Machine-Learning-Based Prediction of Algal Density Using Algal Volatile Organic Compounds for Bloom Early Warning

Area of Science:

• Environmental Science
• Analytical Chemistry
• Biotechnology

Background:

• Harmful algal blooms (HABs) threaten aquatic ecosystems.
• Predicting algal density accurately and rapidly is challenging.
• Algal volatile organic compounds (AVOCs) may offer early warning signals.

Purpose of the Study:

• To develop a novel method for predicting algal density using AVOCs.
• To identify specific AVOC biomarkers for HABs.
• To link metabolic shifts to algal bloom dynamics.

Main Methods:

• Proton transfer reaction time-of-flight mass spectrometry (PTR-TOF-MS) for AVOC analysis.
• Interpretable machine learning (extreme gradient boosting) for density prediction.
• Transcriptomic and enzymatic analyses to understand metabolic pathways.

Main Results:

• The model accurately predicted algal density (R²: 0.95-0.98) using AVOCs.
• Butanal and 2-octenal were identified as key biomarkers.
• Metabolic reprogramming during growth influenced AVOC production.
• Preliminary field validation showed potential for HAB monitoring (67-81% bloom risk).

Conclusions:

• AVOCs serve as dynamic indicators of algal physiology.
• Metabolic shifts are mechanistically linked to bloom dynamics.
• This approach offers a transformative tool for aquatic ecosystem management and HAB monitoring.