The dengue-specific immune response and antibody identification with machine learning

Area of Science:

• Immunology
• Bioinformatics
• Computational Biology

Background:

• Dengue virus is a global health threat with no specific treatment.
• Broadly neutralizing antibodies offer a potential therapeutic strategy against all dengue serotypes.
• Adaptive immune receptor repertoire sequencing (AIRR-seq) provides deep insights into B-cell responses.

Purpose of the Study:

• To investigate the dengue antibody response using AIRR-seq and bioinformatics.
• To apply machine learning for identifying rare, broadly neutralizing antibody sequences.
• To validate computational findings with experimental assays.

Main Methods:

• High-throughput adaptive immune receptor repertoire sequencing (AIRR-seq).
• Bioinformatic analysis of antibody repertoires.
• Machine learning algorithms for sequence identification.
• Antibody expression and functional assays for validation.

Main Results:

• Dengue immunization increased CDR3 and germline gene diversity.
• Antibody repertoire architecture shifted, showing power-law distributions and CDR3 enrichment.
• Upregulation of JNK/Fos transcription factors and ribosomal proteins observed.
• Machine learning successfully identified potential neutralizing antibody candidates.

Conclusions:

• AIRR-seq and machine learning are powerful tools for understanding immune responses and identifying therapeutic antibodies.
• Dengue immunization induces specific, measurable changes in the antibody repertoire.
• Identified antibody sequences show promise for dengue treatment development.