Researchers have developed a novel statistical test to interpret the pathogenicity of rare variants and accurately distinguish between benign and pathogenic variants.
Interpreting variant pathogenicity
Advancements in high throughput sequencing technologies have aided in our ability to identify genetic variants. However, interpreting these variants remains an ongoing challenge for precision medicine. Over the past two decades, experts have developed various in silico variant functional prediction tools. These tools are able to distinguish pathogenic from likely benign genetic variants. Nevertheless, these tools are not perfect and are often prone to both false positive and false negative results. In addition, most tools are based on missense variants, leaving other types of variants largely unexplored. Due to large population sequencing data, scientists have been able to identify a lot of benign variants based on allele frequency and disease prevalence. This has also been supported by comparison of allele frequency in patient and normal control populations. However, these allele frequency-based methods have limited power, particularly when the allele frequency of the variant is low in the normal population.
A novel statistical test
In a study, published in Genetics in Medicine, researchers developed a novel statistical method to identify likely benign variants. Dissimilarly to methods that primarily use normal population allele frequency, this method calculates the expected frequency of an allele in the patient cohort. It also takes into account the disease prevalence and the allele frequency within the normal population. Researchers calculate the probability that the variant is pathogenic by comparing the expected and observed frequency of the variant in the patient cohort.
To test the method, the team applied the test to both simulated and real data sets. They evaluated its performance based on literature and other variant prediction methods. The team further examined variants that contradicted previous reports through experimental functional assays.
Test evaluation
Through evaluation of the model using both simulated and real data followed with experimental validation, the team showed that this test is well powered to detect benign variants. Particularly, this test is effective for detecting variants with low allele frequency in the normal population.
Simulation analysis suggested that the test is most powerful when using a patient cohort with a large sample size, for dominant diseases versus recessive diseases, and for rare diseases versus relatively common diseases.
This test is dependent on the availability of patient sequencing data and patient cohort size. Nevertheless, the team suggest that reductions in sequencing costs will help overcome this limitation. Moreover, they believe that this test provides a general framework for filtering benign variants with very low population allele frequency. They hope that its performance will improve as more patient sequences become available.
Image credit: By sergunt – canva.com
