A novel statistical method for interpreting the pathogenicity of rare variants

Area of Science:

• Genetics
• Bioinformatics
• Computational Biology

Background:

• Accurate molecular diagnosis is crucial for personalized medicine.
• Distinguishing benign from pathogenic genetic variants is a major challenge in sequencing.
• Existing population databases struggle to filter extremely rare variants.

Purpose of the Study:

• To develop a novel statistical method for filtering benign genetic variants.
• To improve the identification of rare variants with low population allele frequency.
• To provide a general framework for variant interpretation in Mendelian diseases.

Main Methods:

• Developed a statistical test combining patient cohort sequencing data with normal control population databases.
• Compared expected and observed allele frequencies within the patient cohort.
• Evaluated method performance using simulated and real data with experimental validation.

Main Results:

• The new method is well-powered to identify benign variants.
• Demonstrated effectiveness in filtering variants with low frequency in the normal population.
• Validated performance on simulated and real datasets.

Conclusions:

• The developed statistical test provides a general framework for filtering benign variants.
• The method is particularly effective for extremely rare variants.
• This approach supports accurate molecular diagnosis for personalized treatment.