In April 20, Mr Naoki Konno (University of Tokyo) gave a fantastic talk about the machine leaning method for the prediction of evolution. The talk was entitled, “Machine learning predicts biological system evolution by gene gains and losses”. To my knowledge, this study using genome data of ~3000 species is one of few seminal attempts to theoretically predict long-term and system-level evolution. The seminar by Konno san was very clear and we understood the predictability of the proposed framework, called “Evodictor”. In the beginning of the talk, he kindly explained the theoretical basis about phylogenetic tree estimation and machine learning for non-specialists for which we could enjoy and learn from his seminar a lot. Thank you, Konno-san for the great talk!!

Reported by Gen Kurosawa


  1. Naoki Konno and Wataru Iwasaki, Machine learning enables prediction of metabolic system evolution in bacteria, Science Advances (2023), doi: 10.1126/sciadv.adc9130