Automated quality control for a molecular surveillance system.

Area of Science:

• Virology
• Public Health
• Bioinformatics

Background:

• Hepatitis C virus (HCV) elimination requires robust molecular surveillance and outbreak investigation.
• The Global Hepatitis Outbreak and Surveillance Technology (GHOST) system uses sequencing data for HCV surveillance but is susceptible to data quality impacts.
• Assessing and improving data quality is crucial for accurate epidemiological inferences from GHOST.

Purpose of the Study:

• To analyze data quality in the GHOST system during its pilot phase.
• To identify common data quality concerns affecting GHOST.
• To develop automatic methods for detecting and correcting data quality issues to enhance surveillance accuracy.

Main Methods:

• Examination of GHOST quality control filters and failure rates.
• Development of new filters for primer dimers, product loss, and short products.
• Adjustment of the genotyping tool and incorporation of parameters for anomaly detection into an automatic quality control module.

Main Results:

• Identified new quality control problems in MiSeq data submitted to GHOST.
• Developed and implemented new filters and adjusted genotyping for improved accuracy.
• Enhanced the automatic quality control module to reject erroneous data and guide users in corrective actions.

Conclusions:

• New quality control measures were implemented to protect GHOST from erroneous data.
• The GHOST system was upgraded to identify causes of errors and recommend corrective actions.
• Improved data quality enhances the reliability of GHOST for HCV molecular surveillance and outbreak detection.