Phasing for medical sequencing using rare variants and large haplotype reference panels.

Area of Science:

• Genetics
• Bioinformatics
• Computational Biology

Background:

• Estimating haplotypes from high-coverage sequencing of single samples is crucial for clinical applications.
• Upcoming large-scale genetic datasets necessitate efficient haplotype estimation methods.

Purpose of the Study:

• To develop a method for haplotype estimation in single, high-coverage sequenced samples.
• To utilize large human genetic variation resources and rare variant sharing patterns for improved phasing.

Main Methods:

• A novel method employing a Hidden Markov Model (HMM) with a selected set of informative copying states.
• Exploits rare variant sharing patterns to infer haplotypes, avoiding iterative Markov Chain Monte Carlo (MCMC) methods.

Main Results:

• Significant gains in phasing accuracy and speed compared to methods not using rare variants.
• Reduced switch error rates by up to 50% when phasing high-coverage samples using the UK10K reference panel.
• Demonstrated proof of concept for phasing large datasets like the 100,000 Genomes Project.

Conclusions:

• Rare variant sharing patterns are a powerful resource for accurate haplotype phasing.
• The developed method offers a computationally efficient and accurate solution for large-scale sequencing studies.
• A webserver is available for phasing high-coverage clinical samples.