Quartet DNA reference materials and datasets for comprehensively evaluating germline variant calling performance.

Area of Science:

• Genomics
• Molecular Biology
• Bioinformatics

Background:

• Genomic DNA reference materials are crucial for omics research data quality.
• Existing reference datasets are limited to specific regions, hindering whole-genome variant analysis.
• There is a need for DNA reference materials that assess variant detection across the entire genome.

Purpose of the Study:

• To develop and validate novel DNA reference materials for comprehensive variant calling assessment.
• To enable accurate germline variant detection across whole-genome regions.
• To improve the reliability of large-scale genomic profiling.

Main Methods:

• Established a DNA reference material suite using four cell lines from a family (parents and monozygotic twins).
• Integrated and certified comprehensive reference datasets for small and structural variants.
• Utilized a quartet family design for precision estimation outside benchmark regions.
• Employed machine learning models trained on reference data to monitor and alleviate batch effects.
• Enabled cross-omics validation using matched RNA and protein reference materials.

Main Results:

• Developed Quartet DNA reference materials and datasets with 4.2 million small and 15,000 structural variants.
• Demonstrated accurate germline variant call reliability within benchmark regions.
• Enabled precision estimation of variant calls outside benchmark regions.
• Successfully monitored and alleviated batch effects using machine learning.
• Facilitated cross-omics validation of variant calls.

Conclusions:

• Quartet DNA reference materials offer a unique resource for whole-genome variant call quality assessment.
• These materials objectively assess germline variant calls across entire genome regions.
• The study improves the reliability of large-scale genomic profiling through enhanced variant detection.