Rapid genotype imputation from sequence with reference panels.

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Accurate and cost-effective genotyping is crucial for advancing genomic research and applications.
• Existing imputation methods often struggle with low-coverage sequencing data, limiting their utility for large-scale or cost-sensitive projects.
• Diverse populations are underrepresented in genomic datasets, hindering the discovery of genetic variants relevant to all individuals.

Purpose of the Study:

• To develop and validate QUILT, a novel computational method for diploid genotype imputation using low-coverage whole-genome sequencing data.
• To assess QUILT's accuracy and performance across various sequencing technologies and compare it with existing imputation strategies.
• To demonstrate QUILT's utility for imputing complex genetic regions, such as human leukocyte antigen (HLA) types, from low-coverage data.

Main Methods:

• QUILT utilizes Gibbs sampling to partition sequencing reads into maternal and paternal sets, enabling accurate haploid imputation.
• The method leverages large reference panels for imputation, achieving high accuracy over extended genomic regions.
• QUILT was evaluated using diverse sequencing data, including Oxford Nanopore long reads and low-cost haplotagged Illumina data.

Main Results:

• QUILT achieves highly accurate diploid genotype imputation from low-coverage whole-genome sequencing data, approaching theoretical limits.
• The method demonstrates superior performance compared to existing imputation techniques, particularly at low coverages and for rare single nucleotide polymorphisms (SNPs).
• QUILT accurately imputes human leukocyte antigen (HLA) types from low-coverage sequence data, a first for such methods.

Conclusions:

• QUILT provides a cost-effective and accurate solution for genotype imputation using low-coverage sequencing, democratizing genomic analyses.
• The method shows particular promise for improving genomic insights in diverse and traditionally underserved populations.
• QUILT's ability to impute complex genetic regions like HLA types expands its applicability in clinical and research settings.