Deep Learning-Assisted Single-Molecule Detection of Protein Post-translational Modifications with a Biological...

Area of Science:

• Proteomics
• Biomarker Discovery
• Nanotechnology

Background:

• Protein post-translational modifications (PTMs) are vital for biological processes and disease biomarkers.
• Current methods for PTM analysis are limited in sensitivity and single-molecule capability.
• Nanopore sensing offers high sensitivity for detecting low-abundance proteins and PTMs.

Purpose of the Study:

• To demonstrate nanopore sensing for detecting and distinguishing single and multiple PTMs on peptides.
• To explore the potential of nanopore technology in single-molecule proteomics and PTM characterization.
• To establish a framework for PTM analysis applicable to biomarker discovery and diagnostics.

Main Methods:

• Utilized a biological nanopore (aerolysin) to sense alpha-synuclein peptides.
• Introduced phosphorylation, nitration, and oxidation PTMs to peptides at various positions and combinations.
• Applied a deep learning model for processing nanopore current signatures to identify PTM variants.
• Quantified peptide concentrations and detected digested peptides.

Main Results:

• Successfully detected and distinguished alpha-synuclein peptides with single or multiple PTMs.
• Identified characteristic current signatures for different PTMs using deep learning.
• Achieved picomolar concentration quantification of alpha-synuclein peptides.
• Detected C-terminal peptides from full-length alpha-synuclein digestion.

Conclusions:

• Nanopore sensing with aerolysin enables simultaneous detection of multiple PTMs on single peptides.
• Deep learning enhances the identification and quantification of PTMs.
• This approach holds significant promise for biomarker discovery and disease diagnostics.