MAGPIE: A Machine Learning Approach to Decipher Protein-Protein Interactions in Human Plasma

Area of Science:

• Proteomics
• Biochemistry
• Computational Biology

Background:

• Immunoprecipitation coupled to tandem mass spectrometry (IP-MS/MS) is crucial for identifying protein-protein interactions (PPIs).
• Existing methods struggle with false positives from contamination and non-specific binding, especially in human plasma where protein manipulation is limited.
• Current computational models are not optimized for IP-MS/MS data from human plasma without overexpression or inhibition controls.

Purpose of the Study:

• To introduce MAGPIE, a novel machine learning approach for robust PPI identification in human plasma IP-MS/MS experiments.
• To develop a method that effectively utilizes negative controls, including antibodies against non-plasma proteins, for false positive modeling.
• To enhance the reliability of PPI discovery in complex biological fluids like human plasma.

Main Methods:

• Constructed a set of negative controls using antibodies targeting proteins absent in human plasma.
• Developed MAGPIE, a machine learning algorithm to assess PPI reliability in IP-MS/MS experiments targeting known plasma proteins.
• Applied MAGPIE to five IP-MS/MS experiments for validation.

Main Results:

• MAGPIE identified 68 PPIs with a False Discovery Rate (FDR) of 20.77% across five proof-of-concept experiments.
• The algorithm demonstrated superior performance compared to a state-of-the-art PPI discovery tool.
• MAGPIE successfully identified both known and predicted PPIs, validating its efficacy.

Conclusions:

• MAGPIE offers a significant advancement in detecting human plasma PPIs using IP-MS/MS.
• The developed approach overcomes limitations of existing methods for analyzing complex biological samples.
• This tool enables a deeper understanding of biological processes occurring within human plasma.