Interpretating SPR-Derived Reaction Kinetics via Self-Organizing Maps for Diagnostic Applications

Area of Science:

• Biomedical Engineering
• Infectious Disease Diagnostics
• Artificial Intelligence in Healthcare

Background:

• Biosensors offer cost-effective, field-deployable diagnostics for infectious diseases, crucial for resource-limited settings.
• Surface Plasmon Resonance (SPR) biosensors excel in label-free, real-time analysis of biomolecular interactions.
• Canine visceral leishmaniasis (CVL) diagnosis is often delayed, hindering effective disease control in human and canine populations.

Purpose of the Study:

• To develop and evaluate an SPR biosensor integrated with Self-Organizing Maps (SOMs) for enhanced serodiagnosis of CVL.
• To assess the diagnostic performance and kinetic parameters of a multiepitope chimeric protein (PQ20) for CVL detection.
• To demonstrate the utility of AI-driven data analysis for rapid and accurate classification of infected versus healthy individuals.

Main Methods:

• Utilized an SPR biosensor to analyze the binding kinetics of the PQ20 protein with anti-PQ20 antibodies.
• Employed Self-Organizing Maps (SOMs) for high-dimensional data projection and automated classification of samples.
• Evaluated diagnostic performance including detection limit, sensitivity, and specificity using raw serum samples.

Main Results:

• Identified two immunodominant epitopes within PQ20, exhibiting high association rates (k_a1 = 2.4 × 10^5 L mol^-1 s^-1) and dissociation rates (k_d1 = 5.5 × 10^-4 L mol^-1 s^-1).
• Achieved a detection limit of 5.1 nmol L^-1 for the SPR biosensor.
• SOM analysis demonstrated improved diagnostic accuracy (sensitivity and specificity) compared to univariate analysis, enabling classification in under 15 minutes with reduced reaction times (100 s).

Conclusions:

• The integration of SPR biosensing with AI-powered SOM analysis provides a rapid, label-free, and accurate method for CVL serodiagnosis.
• This approach enhances diagnostic capabilities, particularly for neglected tropical diseases in resource-limited settings.
• The developed biosensor system shows significant potential for improved infectious disease surveillance and management.