Contact person: Bastiaan Star
Keywords: biodiversity, ecology, evolution, artificial intelligence, data science
Research groups: Centre for Ecological and Evolutionary Synthesis (CEES), Evolution, eDNA, Genomics and Ethnobotany (EDGE)
Department of Biosciences (IBV), Natural History Museum (NHM)
Data science and artificial intelligence approaches are revolutionizing biodiversity research in ecology and evolution in a similar way that genetics and genomics have done in recent decades. Harnessing the potential of these approaches for research necessitates bridging computer sciences and mathematics to biology. Automated image recognition, feature extraction, semantic segmentation and AI-based quality control of high-throughput sequencing enable acceleration of data capture, analysis and interpretation. Emerging research shows novel applications in population monitoring, biodiversity tracking through time and space, conservation planning and biodiversity protection through deep learning and digital twin approaches with relevance to sustainability and the green transition. Research proposals should work across fields for synergistic leveraging of biodiversity, genomics and ecological data using data science approaches.
Research topics:
- Deep learning for species identification based on images
- Semantic segmentation for analysis of collection/field data
- AI-driven morphometrics
- Applying neural networks to infer population demographic parameters
- Phenology analysis from citizen science observations
- AI analyses of whole genome and/or environmental sequence data for biological significance testing
- AI model selection (e.g. recurrent neural networks or spatiotemporal convolutional neural networks) to learn complex patterns in spatiotemporal (sequence) data
- Use machine learning algorithms for real-time ecological data analyses
Application domains for Data Science and Biodiversity:
- Biodiversity/Conservation
- Ecology
- Evolutionary biology
- Natural history collections (including DiSSCo)
- Biological observation data (including GBIF)
Research team:
- Sanne Boessenkool (CEES/IBV)
- Bastiaan Star (CEES/IBV)
- Mark Ravinet (CEES/IBV)
- Hugo de Boer (EDGE/NHM)
Mentoring and internship will be offered by a relevant external partner.
