Alex Van Dam

Publications

Van Dam, A.R. & Štarhová Serbina, L. (2025) Descriptron: Artificial intelligence for automating taxonomic species descriptions with a user-friendly software package. Systematic Entomology, e70005. doi: https://doi.org/10.1111/syen.70005

GoogleScholar: https://scholar.google.com/citations?hl=en&user=yCDm5YIAAAAJ&view_op=list_works&sortby=pubdate

Projects

• Van Dam, A.R. (2024). Descriptron software — Open‑source toolkit for AI‑assisted taxonomy (GitHub). https://github.com/alexrvandam/Descriptron.
• Van Dam A.R. (2025) DINOSAR software — DINOv3 Species Auto-Recovery (AI Enabled Morphological Species Delimitation) (GitHub) https://github.com/alexrvandam/DINOSAR
• Van Dam, A.R. (2025) PHYLUCE_GUI software — A graphical user interface (GUI) for the PHYLUCE pipeline. (GitHub) https://github.com/alexrvandam/PHYLUCE_GUI
• Van Dam, A.R (2026) SAM2-PAL software — SAM2-Palindrome Self-Training with Cycle Consistency. (GitHub) https://github.com/alexrvandam/SAM2-PAL

Tasks

Alex R. Van Dam is a research scientist at the Museum für Naturkunde Berlin (MfN) working at the interface of biodiversity discovery, entomological systematics, and machine learning. He develops open, reproducible workflows that connect specimen imaging, quantitative morphology, and vouchered sequence data to accelerate species delimitation and diagnosis at scale. His work combines practical computer-vision tools for annotation and segmentation (e.g., the Descriptron pipeline) with large language models to help translate complex morphological evidence into consistent, high-quality taxonomic descriptions. In parallel, he explores geometric morphometrics—such as semi-landmarks and shape-based analyses—to quantify diagnostic characters and integrate them with image- and sequence-based evidence. The overarching aim is high-throughput taxonomy and biodiversity discovery that can scale from individual projects to collection-wide and global initiatives.