Artificial intelligence (AI) has the potential to revolutionize the way we identify fungal versatility from scientific literature. Fungi play crucial roles in our ecosystems, from symbiotic relationships with plants to decomposing dead wood. However, many fungi have the ability to switch lifestyles depending on their environment, making it challenging for researchers to understand their flexibility and predict how they will respond to climate change.
Fortunately, a new study published in the open-access journal Research Ideas and Outcomes by Northern Arizona University doctoral student Beatrice M. bock demonstrates how AI can solve this problem. By utilizing a specialized language model called BioBERT, Bock developed an automated workflow that accurately identifies whether a fungus has a single lifestyle or a dual one.
A high-accuracy solution
Bock explains that manually identifying fungal versatility from literature is time-consuming and difficult to scale. With machine learning, thousands of papers can be scanned in just minutes to flag species that may have dual roles - such as helping plants grow while also acting as decomposers when the plant dies.
“Manually identifying fungal versatility from literature is time-consuming and difficult to scale. By using machine learning, we can now scan thousands of papers in just a few minutes to flag species that might be switching roles – such as a fungus that normally helps a plant grow but also turns into a decomposer when the plant dies.”
Beatrice M. Bock
The pilot study tested four different AI models to see which was best at understanding the nuances of biological language. The top-performing model, BioBERT, achieved nearly 90% accuracy in identifying fungal lifestyles.
What sets BioBERT apart from other models? One key factor is its ability to recognize capital letters. Bock found that “cased” models – those that recognize capitalization – performed significantly better than those that did not. This is likely because scientific names of species, such as Fusarium, are often capitalized and crucial for AI to understand the context of research.
The path ahead
Bock has made all the code and data from her study available for free online, demonstrating her commitment to transparency and allowing other scientists to build upon her work and track traits in other organisms like insects or plants.
While this study focused on a small group of papers as proof-of-concept,it opens doors for much larger projects. Future versions of this tool could predict how fungal behavior may change under specific environmental conditions like drought or extreme heat.
“As fungal trait databases continue to grow in importance for biodiversity assessments, automated text mining offers a path toward more efficient, consistent and comprehensive trait annotation.”
Beatrice M. Bock
New Headings:
- The Power of Artificial Intelligence (AI) in Identifying Fungal Versatility
- fungi: The hidden Architects of Our Ecosystems
- Understanding Fungal Flexibility for Climate Change Predictions
- The Challenge of Manual Data Extraction from Scientific Literature
- Introducing biobert: A Specialized Language Model for Automated Workflow
- The Accuracy and Efficiency of AI in Identifying Fungal Lifestyles
- Why Capitalization Matters in Biological Language Models
- Paving the Way for Future Research with Open Access Code and Data
New Statistics:
In 2026, a study led by Northern Arizona University doctoral student Beatrice M. Bock demonstrated how AI can accurately identify fungal versatility from scientific literature with nearly 90% accuracy.
New Examples:
A fungus that normally helps a plant grow but also turns into a decomposer when the plant dies is an example of dual lifestyle flexibility.
new Tone:
the potential impact of artificial intelligence on identifying fungal versatility from scientific literature cannot be underestimated. With the power to scan thousands of papers in just minutes, AI has opened doors for more efficient and accurate research methods.
Sources:
Bock B (2026) Automated extraction of fungal trophic modes from literature using BioBERT: an open pilot workflow. Research Ideas and Outcomes 12: e176590. https://doi.org/10.3897/rio.12.e176590
Story originally published by: EurekAlert! (2026). Using AI to uncover the secret lives of fungi. [online] Available at: https://www.eurekalert.org/news-releases/1114462 [Accessed 2 Feb. 2026]. Republished with permission.
Follow the latest publications and news from the Research Ideas and Outcomes (RIO) journal on Bluesky,Facebook and LinkedIn.
This is fascinating! Can’t wait to see what new discoveries AI will uncover in the world of fungi.