BICEP: Bayesian inference for rare genomic variant causality evaluation in pedigrees

Area of Science:

• Genetics
• Bioinformatics
• Computational Biology

Background:

• Next-generation sequencing (NGS) is crucial for gene discovery using pedigree data.
• Identifying causative variants within a robust statistical framework remains a significant challenge.

Purpose of the Study:

• Introduce BICEP (Bayesian Inference tool for Causal variant Evaluation in Pedigrees).
• Provide a robust statistical framework for rare variant causality evaluation in pedigree-based cohorts.

Main Methods:

• BICEP employs Bayesian inference to calculate posterior odds of variant causality.
• Integrates variant cosegregation data with a priori evidence (deleteriousness, functional consequence).
• Evaluates variants within pedigree structures for Mendelian and complex genetic architectures.

Main Results:

• BICEP accurately identifies causal rare variants, outperforming existing methods.
• Effectively down-weights common variants unlikely to be causal, even with good cosegregation.
• Provides quantitative metrics for comparing variant causality within and across pedigrees.

Conclusions:

• BICEP offers a superior approach for rare variant causality assessment in family studies.
• Enables more accurate gene discovery for both simple and complex genetic diseases.
• Facilitates quantitative, cross-pedigree variant causality comparisons.