Machine Learning Analysis of RNA-Seq Data Identifies Key Gene Signatures and Pathways in Mpox Virus-Induced...

Area of Science:

• Virology and Molecular Biology
• Gastroenterology
• Genomics

Background:

• Mpox virus (MPXV) emerged globally in 2022, causing public health concerns.
• Gastrointestinal complications, including diarrhea and proctitis, are observed in MPXV infections.
• Understanding MPXV's impact on the colon is crucial for managing GI symptoms.

Purpose of the Study:

• To investigate the impact of MPXV infection on colon gene expression using human-induced pluripotent stem cell-derived colon organoids.
• To compare the effects of different MPXV clades (I, IIa, IIb) on colonic cells.
• To identify key genes and pathways affected by MPXV infection in the colon.

Main Methods:

• Reanalysis of existing RNA-seq data (GSE219036) from MPXV-infected colon organoids.
• Application of Generalized Linear Models with Quasi-Likelihood F-tests and Relaxed Magnitude-Altitude Scoring (GLMQL-RMAS) for differential gene expression analysis.
• Utilized a novel Cross-RMAS methodology and machine learning for gene ranking and strain differentiation.

Main Results:

• MPXV Clade I induced the most significant gene expression alterations, upregulating stress response genes (e.g., HSPA6, FOS) and downregulating cytoskeletal/trafficking genes (e.g., PSAP, CFL1).
• MPXV Clade IIb exhibited the least impact on colon gene expression.
• Machine learning models achieved 100% accuracy in differentiating MPXV strains and mock samples based on identified genes.
• Gene Ontology analysis revealed disruptions in protein folding, immune response, and epithelial integrity pathways.

Conclusions:

• MPXV infection differentially affects colon gene expression depending on the clade.
• MPXV Clade I significantly impacts cellular processes, potentially explaining severe GI symptoms.
• The identified gene expression changes provide insights into MPXV pathogenesis and potential therapeutic targets for GI manifestations.