Identifying rare variants with optimal depth of coverage and cost-effective overlapping pool sequencing

Area of Science:

• Genomics
• Genetic Epidemiology
• Bioinformatics

Background:

• Genome-wide association studies (GWAS) have identified numerous genetic variants for complex diseases, but explain limited heritability.
• The missing heritability is hypothesized to stem from rare variants, necessitating efficient screening methods.
• Large-scale resequencing for rare variants remains costly, despite declining sequencing expenses.

Purpose of the Study:

• To propose and evaluate an overlapping pool sequencing strategy for cost-effective screening of rare variants.
• To develop a model for determining optimal sequencing depth in pooled samples.
• To minimize costs and ensure accuracy in rare variant detection using pooled sequencing.

Main Methods:

• Developed a cost model and formulated a method to compute optimal sequencing depth for pooled samples.
• Employed a shifted transversal design algorithm to select appropriate overlapping pool sequencing parameters.
• Implemented a strategy of dividing large populations into smaller blocks for independent, optimized testing.

Main Results:

• Overlapping pool sequencing demonstrated significant cost-effectiveness, reducing expenses by at least 66% for screening 1% frequency variants in 200 individuals (30 Mb target region).
• The experimental screening of variant carriers with 1% frequency achieved 99.93% accuracy using simulated pools and public exome data.
• Dividing populations into smaller blocks proved more cost-effective due to mixing constraints and high depth requirements in pooled sequencing.

Conclusions:

• Overlapping pool sequencing is a highly accurate and cost-efficient method for identifying rare genetic variants.
• The proposed strategy and cost model provide a framework for optimizing rare variant screening in large populations.
• This approach significantly lowers the financial barrier to investigating the role of rare variants in complex diseases.