Insight into the Relationships Between Chemical, Protein and Functional Variables in the PBP/GOBP Family in Moths...

Area of Science:

• Insect olfaction
• Molecular biology
• Computational chemistry

Background:

• Insects detect environmental chemicals using olfactory proteins in their antennae.
• Odorant-binding proteins (OBPs), including Pheromone-binding proteins (PBPs) and general-odorant-binding proteins (GOBPs), are crucial for this process in moths.
• Limited understanding exists regarding specific OBP-chemical interactions and functional roles.

Purpose of the Study:

• To integrate chemical, protein, and functional data for moth OBPs.
• To develop predictive models for OBP-chemical interactions.
• To leverage existing data for a deeper understanding of olfactory mechanisms.

Main Methods:

• Curated chemical, protein, and functional datasets.
• Developed descriptors for dataset features.
• Implemented and evaluated regression algorithms, including XGBoostRegressor, GradientBoostingRegressor, and LightGBMRegressor.

Main Results:

• XGBoostRegressor achieved the highest predictive performance with an R² of 0.76, RMSE of 0.28, and MAE of 0.20.
• GradientBoostingRegressor and LightGBMRegressor also showed strong performance.
• Demonstrated a significant correlation between chemical, protein, and functional data for moth PBP/GOBP families.

Conclusions:

• Machine learning models can effectively predict OBP-chemical interactions.
• This integrated data approach provides novel insights into moth olfaction.
• The findings establish a foundation for future research on insect chemical communication.