Deep Learning Model Predicts Invasive Plant Spread Across Michigan Under Climate Change
Researchers used a deep learning framework called Deepbiosphere, combined with citizen science and remote sensing data, to predict the distribution of 1,553 vascular plant species in Michigan, with a focus on two invasive species. The model outperformed baseline approaches by nearly 11% on average, and by over 56% and 74% respectively for the two target invasive species, Rhamnus cathartica and Ailanthus altissima. The findings matter because they provide high-resolution risk maps that could guide early intervention efforts as climate change is projected to drive both species northward.
A new preprint study published on bioRxiv applied the Deepbiosphere deep learning framework to model the distributions of 1,553 vascular plant species across Michigan, integrating citizen science observations, remote sensing imagery, and climate data. The model achieved a mean AUC-ROC of 0.79 across all species and outperformed traditional machine learning baselines by an average of 10.98%. For two focal invasive species — common buckthorn (Rhamnus cathartica) and tree of heaven (Ailanthus altissima) — accuracy improvements were especially pronounced, at 56.41% and 74.99% respectively. Current projections show R. cathartica is already broadly suitable across much of Michigan, while A. altissima remains more restricted to southern areas but is projected to expand strongly northward under future climate scenarios. The study also mapped prediction uncertainty, finding that differences between general circulation models (GCMs) were the dominant source of spatial uncertainty across most of the state. The authors argue that explicitly quantifying and communicating this uncertainty is critical for informing management decisions, particularly given the range of possible future climate trajectories.
What's missing
As a preprint, this study has not yet undergone formal peer review, so its methods and conclusions have not been independently validated. The study does not report field-validation of the generated risk maps against observed on-the-ground invasive species presence, which would be important for assessing real-world applicability. Additionally, the model's transferability to other regions or species beyond Michigan's vascular flora is not assessed.
What different sources said
- bioRxivCenter
Mapping distribution of invasive plant species and uncertainty using citizen science, remote sensing, and deep learning
Related
Multiscale Brain Model Predicts Novel Propofol Anesthesia Biomarker Without Training on Clinical Data
Researchers developed a mechanistic computational model of thalamocortical brain circuits that successfully predicted a previously unnoticed dose-dependent biomarker of propofol anesthesia. The model, driven solely by GABA-A receptor modulation, reproduced empirical data from both macaques and humans without being fitted to any anesthesia-specific data. The findings suggest that simulation-first approaches could accelerate biomarker discovery in neuropharmacology without requiring large clinical datasets.
Green-Synthesized Zinc Oxide Nanoparticles from Mimosa pudica Show Biocompatibility with Bone Marrow Stem Cells in Lab Study
Researchers synthesized zinc oxide nanoparticles using Mimosa pudica leaf extract and tested their effects on human bone marrow mesenchymal stromal cells, finding the nanoparticles preserved cell viability, structure, and bone-forming capacity. The plant-derived nanoparticles outperformed both the raw plant extract and conventionally synthesized zinc oxide in maintaining cell metabolic activity over five days. The findings suggest these bioactive nanomaterials could be candidates for musculoskeletal tissue engineering, though the research remains at an early in vitro stage.
Study Compares Genetic Modeling Approaches for Dyadic Social Interactions in Animals
A new preprint study compared two statistical modeling approaches for analyzing the genetic basis of social interactions in animals, finding that dyadic models outperform marginal models that aggregate individual-level data. The research used pig aggression data from 797 finishing pigs across 59 social groups as a test case. The findings have implications for how animal geneticists model and interpret the heritable components of social behavior.