Quantifying unobserved protein-coding variants in human populations provides a roadmap for large-scale sequencing...

Area of Science:

• Genomics
• Population Genetics
• Bioinformatics

Background:

• Large-scale human genome sequencing projects are expanding.
• Evaluating the statistical power of these projects requires robust quantitative frameworks.
• Understanding the full spectrum of genetic variation, especially rare variants, is crucial.

Purpose of the Study:

• To develop a novel algorithm, UnseenEst, for estimating the frequency distribution of all protein-coding genetic variants.
• To assess the expected number of new variants discoverable in future large sequencing cohorts.
• To quantify the proportion of all possible loss-of-function and missense variants captured with increasing cohort sizes.

Main Methods:

• Development of the UnseenEst algorithm.
• Application of UnseenEst to exome data from 60,706 individuals.
• Extrapolation of variant frequencies to predict discovery in larger cohorts (up to 500,000 individuals).

Main Results:

• UnseenEst accurately estimates the frequency distribution of protein-coding variants, including unobserved rare variants.
• With 500,000 individuals, approximately 7.5% of all possible loss-of-function variants and 12% of missense variants are expected to be captured.
• An estimated 2,900 genes exhibit loss-of-function frequencies below 0.00001 in healthy populations, indicating strong intolerance to gene inactivation.

Conclusions:

• The UnseenEst algorithm provides a quantitative framework for evaluating the power of large human sequencing initiatives.
• Future large cohorts will significantly increase the identification of rare genetic variants.
• A substantial number of genes are highly intolerant to loss-of-function mutations in the human population.