This research demonstrates how retrieval-augmented LLMs can effectively analyze proteomics data to predict protein interactions in cardiac cells.

• Applied AI to identify protein fragments that bind to SERCA calcium pumps
• Overcame limited training data challenges through retrieval augmentation
• Successfully classified cardiac protein interactions with high accuracy
• Established a novel approach for analyzing complex proteomics datasets

This innovation enables more efficient discovery of cardiac protein interactions that could impact heart function and disease treatment, potentially accelerating therapeutic development for cardiac conditions.

Leveraging Retrieval-Augmented Generation and Large Language Models to Predict SERCA-Binding Protein Fragments from Cardiac Proteomics Data