Correlation between variant call accuracy and quality parameters in comprehensive cancer genomic profiling tests

Area of Science:

• Genomics
• Molecular Diagnostics

Background:

• Comprehensive genomic profiling (CGP) is a key clinical tool.
• Automated variant calls in CGP often include false positives.
• The relationship between CGP quality metrics and false-positive rates is not well understood.

Purpose of the Study:

• To investigate the correlation between quality parameters and false-positive variants in CGP tests.
• To identify key quality indicators for improving CGP accuracy.

Main Methods:

• Analysis of 125 CGP tests from a single laboratory.
• Manual identification of false-positive variants through visual inspection.
• Evaluation of quality parameters from both wet and dry laboratory processes.

Main Results:

• Over 40% of CGP tests required variant correction, with over 15% needing multiple corrections.
• Significant correlations were found between somatic false positives and various wet (e.g., library concentration) and dry (e.g., mean depth) process parameters.
• Capture library concentration and mean depth emerged as strong independent predictors of somatic false positives.

Conclusions:

• A clear correlation exists between somatic false-positive variants and quality parameters in CGP testing.
• Understanding these correlations is crucial for effective quality management of CGP assays.
• This research supports enhanced quality control strategies for genomic profiling.