Motifier: An IgOme Profiler Based on Peptide Motifs Using Machine Learning

Area of Science:

• Immunology
• Bioinformatics
• Computational Biology

Background:

• Antibodies in serum offer insights into immune system encounters.
• Analyzing antibody repertoires is crucial for understanding clinical conditions.
• Phage display and Next-Generation Sequencing (NGS) enable deep probing of antibody spectrums.

Purpose of the Study:

• To introduce Motifier, a computational pipeline for analyzing Deep-Panning data.
• To systematically generate discriminatory peptide motifs from affinity-selected peptides.
• To enable accurate classification of antibody mixtures for various biological conditions.

Main Methods:

• Utilized Deep-Panning technology combined with NGS for antibody repertoire analysis.
• Developed Motifier, a computational pipeline with algorithms for motif generation.
• Implemented machine-learning protocols for classification of antibody mixtures.

Main Results:

• Motifier systematically generates discriminatory peptide motifs from Deep-Panning data.
• These motifs effectively characterize specific antibody binding activities.
• Machine-learning classification accurately distinguished antibody mixtures representing different biological states.

Conclusions:

• Motifier provides a powerful bioinformatics approach for analyzing antibody repertoires.
• The generated peptide motifs serve as effective biomarkers for clinical situations.
• This methodology advances the potential for data-driven immunological diagnostics.