservoir of diversity. However, it remains little explored due, on the one hand, to the too limited biochemical, functional and structural information available and, on the other hand, to a lack of powerful algorithms to classify protein functions. A research team from the Laboratory of Computational and Quantitative Biology (Sorbonne University, CNRS), in collaboration with the Laboratory of Chloroplast Biology and Light Perception in Microalgae (Sorbonne University, CNRS), has developed a novel computational approach for the classification function of protein families. Presented in an article published on March 30, 2022 in Molecular Biology and Evolutionthis approach aims to allow the discovery of proteins playing key functional roles in a wide variety of fields such as biology, human health Where lit is biotechnologys.