Seminar: Kevin Karbstein

Species are the central units for taxonomic research and measuring Earth’s biodiversity. Recent findings in evolutionary genomics are raising awareness that what we call species can be ill-founded entities due to solely morphology-based or regional species descriptions. This particularly applies to plant groups characterized by intricate evolutionary processes. Here, challenges of modern integrative taxonomy, that is genomics combined with morphological, ecological, and other datasets, become apparent: (i) different favored species concepts (e.g., genetic vs. morphological concepts), (ii) missing appropriate analytical tools for intricate evolutionary processes, and (iii) highly subjective ranking and fusion of datasets for final taxonomic treatments (e.g., whether genetics or morphology is taxonomically most important). I will introduce modern integrative taxonomy combined with machine learning (ML) under a unified species concept that enables systematic data integration to reduce subjectivity in species classification and delimitation. I will also present pioneering ML approaches of the group that fuse genetic information with other sources (e.g., ‘DNA’, ‘DNA+Image’), and look inside the black box of the ML process by visualizing highly important plant features for classification (‘XAI’).