Quantitative differences between intra-host HCV populations from persons with recently established and persistent...

Area of Science:

• Virology
• Computational Biology
• Infectious Disease Epidemiology

Background:

• Accurate detection of incident hepatitis C virus (HCV) infections is critical for public health interventions and outbreak identification.
• Current diagnostic assays lack the ability to reliably distinguish between recent and persistent HCV infections.
• Understanding the evolutionary dynamics of intra-host HCV populations is incomplete.

Purpose of the Study:

• To investigate the genetic structure and evolutionary dynamics of intra-host HCV populations.
• To identify genetic features that differentiate recent from persistent HCV infections.
• To develop a machine learning classifier for accurate HCV infection staging.

Main Methods:

• Employed next-generation sequencing to analyze intra-host HCV populations from 98 recently and 256 persistently infected individuals.
• Evaluated genetic population structure using 245,878 viral sequences and features measuring diversity, complexity, selection, and evolutionary dynamics.
• Developed and validated a machine learning classifier for infection staging.

Main Results:

• Significant differences in viral population features were observed between recent and persistent HCV infections.
• Viral genetic diversity generally increased from recent to persistent infections, accompanied by decreased genomic complexity and increased population structuredness.
• The developed machine learning classifier achieved a high detection accuracy of 95.22% for infection staging.

Conclusions:

• Intra-host HCV population development follows a complex, regular, and predictable pattern.
• Genetic factors are strongly associated with different stages of HCV infection.
• The proposed cyber-molecular assays could complement or replace standard laboratory methods for HCV infection staging.