Unveiling Immune Response Mechanisms in Mpox Infection Through Machine Learning Analysis of Time Series Gene...

Area of Science:

• Immunology
• Genomics
• Computational Biology

Background:

• Monkeypox virus (Mpox) outbreaks have increased globally, necessitating a deeper understanding of the host immune response.
• Effective disease management and vaccine development for Mpox depend on identifying key immune mechanisms.
• Macaque models are crucial for studying Mpox pathogenesis and immune responses.

Purpose of the Study:

• To identify key immune-related genes associated with different stages of Mpox infection using machine learning.
• To uncover novel molecular targets for monitoring and intervention strategies against Mpox.
• To build predictive models and classification rules for Mpox infection stages based on gene expression.

Main Methods:

• Analysis of time series gene expression data from Mpox-infected macaques.
• Application of nine feature ranking methods and incremental feature selection to identify important genes.
• Utilized twelve classification algorithms and Synthetic Minority Oversampling Technique for model building.

Main Results:

• Identified repeatedly significant immune genes like CD19, MS4A1, and TLR10, involved in B-cell activation and innate immunity.
• Discovered novel key genes, including HS3ST1, SPAG16, and MTARC2, potentially implicated in Mpox pathogenesis.
• Developed efficient prediction models and classification rules for different infection phases.

Conclusions:

• Machine learning effectively identified critical immune genes in Mpox infection.
• The study provides insights into the host immune response to Mpox, highlighting both known and novel molecular players.
• Identified genes offer potential targets for future Mpox diagnostics and therapeutics.