What is it about?
Nanoparticles are increasingly used in agriculture as fertilizers, pesticides, and plant growth enhancers. While they offer many benefits, their effects on plant growth are not always well understood and can vary depending on their properties. In this study, we developed an artificial intelligence (AI) approach to predict how nanoparticles influence plant growth, specifically plant length (such as root or shoot growth). We first carefully cleaned and improved existing experimental data, enriching it with detailed information about nanoparticle structure. This ensured that the data used for modeling was reliable and comprehensive. We then applied automated machine learning (AutoML) to test multiple algorithms and identify the best-performing model. The final model, based on XGBoost, achieved high predictive accuracy (around 85%) in classifying how nanoparticles affect plant growth.  Importantly, the model does not require new experimental data to make predictions. Instead, it can estimate the effects of nanoparticles virtually, allowing researchers to screen materials before testing them in the lab. The dataset and model are openly available and accessible through the Enalos Cloud platform as an easy-to-use web tool
Featured Image
Photo by Katrin Hauf on Unsplash
Why is it important?
Agriculture is facing major challenges, including climate change, increasing food demand, and the need for more sustainable practices. Nanotechnology offers promising solutions, but assessing the safety and effectiveness of nanoparticle-based products remains a key barrier. This work is important because it provides a fast, reliable, and reproducible way to predict how nanoparticles affect plant growth without relying on time-consuming experiments. By combining rigorous data curation, enrichment, and advanced AI modeling, the study ensures high-quality and trustworthy predictions. The approach supports “safe-and-sustainable-by-design” development of nano-enabled agricultural products, helping scientists and industry identify formulations that enhance plant growth while minimizing environmental risks. It also aligns with FAIR data principles, making both the data and models accessible and reusable for the wider scientific community.
Perspectives
This work highlights the critical role of high-quality data in building reliable AI models. By focusing on rigorous data curation and enrichment, the study demonstrates that improving data quality can significantly enhance model performance and applicability. From a personal perspective, the integration of AutoML with curated datasets shows how advanced modeling can become more accessible to researchers without deep expertise in machine learning. The deployment of the model through the Enalos Cloud platform further emphasizes the importance of making scientific tools openly available and easy to use. Looking ahead, this approach can be extended to other agricultural endpoints and environmental systems, supporting smarter, data-driven decision-making in agriculture. It has strong potential to accelerate innovation in sustainable farming, helping balance productivity with environmental protection in a rapidly changing world.
Dr Antreas Afantitis
NovaMechanics Ltd
Read the Original
This page is a summary of: Rigorous data curation, enrichment and meta-analysis enable autoML prediction of plant length responses to nanoparticles powered by the Enalos Cloud platform, Environmental Science Nano, January 2026, Royal Society of Chemistry,
DOI: 10.1039/d5en00897b.
You can read the full text:
Contributors
The following have contributed to this page
