Sign in Subscribe
Genetic research

Leveraging Machine Learning to Classify Coral Diseases

Strategic Insights

— 1 min read

Researchers from the University of Texas Arlington have made a breakthrough in developing a machine learning model to classify and characterize gene expression patterns in two common coral diseases, stony coral tissue loss disease (SCTLD) and white plague (WP). By identifying 463 gene expression biomarkers, the study provides a preliminary disease classification model that distinguishes between SCTLD and WP, offering valuable insights into their underlying cellular responses. The model has high predictive performance, with an Area Under the Receiver Operating Characteristic (ROC) Curve (AUC) of 0.9895, and has the potential to develop rapid diagnostic tools to identify and mitigate future coral disease outbreaks.

Key Takeaways:

  • The study highlights the devastating effects of anthropogenic climate change on coral reef systems, including increased disease outbreaks and declining reef biodiversity.
  • The researchers identified 463 gene expression biomarkers, with 275 unique to SCTLD and 167 unique to WP, revealing distinct immune responses between the two diseases.
  • The machine learning model, built with partial least squares discriminant analysis (PLS-DA), achieved high predictive performance, with an AUC of 0.9895 and average balanced error rate (BER) of 0.0799.
  • The study provides a foundation for the development of rapid diagnostic tools to identify and mitigate future coral disease outbreaks.
  • The researchers emphasize the need for more sophisticated methods to classify diseases that appear visually similar and require distinct treatment protocols.
  • The study highlights the potential of machine learning to classify and characterize gene expression patterns in coral diseases, offering valuable insights into their underlying cellular responses.

Statistics:

  • 463 gene expression biomarkers were identified, with 275 unique to SCTLD and 167 unique to WP.
  • The machine learning model achieved an AUC of 0.9895 and average overall error rate of 0.0754.
  • The model delivered an average balanced error rate (BER) of 0.0799.
  • The study provides a preliminary disease classification model that distinguishes between SCTLD and WP.
  • Coral disease outbreaks have resulted in high mortality rates and declining reef biodiversity.

Sources:

  • "Leveraging machine learning to classify and characterize gene expression patterns in two coral diseases." Discover Applied Sciences, 2025,7(9):1-24. The publisher for Discover Applied Sciences is Springer.
  • A free version of this journal article is available at https://doi.org/10.1007/s42452-025-07382-7
  • NewsRx. University of Texas Arlington Researchers Release New Data on Climate Change (Leveraging machine learning to classify and characterize gene expression patterns in two coral diseases). Health & Medicine Week. September 19, 2025; p 8494.

Read more